Oneflare Pty Ltd v Chernih - [2016] NSWSC 1271

Supreme Court

New South Wales

Medium Neutral Citation: Oneflare Pty Ltd v Chernih [2016] NSWSC 1271
Hearing dates: 14/03/2016, 15/03/2016, 16/03/2016, 17/03/2016, 18/03/2016 and 19/08/2016
Date of orders: 13 September 2016
Decision date: 13 September 2016
Jurisdiction: Equity - Commercial List
Before: McDougall J
Decision:
Judgment for defendants on plaintiff’s claim. Reserve costs. Directions for submissions on costs.

Catchwords:
CONTRACT – contract for search engine optimisation services – where scope and terms of retainer not fully expressed – implied contractual and common law duty to use reasonable care and skill – whether duty extended to duty to warn – whether scope and terms informed by subsequent conduct and performance of retainer – whether plaintiff or defendants responsible for design of search engine optimisation strategy – whether defendants used reasonable care and skill in the performance of the retainer
DAMAGES – pure economic loss – whether breach of retainer results in loss – where direct correlation between alleged breach and loss – assessment of expert evidence – whether discount should be allowed if damages awarded
Legislation Cited: Civil Liability Act 2002 (NSW)
Trade Practices Act 1974 (Cth)
Fair Trading Act 1987 (NSW)
Cases Cited: Brambles Holdings Ltd v Bathurst City Council (2001) 53 NSWLR 153
Commercial Union Assurance Company of Australia Ltd v Ferrcom Pty Ltd (1991) 22 NSWLR 389
County Securities Pty Ltd v Challenger Group Holdings Pty Ltd [2008] NSWCA 193
Liverpool City Council v Irwin [1977] AC 239
Prestia v Aknar (1996) 40 NSWLR 165
Wallace v Kam (2013) 250 CLR 375
Watson v Foxman (1995) 49 NSWLR 315
Category: Principal judgment
Parties: Oneflare Pty Ltd (Plaintiff)
Nicholas Chernih (First Defendant)
Linkbuild SEO Pty Ltd (Second Defendant)
Representation:
Counsel:
G Lucarelli (Plaintiff)
D W Robertson (Defendants)
Solicitors:
Levitt Robinson Solicitors (Plaintiff)
Robertson Saxton Primrose Dunn (Defendants)

File Number(s): 2014/336363

Judgment

HIS HONOUR: The plaintiff (Oneflare) operates an internet business which seeks to match service providers (such as removalists, plumbers, pest control experts and no doubt many others) with prospective customers. Service providers register with Oneflare. Prospective customers post details of the services they require on Oneflare’s website. Service providers who are interested will provide quotations for the work. The prospective customer can choose from among the quotations so provided.
Like other internet-based businesses, Oneflare depends critically on search engines, such as (and primarily) Google, to obtain business. Prospective customers who know the kind of service they require but do not know of Oneflare’s website will typically conduct a Google search. (In what follows, I shall focus on Google to the exclusion of other search engines, because the vast majority of searches in this country are conducted using Google.) The Google search results display, on a page by page basis, possible answers to the searcher’s request. Those pages are known as “Search Engine Results Pages”, or SERPs.
Oneflare, like other internet-based businesses, does what it can to improve its “page ranking”; to ensure that its website appears on the first page of the search results, and, preferably, towards the top of that page. To that end, it uses techniques known as “Search Engine Optimisation” (SEO) to improve its chances of being listed favourably in answer to a Google search. Oneflare engaged the first defendant (Mr Chernih) to provide SEO services. He did so, at first on his own account and, later on, through his company the second defendant (Linkbuild).
Oneflare says that on Mr Chernih’s advice, it adopted SEO techniques that were contrary to guidelines published by Google for webmasters, and that, as a result, it was penalised. It claims damages based on the loss of revenue that it says is attributable to the defendants’ activities. That claim is brought in contract and, in tort. A case based on statutory proscriptions of misleading or deceptive conduct was dropped in final submissions.

The real issues in dispute

Slightly rephrased, the agreed real issues are:

what were the terms of Oneflare’s retainer of Mr Chernih (since there is no relevant distinction between Mr Chernih and his company Linkbuild, I shall refer only to him unless it is necessary to distinguish between the two defendants)?

Did Mr Chernih devise and recommend the SEO strategy that Oneflare in fact pursued over the relevant period?

Did Mr Chernih use appropriate skill and competence in formulating any SEO strategy that he recommended to Oneflare?

If Mr Chernih did not use appropriate skill and competence, what if any damage has Oneflare proved?

That commendably brief statement of the real issues in dispute conceals a multitude of sub-issues. There is little point in mentioning them at this stage.
The issues do not make any reference to two “pleaded” defences. One raised contributory negligence (ss 5R and 5S of the Civil Liability Act2002 (NSW)). The other raised proportionate liability (Pt 4 of the Civil Liability Act). Mr Robertson of Counsel, who appeared for the defendants, confirmed that those defences were not pressed.

The witnesses

Witness of fact

Oneflare called four witnesses of fact. They were Mr Adam Dong, Mr Marcus Lim, Mr James Martin and Mr Ken Tabuki. The defendants called as their sole witness of fact Mr Chernih.
Messrs Dong and Lim are the principals of Oneflare. Mr Dong had a background in website development. He was, at the relevant time, Oneflare’s Chief Technology Officer. Mr Lim’s background appears to have been in business. However, it is clear that he was computer-literate. He was, at the relevant time, Oneflare’s Chief Executive Officer. Messrs Martin and Tabuki were employees of Oneflare.
I start by observing that each of the three principal witnesses (Messrs Dong, Lim and Chernih) had a significant interest in the outcome of the case. Each was, in my view, keenly aware of that fact. And in my view each was, consciously or otherwise, influenced by that fact in giving his evidence.
Messrs Dong and Lim were cross-examined extensively, and their credibility was put in issue. Mr Chernih, too, was cross-examined extensively, and his credibility was put in issue. Mr Martin was cross-examined briefly, but in a way that did not suggest that his credibility was in issue. Mr Tabuki was not cross-examined.
Messrs Dong and Lim swore several affidavits. It became clear that, in those affidavits, they sought to downplay their knowledge and deployment of SEO techniques, in particular, the technique of “linkbuilding” (as to which, see at [42] and following below). I have come to the conclusion, on the whole of their evidence, that each of them knew, and did, far more than he suggested in his affidavit evidence as to those topics. In my view, each sought to downplay his knowledge and actions because he feared that the truth would be prejudicial to Oneflare’s claim.
Each of Messrs Dong and Lim was very concerned to ensure that the Court should understand, and conclude, that all SEO actions taken by them or by Oneflare’s employees at their directions were undertaken “under the instruction” or “under the advice” of Mr Chernih. Their concern to ensure that the Court understood this aspect of their case was such that they sought to insert that qualification, or rider, on numerous occasions in the course of their cross-examinations.
On most occasions when they did so, that rider could not be thought to have been responsive to the particular question asked. The liberality of use of those forced responses in cross-examination stands in stark contrast to the absence of any such evidence in the affidavits in chief. Some attempts were made to rectify that startling deficiency in the affidavits in reply. Those attempts were manifestly inadmissible and, on objection being taken, were rejected.
In similar vein, each of Messrs Dong and Lim was concerned to suggest that he did not give instructions to Mr Chernih as to particular SEO activities to be undertaken but, rather, relied on him to tell them what to do. On occasions, this evidence was flatly inconsistent with contemporaneous documents. That led each from time to time to insist that he did not understand the content of emails from the other or from Mr Chernih; but both the form of those emails and the contemporaneous circumstances surrounding them suggest very strongly, and I find, that he did.
In my view, the evidence to which I have referred in the three preceding paragraphs was dishonest. And in each case, it is dishonest as to fundamental parts of Oneflare’s case.
There were other specific instances of what in my view was unsatisfactory evidence on the part of Messrs Dong and Lim. There were, further, instances of conflict between them. I shall refer to those matters, to the extent necessary, when dealing with the issues.
Having listened to each of Messrs Dong and Lim in the witness box, and having reread their affidavits and reread the transcript of their cross-examinations, I have come to the view that their evidence, in particular of their states of knowledge or understanding of SEO strategies and of their conversations with Mr Chernih, cannot be accepted to the extent that it was put in issue. That is of some significance in this case, bearing in mind that Oneflare relies (among other things) on oral representations said to have been made by Mr Chernih and on a contract said to arise from discussions between Messrs Dong and Lim, on behalf of Oneflare, and Mr Chernih.
In those circumstances, I proceed on the basis that the assessment of the key evidence should be guided by what McClelland CJ in Eq said in Watson v Foxman (1995) 49 NSWLR 315 at 318-319, in relation to “a party relying upon spoken words as the foundation of a cause of action…”. (His Honour was talking of a cause of action based on misleading or deceptive conduct in contravention of s 52 of the Trade Practices Act1974 (Cth), but what he said is equally apposite to a cause of action based on an oral contract.) I set out the relevant passage:

Where, in civil proceedings, a party alleges that the conduct of another was misleading or deceptive, or likely to mislead or deceive (which I will compendiously described as “misleading”) within the meaning of s 52 of the Trade Practices Act 1974 (Cth) (or s 42 of the Fair Trading Act), it is ordinarily necessary for that party to prove to the reasonable satisfaction of the court: (1) what the alleged conduct was; and (2) circumstances which rendered the conduct misleading. Where the conduct is the speaking of words in the course of a conversation, it is necessary that the words spoken be proved with a degree of precision sufficient to enable the court to be reasonably satisfied that they were in fact misleading in the proved circumstances. In many cases (but not all) the question whether spoken words were misleading may depend upon what, if examined at the time, may have been seen to be relatively subtle nuances flowing from the use of one word, phrase or grammatical construction rather than another, or the presence or absence of some qualifying word or phrase, or condition. Furthermore, human memory of what was said in a conversation is fallible for a variety of reasons, and ordinarily the degree of fallibility increases with the passage of time, particularly where disputes or litigation intervene, and the processes of memory are overlaid, often subconsciously, by perceptions or self-interest as well as conscious consideration of what should have been said or could have been said. All too often what is actually remembered is little more than an impression from which plausible details are then, again often subconsciously, constructed. All this is a matter of ordinary human experience.
Each element of the cause of action must be proved to the reasonable satisfaction of the court, which means that the court “must feel an actual persuasion of its occurrence or existence”. Such satisfaction is “not … attained or established independently of the nature and consequence of the fact or facts to be proved” including the “seriousness of an allegation made, the inherent unlikelihood of an occurrence of a given description, or the gravity of the consequences flowing from a particular finding”: Helton v Allen (1940) 63 CLR 691 at 712.
Considerations of the above kinds can pose serious difficulties of proof for a party relying upon spoken words as the foundation of a causes of action based on s 52 of the Trade Practices Act 1974 (Cth) (or s 42 of the Fair Trading Act), in the absence of some reliable contemporaneous record or other satisfactory corroboration.

There were aspects of Mr Chernih’s evidence that gave me some cause for concern: for example, his evidence on the topic of building a “linkwheel” (see, generally, T176-181) and his evidence on the topic of “scraped content” (T212-213). There were other attacks on Mr Chernih’s credibility that misfired. For example, Mr Lucarelli of Counsel, who appeared for Oneflare, attacked Mr Chernih’s evidence on the topic of “automated linkbuilding”. On analysis, it became clear that Mr Lucarelli had not been sufficiently precise in his question (because it covered several levels of links), and that Mr Chernih’s answer was correct in respect of linkbuilding at one level.
In assessing the impact of the matters to which I have referred (and many of the others on which Mr Lucarelli relied) on Mr Chernih’s credibility, one important thing must be borne steadily in mind. That is, that in almost every case, the attack was mounted by reference to evidence that Mr Chernih had given in his extremely lengthy and detailed principal affidavit, or that he had adduced in the form of emails contained in the exhibit to that affidavit. For example, in relation to the linkwheel issue referred to in the previous paragraph, the attack that Mr Lucarelli mounted was based on the discrepancy between Mr Chernih’s evidence in chief (which had volunteered all the relevant material) and his failure of recollection in the witness box. Whilst that might be taken to indicate that Mr Chernih’s recollection, at least in the isolation of the witness box, was faulty, it does not indicate that he was seeking dishonestly to conceal relevant material from the Court. The same may be said of many of the other matters on which Mr Lucarelli relied.
Finally, in fairness to Mr Chernih, I should record that my impression of him is that he is a man who by nature took a precise and literal approach to questions put to him. In circumstances where (as happened frequently) questions were embellished (and confused) by adjectival comment, Mr Chernih’s disagreement with what might appear to have been fundamentally a correct proposition of fact reflected, on examination, his disagreement with the adjectival embellishment of that fact.
On balance, although there are some aspects of Mr Chernih’s evidence that are difficult to accept without hesitation, I conclude that he did not seek to mislead the Court. On the contrary, reviewing the whole of his evidence, I conclude that he sought to tell the truth. Certainly, it could not be suggested that Mr Chernih sought to conceal matters that could be thought to be unfavourable to his case. Overall, to the extent that there is a conflict between Mr Chernih’s evidence and that of Messrs Dong and Lim, I prefer Mr Chernih’s.
Before leaving this topic, I should refer to one episode that in my view does no credit to any of the individuals involved: Mr Dong, Mr Lim and Mr Chernih. That relates to actions taken by Oneflare when Google had imposed a “manual penalty” on Oneflare (see at [65] and following below). Messrs Dong, Lim and Chernih collaborated in the preparation of Oneflare’s response to that manual penalty. Mr Chernih was the originator of the idea, but Messrs Dong and Lim were happy to go along with it. As I state at [70] below, the response was patently false. I am satisfied that each of them was prepared to collaborate in the preparation and dispatch of that response, knowing that it was false in material respects, because he perceived it would suit his particular business interests to do so.

Expert witnesses

Each party called expert evidence. Oneflare called an accountant, Mr Shannon Cavanagh, on the question of damages. It called also Mr Andre Weyher, an expert in SEO techniques (and a former employee of Google).
The defendants did not call expert accounting evidence. They did however call Mr David Hamann, who is also an expert in SEO techniques (although not a former Google employee).
Mr Robertson attacked Mr Weyher’s independence, in a manner that I think was intended to undermine his credibility. Since in my view the attack did not succeed in demonstrating any relevant want of independence, and did not cause me to form an unfavourable view of Mr Weyher’s credibility, I shall not go to the detail.
The evidence of the experts is to be assessed in each case by reference to its intrinsic merits, not by reference to any considerations of “credibility” in the wider sense.

Some technical matters

The aim of SEO techniques is to increase the amount of “organic traffic” to a website. Organic traffic is traffic that is directed to the website as a result of a search undertaken using a search engine such as Google. It is to be contrasted with traffic generated through the use of paid advertising.
The Google search engine is driven by a complex and confidential algorithm, which is very frequently updated. Some updates are minor; others are substantial. The essential function of the search engine is to take keywords identified by searchers in their search queries and, by an analysis of those keywords and of websites that Google has “indexed”, to direct the searchers to what the search engine considers to be relevant websites.
At all relevant times, Google published what it called “Webmaster Guidelines” (Guidelines). The Guidelines said to have been current as at 2007 contained the following statement:

Following these guidelines will help Google find, index and rank your site. Even if you choose not to implement any of these suggestions, we strongly encourage you to pay very close attention to the “Quality Guidelines”, which outline some of the illicit practices that may lead to a site being removed entirely from the Google index or otherwise penalized. If a site has been penalized, it may no longer show up in results on Google.com or any of Google’s partner sites.

Essentially similar statements appeared in the Guidelines said to have been current as at 2013 and as at 2015, although the reference to a site’s being “penalized” was changed to refer to the site’s being “otherwise impacted by an algorithmic or manual spam action.”
The reference to “algorithmic or manual spam action” in the 2013 and 2015 Guidelines needs some explanation. Apparently, if the Google algorithm detected a website using what it regarded as “illicit practices”, the algorithm might automatically demote that website in the page rankings. As a result, a site which had appeared, for example, high on the first page of the search results might be demoted a number of places so that it did not appear until the second or a later page. No human intervention was necessary to trigger an algorithmic spam action.
There was some dispute as to whether algorithmic spam action should be described as an “algorithmic penalty” or by some other term. To my mind, this semantic issue is devoid of any relevance to the real issues in this case. However, to avoid controversy, I note that the SEO experts agreed that an algorithmic spam action was not a “penalty”. Accordingly, adopting Mr Weyher’s terminology, I shall use the phrase “algorithmic demotion”.
A “manual spam action” was a step taken by an individual in the employ of Google: someone like Mr Weyher, who had been engaged in the creation and implementation of Google’s organic search and spam policies, and who had contributed to the development of updates to the algorithm driving the Google search engine. Manual spam action could take a number of forms. The detail does not matter, but the outcome was, at best, to demote the website in the SERPs and, at worst, “de-indexing”.

The Guidelines included guides to design and content, and to quality. In general terms, those guides were intended to persuade webmasters to improve their site’s search results by improving “on-page” quality: the quality of content on their website, so that (among other things) other sites would link to it. The guidelines also sought to discourage webmasters from engaging in “tricks intended to improve search engine rankings”. Those “tricks” included “automatically generated content”, “participating in link schemes”, “creating pages with little or no original content” and “loading pages with irrelevant keywords”.
That itemisation of “tricks” was taken from the 2013 Guidelines. However, those aspects of the 2013 Guidelines appear to have been constant features of Google’s Guidelines both before and after that year. As Mr Weyher put it, in a paragraph of his report with which Mr Hamann agreed, the Guidelines “are changed from time to time but have not been modified significantly for many years”.
Mr Robertson submitted that Oneflare’s evidence was defective because it did not prove the relevant version of Google’s Guidelines. Specifically, he submitted, the material did not prove that the Guidelines at the relevant time (April 2012 to June 2013) identified link schemes as an SEO technique to be avoided. That is not correct. It is apparent that from 2007, the Guidelines had warned against the use of link schemes. In any event, this does not seem to be particularly important, because Mr Chernih acknowledged that as at April 2012, he was aware that the Guidelines warned against the use of link schemes (T174.26-.34; T175.8-.11).
Although it is a slight diversion from the present topic, I should set out at this point the whole of Mr Chernih’s cross-examination on the relevance of Google’s warning against the use of link schemes (T175.8-.49):

Q.   Were you aware as at April 2012 that Google was telling webmasters in what I suggest is fairly clear and simple language no [sic] to engage in link schemes but instead to engage in quality content. Correct?
A.   Yes.
Q.   Those statements by Google, the warning that link schemes are in violation of the webmaster guidelines, and the recommendation or focus on quality content were highly relevant, were they not, to an SEO professional as at April 2012 seeking to improve a client’s organic search results. Correct?
A.   No.
HIS HONOUR
Q.   Why not, Mr Chernih.
A.   I just objected to the use of “highly relevant” It was -
Q.   They might be relevant, but not highly relevant.
A.   Yeah, not highly relevant. That’s right, your Honour.
Q.   To put it another way, an SEO professional in your position in April 2012 would have had to considers these matters -
A.   That’s right, your Honour.
Q.   - and then make a decision based on knowing Google’s attitude -
A.   That’s right, your Honour.
Q.    - whether, and if so to what extent, to use a link scheme to boost page ranking.
A.   That’s right, your Honour. This is what Google wanted you to do.
Q.   Google being the beast it is and was, if you don’t do what it wants, you’re likely to be punished, aren’t you?
A.   I wouldn’t say likely. No. Potential to be punished.
Q.   If they catch you.
A.   Yes.
Q.   In effect, link schemes are only good so long as you don’t get caught out.
A.   Yes. Yes, your Honour.
Q.   That was so in 2012.
A.   Yes, your Honour.

Two things flow from that. One, harking back to what I said at [22] above, is that Mr Chernih’s disagreement with what might have seemed to be a straightforward proposition was based not on the essence of the proposition but on its adjectival embellishment. The other, and more significant (in terms of the issues) is that it demonstrates both Mr Chernih’s awareness, at the outset of the retainer, of Google’s attitude to linkbuilding schemes and his understanding that such schemes were only useful if (and for as long as) Google did not detect them.
In SEO parlance, SEO strategies that are consistent with the Guidelines are often referred to as “white hat” strategies, whereas strategies that are inconsistent with those Guidelines (or that employ “illicit practices”) are referred to as “black hat” strategies.
The practice of linkbuilding is intended to take advantage of a particular operational feature of the Google search engine. Mr Weyher said, at paras 6 and 7 of his report (he did not number the paragraphs individually, but I have used the numbers allocated to them by Mr Weyher and Mr Hamann in their joint report):

6. The quantity and quality of links to a given website determine that websites ranking in organic search results. “Link building” is the process of obtaining and acquiring links from other pages on the internet in order to improve organic traffic. This is one of the key areas of SEO and an integral part of any SEO campaign.
7. Search engines register these links and count them as a ‘vote of confidence’ in the website that acquired them. The higher the quality of the site that link to another, the higher the value of the link. Acquiring links from high-end websites can be compared to a Public Relation campaign, an example of this would be writing an article about the topic of ones website that might be of interest to a journalist working for “news.com.au”. If the article is hosted on “news.com.au”, the author would often receive a link to his page as he would be seen as an authority on the subject he has written about. Having a large amount of links like these to the authors website will result in higher rankings on Google and other search engines.
A “natural link” is a link which is the product of an authentic editorial decision made by a real human being. The greater the online reputation of the site that provides the natural link, the greater the effect will be to a websites ranking in organic search results.

Mr Hamann agreed with those paragraphs, but added that linkbuilding could also be achieved by “creating links by adding content with links on other webpages”. Mr Weyher agreed with that qualification.
Mr Weyher described a link scheme, and the concept of an artificial or unnatural link, at paras 8 to 11 of his report (and again, Mr Hamann did not disagree with those paragraphs):

8.   Google defines a link scheme as “any links intended to manipulate PageRank or a site’s ranking in Google search results may be considered part of a link scheme and a violation of Google’s Webmaster Guidelines” (“the guidelines”).
9.   Acquiring a vast amount of links from low quality websites may provide some short-term SEO benefits but such benefits will only last until a Google algorithm and/or reviewer detects that the scheme is a violation of the guidelines. SEO techniques that deliberately violate the guidelines are called “Black Hat”.
10.   Google’s guidelines prohibit unnatural links because they provide no useful or valuable information to anyone and are effectively just ‘internet pollution’ that only exists in order to trick Google’s PageRank algorithm.
11.   An artificial or unnatural link is a link that is being created with nothing else in mind than SEO value. Google counts links as an organic ranking factor but wants these links to be, first and foremost, of value to the website visitor and placed by the webmaster for authentic, value-based reasons.

Google updates its algorithm very frequently, in part to defeat the activities of black hat SEO strategists. Of particular significance in this case is the “Penguin” series of updates. Penguin 1.0 was introduced in April 2012. A refined version, known as Penguin 2.0, was introduced on 22 May 2013.
Mr Weyher said, in a paragraph of his report with which Mr Hamann agreed, that the Penguin updates were intended “to automatically filter out websites that were violating the [Guidelines] by causing offending websites to be demoted from Google’s organic search results”. Mr Weyher added (and again Mr Hamann agreed) that Penguin 2.0 took into account specifically “on page content and the quality of links flowing to the page [his emphasis]”.
The Penguin 2.0 update is of particular significance in this case because, as is common ground, it was the introduction of that update that led to an algorithmic demotion very shortly thereafter. In short, the effect of the Penguin 2.0 algorithmic demotion was to push Oneflare’s website off the first page of the search results.

Factual background

Messrs Dong and Lim founded a business called “PickAQuote”. Mr Dong created, or “built”, its website, PickAQuote.com.au. PickAQuote operated essentially as Oneflare now operates.
Oneflare was incorporated in November 2011. Thereafter, it took over the operation and control of the PickAQuote website. At all material times, Messrs Dong and Lim have been shareholders in, directors of, and the principal executive officers of, Oneflare.
On 21 January 2012, the PickAQuote website was, in the words of Mr Lim, “rebranded” (T68.34). The precise nature of this rebranding is obscure. Oneflare’s case is that as a result of the rebranding, there were two separate websites, and that any inquiries to the old PickAQuote website were automatically redirected to the Oneflare website.
Prior to April 2012, Messrs Dong and Lim had arranged for some SEO work to be done. Presumably, until the Oneflare website was launched in January 2012, that work would have been done for the PickAQuote website. A “freelancer from eLance” was engaged to carry out SEO work in June/July 2011. Between January 2011 and April 2012, a company known as Arrow Internet Marketing was retained to do SEO work. Each of Mr Dong and Mr Lim sought to downplay the significance of this. Each suggested that the work had been arranged by the other. Mr Lim was less than clear as to whether Arrow Internet Marketing did work on the PickAQuote website, the Oneflare website or both.
Although it is a little out of chronological sequence, I note at this point that on 28 May 2012, Mr Lim sent Mr Chernih an email attaching an excel spreadsheet titled “LinkBuilding_Report_Oneflare.xls”. Mr Lim’s email suggested much deeper involvement in SEO work than his evidence acknowledged. It said, among other things:

Moving forward, I was thinking of having some process around tracking and reporting for SEO.
I have attached a spreadsheet that we previously used with other SEO companies to track our progress with them.

In about March 2012, Messrs Dong and Lim learned of a business in America, known as Thumbtack, which had a business model similar to Oneflare’s. They became aware that Thumbtack had improved its business very substantially by using SEO strategies. They decided to seek expert help to assist Oneflare in developing a similar SEO strategy.
At that time (March 2012), Mr Lim at least had an understanding that SEO strategies could encompass both onpage and offpage activities. He understood that those strategies included linkbuilding, and that linkbuilding was an offpage SEO strategy.
As at April 2012, Mr Chernih was a person who had had experience in SEO techniques. He had just set up his own business, offering his services as consultant to whomever might need them. Mr Jonathon Low, a mutual friend of Messrs Lim and Chernih, gave the latter’s name to the former.
Messrs Dong, Lim and Chernih met on 27 April 2012. They discussed the services that Mr Chernih might be able to provide to Oneflare. There is some dispute as to precisely what was said (both then and in later meetings). I shall return to that dispute when dealing with the first issue. What is clear is that Mr Chernih undertook to provide some detail as to what he could do for Oneflare, and Messrs Dong and Lim undertook to provide Mr Chernih with some information to help him understand Oneflare’s needs.
After that meeting (and on the same day), Mr Lim sent Mr Chernih an email. The email attached an excel spreadsheet. The spreadsheet showed a report on Thumbtack’s website produced by a software program known as “Keyword Spy”. That program can be used to identify the keywords used on a competitor’s website that rates highly in search results. Of course, Thumbtack and Oneflare were not competitors. However, there is no doubt that Messrs Dong and Lim hoped to adapt Thumbtack’s SEO strategy. They thought that sending the keyword report would enable Mr Chernih to undertake some analysis for their benefit.
Mr Chernih worked on the material given to him, and reported by email to Oneflare. Oneflare provided him with more information. There was a further meeting on 3 May 2012. Again, there is some dispute as to precisely what was said. Again, I shall return to that dispute.
On 9 May 2012, Mr Chernih sent an email to Mr Lim attaching a “Budget” spreadsheet. That spreadsheet set out a budget for various items of work that were briefly described, and the projected returns to Oneflare.
After some further emails were exchanged, Oneflare retained Mr Chernih to provide SEO services. That was done by email dated 17 May 2012. Omitting formal parts, the email read:

It’s been a week since we spoke, I tried calling you earlier but I think you might have been busy.
The team has decided to go ahead with the SEO initiative and would like to commence at your earliest convenience.
As a trial to understand the up take of SEO we would like to spend on the initial set up cost ($2,000) and $1,000 retainer for the first three months.

Mr Chernih replied the same day saying, relevantly:

Sorry you couldn’t get through, but glad we can get to working together. I’ll start setting things up ASAP.

The services that Mr Chernih provided were essentially linkbuilding services. Among other things, he suggested that Oneflare should buy “aged domains”. The intention was to post material onto the websites of those domains, which material would include links back to Oneflare’s website. As will become apparent, that was regarded by Google as a “black hat” tactic, and one that it sought to discourage.
On 23 May 2013, Mr Dong became aware that the amount of organic traffic to Oneflare’s website had diminished substantially. In the two weeks preceding 23 May, there were, respectively, 22,451 and 25,498 visits. In the two weeks following, there were, respectively, 17,968 and 17,643 visits. Thereafter, until late July 2013, organic traffic fluctuated from a low of 15,481 visits to a high of 18,227 visits. It is now accepted that the diminution in traffic was a result of the Penguin 2.0 algorithmic demotion of Oneflare’s website.
Mr Dong asked Mr Chernih for advice. Mr Chernih provided some recommendations, which Messrs Lim and Dong considered. In the course of doing so, Mr Dong said, he became aware for the first time of the Guidelines. He said that he then recognised that the SEO strategies that had been implemented were in contravention of aspects of those Guidelines. There were further discussions and meetings, and again there is some controversy as to what was said.
On 26 July 2013, Google sent an email to Oneflare through the “Webmaster Tools” portal. That email stated, among other things, that:

Google has detected a pattern of artificial or unnatural links pointing to your site. Buying links or participating in link schemes in order to manipulate PageRank are violations of Google’s Webmaster Guidelines.
As a result, Google has applied a manual spam action to oneflare.com.au/. There may be other actions on your site or parts of your site.

The email set out “recommended action” that Oneflare might take. It stated that once the site was found no longer to be in violation of the Guidelines, Google would revoke the manual action.
Following the imposition of the manual penalty, the amount of weekly organic traffic to the Oneflare website dropped almost to zero. When the manual penalty was lifted, organic traffic recovered somewhat, but to nothing like the levels that had been enjoyed prior to 23 May 2013.
Again, it appears by now to be common ground that the drop in (or virtual disappearance of) organic traffic following 26 July 2013 was a result of the imposition of the manual penalty.
Messrs Dong, Lim and Chernih met. They decided to respond to the manual spam notice. Although there is some controversy as to who prepared the request for reconsideration, I come to the view, on balance, that it was Mr Chernih who was, at least in substance, the author. However, I conclude that Messrs Dong and Lim participated in its preparation, and approved of its being provided to Google.
Among other things, the response blamed the problems on “our SEO Company, Wildnet Technologies”, which company “[had] subsequently been dismissed”. That was false. Oneflare had not engaged the services of an SEO company known as Wildnet Technologies. So far as the evidence goes, Wildnet Technologies was a fictitious name. The only person from whom Oneflare had obtained SEO services was Mr Chernih.
Google rejected the reconsideration request. Mr Dong undertook further work himself. On 11 August 2013, he sent his own reconsideration request to Google. Google agreed. On 15 August 2013, it lifted the manual penalty.
At about this time, Oneflare and Mr Chernih parted company. Oneflare said that Mr Chernih should work without remuneration to rectify the damage caused by what it said were his SEO strategies. Mr Chernih was prepared to do that work, but only on condition that Linkbuild’s outstanding invoice was paid, and that Linkbuild would be paid for the further work that he did.
Mr Dong continued to work to remove “toxic links”, and to disavow those links that could not be removed. However, on 21 November 2013, Google imposed a further manual penalty. The defendants say that Oneflare should not be permitted to rely on that second manual penalty, because it forms no part of Oneflare’s pleaded case, either as to liability or as to damages.
It is factually correct to say that the manual penalty of 21 November 2013 is not mentioned in Oneflare’s Commercial List Statement. Nor was it mentioned in the reports (individual or joint) of the SEO experts. It was however dealt with by Mr Cavanagh, who sought to calculate its monetary consequences to Oneflare.
In my view, Oneflare should not be allowed to press any claim for damages based on the imposition of the manual penalty on 21 November 2013. In circumstances where the defendants’ attention was not drawn to the matter by any pleading, there is I think a real risk of injustice, having regard to the facts as to how that manual penalty was triggered. I turn to those facts.
Oneflare and Mr Chernih parted company in August 2013. Mr Dong said that Mr Chernih last worked for Oneflare on about 6 August 2013.
In September 2013, either Mr Dong or Mr Lim bought an “aged domain” known as ojr.org, and one or other of them thereafter uploaded content to that domain. Mr Dong said that it was the existence of that domain that “attributed” to the imposition of the second manual penalty on 21 November 2013 (T60.29-31). He sought to characterise this as a “legacy strategy” (T60.41). He then gave the following evidence (T60.50-61.6):

Q.   In about December 2013 you made public comments by which you stated that an external consultant purchased the domain ojr.org on behalf of Oneflare?
A.   Yes.
Q.   That was untrue, wasn’t it?
A.   Yes.

Mr Dong was then re-examined as to what he meant by “it was a legacy issue”. He said (T61.38-62.4):

Q.   What do you mean by that?
A.   Essentially this was a technique that Nick had taught us during his time. After the penalty we had made the mistake internally of thinking that the issue – I mean, that technique would still work.
Q.   What technique?
A.   The technique of purchasing aged domains.
Q.   You gave some answer concerning making public comments attributing the acquisition of that ojr.org aged domain to an external consultant. Do you remember that?
A.   Yes
Q.    Was that something else you had learned during the period of the retainer?
A.   No, that was simply referring to the fact that we were simply trying to say in a public statement that the reason we had bought OJR was the result of an advice previously taken on from our SEO consultant.

I regard Mr Dong’s reference to this as “a legacy issue,” and his attempts to explain it in re-examination, as being self-serving and unreliable. If that had been the true explanation for this episode, why would Mr Dong have lied about it in December 2013? In my view, this aspect of his evidence is a perpetuation of the pattern, referred to earlier, of his seeking to blame Mr Chernih for every problem that Oneflare had suffered. I add that neither his nor Mr Lim’s affidavit in chief mentioned this “legacy strategy”.
Regardless, the fact is that the apparent trigger for the imposition of the second manual penalty was the existence and content of the domain ojr.org (and, presumably, links from that domain back to the Oneflare website). If the matter had been pleaded, or otherwise properly flagged to the defendants, they would have had an opportunity of investigating it. I do not think that it is sufficient for Oneflare to say that, since Mr Cavanagh had sought to calculate damages flowing from the second manual penalty, the defendants were thereby given sufficient notice.
I add that it is not as though the lapse of time between the cessation of Mr Chernih’s SEO activities and the imposition of the second manual penalty was so brief that the “legacy” explanation might have some inherent plausibility. On the evidence, a month or more elapsed between the former of those events and the time when Mr Dong or Mr Lim purchased the domain. And another two months or so elapsed before the second manual penalty was imposed. It is very difficult to understand how Mr Dong or Mr Lim could have been so under the thrall of Mr Chernih’s discredited strategy, a month or more after Oneflare and Mr Chernih had parted company (on less than amicable terms), that he – whichever of them it was – could have continued to pursue what Oneflare now says (and then should have realised) was the discredited strategy of buying aged domains, stuffing them with content, and building backlinks from them to the Oneflare website.
I note, and take into account, that although Messrs Dong and Lim swore numerous affidavits, containing many pages and hundreds of paragraphs of texts, the only reference to the second manual penalty is in the affidavit of Mr Dong affirmed 12 February 2016, where he sought to say that the manual penalty “was a result of Mr Chernih’s SEO strategy and action that had previously been taken on Mr Chernih’s advice”. That manifestly inadmissible sentence was objected to and rejected. Although Mr Cavanagh prepared his damages report in February 2015, none of the affidavits affirmed by Mr Dong or Mr Lim thereafter made any attempt to prove (at least, in a way that approached the outer limits of admissibility) the facts relevant to the inclusion of that manual penalty in their case against the defendants.
That omission is all the more surprising when one takes into account the fact that a topic specifically covered in Mr Dong’s last affidavit (affirmed 12 February 2016) is “link-clean up from 6 August 2013”.
In all the circumstances, I conclude, as I have indicated, that it would be manifestly unfair to the defendants now to permit Oneflare to claim damages in respect of the second manual penalty.

First issues: the retainer and its terms

The relevant principles

This is a case where the parties (Mr Chernih and Oneflare) undoubtedly reached a binding contract. That happened no later than 17 May 2012, when Mr Lim emailed Mr Chernih, and Mr Chernih replied, in the terms set out at [60], [61] above.
Thus, although the parties struck a contract, and agreed on the principal terms of their bargain, the full identification of the subject matter of their contract, and the identification of all its terms, is not recorded in any document (nor, for that matter, apparent from any conversation or set of conversations). So much, I think, was common ground.
Each of Mr Lucarelli and Mr Robertson submitted that, in those circumstances, it is both legitimate and necessary to have regard to subsequent conduct to ascertain the full scope of the contract and the full extent of its terms. In my view, that is correct.
In County Securities Pty Ltd v Challenger Group Holdings Pty Ltd [2008] NSWCA 193, the Court of Appeal was required to consider (as Spigelman CJ put it at [2]) “a contract partly in writing, partly oral and partly to be inferred from conduct”. In those circumstances, Spigelman CJ said (again at [2]), reliance on conduct is necessary “because important aspects of the agreement were not the subject of express statements, either written or oral, in a form which could constitute the making of an agreement”. His Honour said at [20] to [25] that where a contract is not wholly in writing, it is legitimate to have regard to subsequent conduct to ascertain its subject matter (including identifying necessary but not expressed terms).
It will be seen that his Honour’s approach is not far removed from that of Heydon JA in Brambles Holdings Ltd v Bathurst City Council (2001) 53 NSWLR 153 at [81] and following. Indeed McColl JA, who wrote the principal judgment in County Securities, referred to Brambles at [151], [153].

The first meeting, 27 April 2012

Mr Chernih had a LinkedIn profile. On that profile, he described himself as a “Web Dev Guru” and a “Link Building Expert”. Under the latter heading, he advised readers of his profile that:

To get to the top, you have to use a plethora of tools, to create the variety of quality plus quantity that Google loves.

The obvious inference is that Mr Chernih, either in his guru capacity or in his expert capacity, was holding himself out as an appropriately skilled person to create the requisite “variety of quality plus quantity that Google loves”.
The first meeting between Messrs Dong, Lim and Chernih occurred on 27 April 2012. Mr Lim gave evidence in chief, through his primary affidavit, as to what was said at this meeting. So did Mr Chernih.
Since Mr Dong was present at the meeting, one would expect him to have given evidence in chief of what was said at it. He did not do so. There was no explanation. In the circumstances, there is available an inference of the kind to which Handley JA referred in Commercial Union Assurance Company of Australia Limited v Ferrcom Pty Ltd (1991) 22 NSWLR 389 at 418.
In view of Mr Dong’s silence on the topic of what was said at the meeting, his cross-examination on this point was limited. However, Messrs Lim and Chernih were cross-examined extensively. For the reasons that follow, I conclude that Mr Chernih’s account of the meeting is to be accepted in preference to that given by Mr Lim, to the extent that the two accounts differ in substance. In reaching that conclusion, I do rely on the fact that Mr Dong could have given, but did not give, evidence in chief on this topic.
According to Mr Lim, Mr Chernih, in the course of discussing his background, said that he was doing independent contract SEO work “with specific focus on link-building”. Mr Lim said that Mr Dong asked “what’s link-building?” and that Mr Chernih gave an explanation.
Mr Chernih denied that Mr Dong had asked that question. I find that Mr Dong did not do so. There are several reasons.
First, as I have said, Mr Dong did not give that evidence. He did not say that he asked the question “what’s link-building?”.
Secondly, in my view, Mr Dong was well and truly aware of what was involved in linkbuilding. He strove to conceal the extent of his knowledge, and to convey the impression that he knew nothing about this before 27 April 2012. He said in evidence (both in his principal affidavit and in cross-examination) that before about May 2012, Oneflare had not engaged in linkbuilding. That evidence was directly inconsistent with evidence given by Mr Lim.
Ultimately, Mr Dong had to agree that Oneflare had engaged in linkbuilding prior to May 2012. He sought to cover himself by saying that this had been done by Mr Lim and that he did not know of it. I do not accept that explanation. It is obvious that Messrs Dong and Lim worked closely together. It is not credible that Mr Lim would have retained (as he said he did) a consultant to engage in linkbuilding without discussing this with Mr Dong. That conclusion is reinforced by the fact that it was Mr Dong who was the “computer expert”, and Mr Dong indeed who, by October 2013 at least, regarded himself as sufficiently expert to present himself as an SEO expert.
I add that, in his affidavit in reply to Mr Chernih’s affidavit, Mr Dong denied that he said that he knew anything about linkbuilding, or that Oneflare had had some SEO work done. He accepted that Oneflare (under its previous name) had retained the services of an SEO specialist. He said, however, that the services provided did not extend to linkbuilding. For the reasons I have just given, I do not accept the first and third aspects of that evidence, which in my view were dishonest.
Third, Mr Dong agreed that, in the course of the meeting of 27 April 2012, he said words to Mr Chernih to the effect:

I was looking at your LinkedIn profile and saw that you mention that you are across all of the Google updates like Penguin and Panda.

Mr Lim said in his affidavit that Mr Dong had made a comment to that effect to Mr Chernih. However, the version of Mr Chernih’s LinkedIn profile current as at 27 April 2012 did not say anything about his being “across all of the Google updates like Penguin and Panda”. It is no doubt possible that this aspect of Mr Lim’s evidence, and of Mr Dong’s acceptance of it, could have been mistaken. However, in my view, the real explanation is that Messrs Dong and Lim were seeking to bolster their case, in circumstances where they knew that the Penguin updates (in particular) were of great significance.
Fourth, as I have said, Messrs Dong and Lim had become interested in the topic of SEO techniques because of the apparent success of the Thumbtack website. Mr Lim agreed that he had read about linkbuilding in relation to Thumbtack, believed that it was improving organic traffic to Thumbtack’s website, and that it was a strategy that Oneflare should adopt (T77.6-.15). Messrs Dong and Lim had discussed Thumbtack prior to the meeting, and were keen to emulate its success. It is not credible that before the meeting, Mr Lim would have become aware of those matters relating to linkbuilding but would not have discussed them with Mr Dong.
As I have said, I accept Mr Chernih’s evidence as to this part of the meeting. I find that Mr Dong did not ask “what’s link-building”, and that Mr Chernih did not respond as Mr Lim said he had done. I find that by 27 April 2012, Messrs Dong and Lim well understood what was involved in the practice of linkbuilding, and wanted to have it done for Oneflare.
Mr Chernih said in his account of the conversation that Mr Lim said among other things that Oneflare was trying to copy the Thumbtack business model, and that “we know they are ranking for a lot of keywords and they have a very aggressive SEO strategy”. Messrs Dong and Lim did not specifically deny this in their affidavits in reply. I find that Mr Lim did say words to the effect of those attributed to him.
According to Mr Chernih, Mr Lim followed up the reference to “aggressive SEO strategy” by saying “we want to get our business to a similar point that [Thumbtack is] at, as quickly as possible”. I find that those words were said. In my view, in using those words, Mr Lim made it clear to Mr Chernih that Oneflare wished to adopt an “aggressive SEO strategy”, of the kind adopted by Thumbtack (I interpolate, including the use of linkbuilding), to increase Oneflare’s organic traffic “as quickly as possible”.
Following and on the day of the meeting, Mr Lim sent an email to Mr Chernih which said, among other things:

As mentioned we have a counterpart [Thumbtack] in the US that has implemented an aggressive SEO strategy with over 49k organic keywords.

As Mr Lim recognised, that email did confirm Mr Chernih’s account of the conversation insofar as it dealt with Thumbtack. Mr Lim sought to downplay his use of the adjective “aggressive”. He said it “meant … a strategy that produced good results”; “a strategy that achieved results in a shorter period of time” (see generally T90-91). In my view, Mr Lim was dissembling when he said that, in an attempt to deflect the consequences that might follow from an understanding that a strategy involving linkbuilding (which to his knowledge Thumbtack had adopted) could be described as “aggressive”.
In my view, as I have said, it is clear that Mr Lim conveyed to Mr Chernih that Oneflare wanted to adopt an aggressive SEO strategy, similar to that employed by Thumbtack, which would deliver similar results.
It is also clear that the strategy that Oneflare told Mr Chernih was to be adopted included, on Oneflare’s express instructions, linkbuilding. Mr Chernih said that, in the course of this meeting, he recommended that Oneflare should “focus on making sure your on-page is right” and suggested that “[a]fter that, it is mostly about off-page, in particular link-building, building links to artificially inflate your link profile in Google’s eyes”.
According to Mr Chernih, Mr Dong replied that “we know a bit about link-building” and added that “we have had some SEO work done before, when we were PickAQuote”. I accept that evidence.
Mr Lim accepted that Oneflare had “sought out Mr Chernih to engage in linkbuilding”, and had hired him “specifically to do link-building” (T76.48-77.19):

Q.    I take it that you asked Mr Lowe [sic] whether he knew anyone who could do link building for Oneflare's website. Is that correct?
A.    Correct.
Q.    That was the basis on which Mr Chernih was hired, was it, specifically to do link building for Oneflare?
A.    Correct.
Q.    So the link building strategy was something that you had read about in respect of Thumbtack? Is that correct?
A.    Correct.
Q.    And you were of the belief that link building done by Thumbtack was driving organic traffic through Google search engines to their website?
A.    Correct.
Q.   And that was an SEO strategy that you wanted to adopt for Oneflare?
A.    Correct.
Q.    And you sought out Mr Chernih to engage in link building for Oneflare. Is that correct?
A.    Correct.

Given that this was Oneflare’s requirement, it is inherently likely that the conversation, in respect of linkbuilding, was as Mr Chernih said.
Mr Chernih said that, as a result of the meeting of 27 April 2012, he “believed that Mr Lim and Mr Dong wanted me to devise and implement an SEO strategy that was similar to that implemented by the US website Thumbtack”, and that the strategy was to include “all aspects of on-page optimisation and off-page optimisation, for the purposes of increasing the ranking of Oneflare’s website and increasing traffic flows to Oneflare’s website”. (See paras 42, 41 of his affidavit sworn 5 June 2015.) In my view, he was correct to conclude, from the conversation of 27 April 2012 and from the email of the same date, this was what he was required to do.
Mr Chernih said in his affidavit that, in the course of the conversation, he said words to the effect:

Well, its not just about the quantity of the links. You need to get good quality links as well. Google is like a popularity contest, and links are like votes. Though its not just a numbers game, it’s a quality thing too … I can build you a million links in an hour if you want, but that is more likely to get you penalised than to help your rankings … If you get penalised, you then have to get Google to restore you. If that’s an algorithmic penalty, you don’t get restored until the algorithm updates again.

Mr Chernih did not say that Mr Dong or Mr Lim questioned what he said. (Indeed, on Mr Chernih’s evidence, they did not reply at all). Messrs Dong and Lim did not specifically deny this aspect of the conversation, in their affidavits in reply. I find that Mr Chernih did say words to the effect of those that I have set out. In my view, what he said made it clear (and it must have been clear to Messrs Dong and Lim as at 27 April 2012) that linkbuilding for its own sake involved risks, including a risk that Google might penalise the Oneflare website. In my view, Messrs Dong and Lim were prepared to accept that risk.
Mr Chernih said, further, that Mr Lim said that Oneflare’s target was to get its organic traffic “up to 100,000 visits per month, within the next 12 months”. Although Mr Lim in particular appeared unwilling to accept that he did say words to that effect, it is in my view clear, in particular from a presentation prepared for potential investors, that this was Oneflare’s aim as at April 2012. It is not at all implausible that Mr Lim would have said words to the effect of those that I have set out, for the purpose of explaining what he wanted the “aggressive” SEO strategy to achieve for Oneflare. I find that Mr Lim did tell Mr Chernih, in the meeting of 27 April 2012, that Oneflare’s aim was to increase organic traffic to 100,000 visits per month within the next 12 months.
I find, specifically, that:

in the 27 April meeting, Mr Lim, with at least the acquiescence of Mr Dong, instructed Mr Chernih that Oneflare wished to pursue an aggressive SEO strategy that included both on-page optimisation (improving the content of Oneflare’s website) and off-page optimisation (including linkbuilding);

as at 27 April 2012, Mr Dong at least understood clearly what was involved in linkbuilding and if (which is unlikely, given his failure to question the reference to it) Mr Lim did not, he could have found out very easily from Mr Dong what was involved;

it was reasonable for Mr Chernih to understand from the conversation that Messrs Dong and Lim had a good understanding of SEO techniques, in particular the technique of linkbuilding;

Mr Chernih warned Messrs Dong and Lim that overindulgence in (or overenthusiastic pursuit of) linkbuilding could result in Oneflare’s website being demoted in the Google search results; and

Messrs Dong & Lim understood that warning.

The formal retainer was not concluded on 27 April 2012. However, nothing that happened thereafter up until 17 May 2012 detracts in any way from the conclusions just expressed. On the contrary, it is plain that, when Mr Chernih was retained to provide SEO advice and assistance to Oneflare, it was a retainer to assist them to pursue an aggressive SEO strategy, including both on-page and off-page optimisation, similar to that employed by Thumbtack. Specifically, as part of the off-page optimisation, Mr Chernih was retained to assist Oneflare in linkbuilding. The object of the retainer was to assist Oneflare in improving its SERP ranking.

Events from 27 April to 3 May 2012

On 1 May 2012, Mr Lim emailed Mr Chernih giving him “admin access to our analytics”. As I understand it, that was intended to assist Mr Chernih in considering Thumbtack’s SEO strategy, and seeing how it might be adapted for Oneflare. Thereafter, Mr Chernih and (mostly) Mr Lim exchanged emails, apparently to assist Mr Chernih’s consideration of “strategy and process”.
On 1 May 2012, Mr Lim sent an email to Mr Chernih asking him to come in for a meeting in two days’ time. Among other things, that email said:

I also have another question regarding the link building process and whether if we are able to activate any resources on our end e.g. our developer or customer care consultants or even Adam to help with the process of link building. Essentially if you are able to come up with the strategy and process, are we able to assist in executing the plan with our resources. I am able to activate a few more resources if needed to expedite the process to achieve quicker results if needed. Let me know if this is possible.

The second meeting took place on 3 May 2012. According to Mr Chernih, Messrs Dong and Lim were present, as were Mr Martin and a Mr Tommy Lim, who was introduced “as a shareholder of Oneflare”. Mr Martin’s affidavit did not touch on the meeting. Mr Tommy Lim was not called to give evidence. Further, neither Mr Dong nor Mr (Marcus) Lim gave evidence in chief of this meeting. In the circumstances, I accept the substance of Mr Chernih’s account of it. Mr Chernih said that in the course of this meeting there was the following exchange between him and Mr (Marcus) Lim:

Mr Lim: “So we will need to adopt an aggressive SEO strategy if we want to replicate Thumbtack’s strategy. Our traffic goal is 100,000 per month, what do you think would be a realistic spend on SEO to get to our target?”
Me: “It’s really hard to say. That is a very large increase, given present traffic volumes. I never promise that SEO will deliver any particular traffic volumes, and I wouldn’t trust anyone who would make such a promise. The return in traffic volume for an increase in spending is not linear. And the situation is always very fluid. Google changes its algorithm, sometimes 1 to 2 times a day, which affects how an SEO strategy drives traffic. And new competitors can also emerge and take traffic. SEO is an ongoing process that has to be continually updated to reflect the changing situation. But a good starting point for spending on SEO is maybe $2,000 to $3,000 per month initially, and then that would need to be reviewed after a few months.”

Events from 3 May 2012 to conclusion of retainer on 17 May 2012

On 9 May 2012, Mr Chernih sent Mr Lim a “budget” email attaching his “breakdown on what I think is achievable”. He asked for further instructions: “let me know what direction you want to go in, and I can go into more detail”.
There was an excel spreadsheet attached to that email. It contained three boxes. One was titled “Campaign”. It referred to various kinds of services that were acceptable through the Oneflare website, and predicted a “conv.” (i.e., the extent to or rate at which inquiries would mature into sales). It set out an average job value and profit for each category, and thus projected a profit “per 1000 searches”.
The second box set out levels of proposed “Budget” expenditure at levels from $1,000.00 to $20,000.00 (presumably, monthly). It projected a “Return” (at 3 months), varying according to the amount spent.
Mr Lucarelli relied on both those boxes to show that Mr Chernih recognised that there was a link between SEO optimisation, visits to the site, revenue and profit. I think that this is a reasonable inference to draw from them.
The third box, titled “Calculations”, identified the SEO activities that would be carried out at various levels of three-monthly budget expenditure.
On 17 May 2012, Mr Dong sent an email to Mr Lim asking a question “regarding the URL structure”. On the same day, Mr Lim forwarded that to Mr Chernih, with the “retainer” email referred to at [57] above.
Mr Chernih replied to Mr Lim’s email stating, among other things:

Sorry you couldn’t get through, but glad we can get to working together. I’ll start setting things up ASAP.

Mr Chernih’s email also replied to Mr Dong’s question.

“Professional”?

Mr Lucarelli sought to characterise Mr Chernih as a “professional”, and to equate his retainer to a retainer of a professional person to provide professional services. He built on that by suggesting that the implied term (and equivalent common law duty) to exercise reasonable care and skill in the execution of the retainer was a “single comprehensive duty” which included, as an incident, an obligation to warn of material risks of harm inherent in a proposed course of SEO activity. Mr Lucarelli relied upon the decision of the High Court in Wallace v Kam (2013) 250 CLR 375 at [8] (from which paragraph the earlier quotation comes).
Mr Robertson denied that his client was a “professional”, and that the duty to warn applicable to skilled medical practitioners should be imposed upon him.
Mr Lucarelli relied on the decision of Santow J in Prestia v Aknar (1996) 40 NSWLR 165 at 184-185. Mr Lucarelli focused on part of one sentence at 184, in which Santow J said, by reference to authority, the citation of which I will spare the reader of these reasons:

The “professional” has been said to possess “special knowledge which is attained after study and application, as distinguished from mere skill”…

That partial sentence comes from Santow J’s detailed consideration of various factors that have been held, in particular cases dealing with particular facts (and sometimes specifically in a legislative context), to justify describing someone as a “professional”. Further, earlier on the same page, Santow J stated that “[i]t is not possible to give a precise legal definition of the term “profession””. And a little later on the same page, Santow J pointed out that the use of the term was not rigid or static, but varied over time. Finally, the question arose for Santow J because of the definition, in s 4(1) of the Fair Trading Act1987 (NSW) of “trade or commerce” and the reference in that definition to “professional business”.
It is an inappropriate use – more accurately, a misuse – of authority to select partial statements from a decided case, made for the purpose of deciding the meaning to be attributed to the words “professional activity” in a particular statutory context, and to seek to apply what was said to the general conception of what constitutes a profession, and makes its practitioners “professionals”.
More fundamentally, the underlying reasoning seems to me to be utterly misconceived. In essence, Mr Lucarelli sought to apply a particular label to the nature of the services to be provided by Mr Chernih to Oneflare, and thus to Mr Chernih as the provider of those services, and, having done so, to submit that contractual or common law duties applicable to some persons fitting that description, in the exercise of that activity, should be implied into or imposed on Mr Chernih’s contract with Oneflare.
I accept, of course, that it is well recognised that particular terms may be implied, as a matter of law, into particular kinds of contracts. That is exemplified in the oft-cited and applied decision of the House of Lords in Liverpool City Council v Irwin [1977] AC 239. It is quite another to say that, in the innumerable kinds of contract that may be made for the provision of the numerous services that may, in particular contexts, merit the description “professional”, duties held to be applicable to one professional in one situation apply to all “professionals” in all contexts.
Moving from the general to the particular, I find it difficult to understand why the content or scope of a duty of care applicable to skilled medical practitioners, whose treatments may involve the risk of physical injury, should be applied without analysis to a contract of the kind made between Oneflare and Mr Chernih, involving no risk of physical injury. Mr Lucarelli’s submissions offered no assistance on this question.

Conclusions to date on scope of retainer

I start by referring back to what I have said at [118], [119] above. At the outset, Oneflare made it clear to Mr Chernih that it wanted him to provide SEO strategies that were “aggressive” and that included linkbuilding. I accept, of course, that the SEO strategies also included improving the content of Oneflare’s website. The reason why Oneflare wanted to pursue an aggressive SEO strategy was, as made known to Mr Chernih, that it wished to drive its sales growth in the way that Thumbtack had apparently done in America.
The request made of Mr Chernih in the email of 1 May 2012 (see at [120] above), that he “come up with the strategy and process”, is, in context, clearly a request for him to design the strategy and process that would implement the overall SEO strategies that Oneflare had told him it wished to employ.
I accept that there was an implied term of the contract of retainer that Mr Chernih would use reasonable care and skill, appropriate to one professing expertise as an SEO specialist, in performing work pursuant to the retainer. I accept, further, that Mr Chernih owed Oneflare an equivalent common law duty of care. The precise content and application of those duties is a matter to which I shall return, in connection with the third issue.
I do not find that the contractual or common law duties that Mr Chernih owed to Oneflare included a general duty to warn of the risks attaching to particular aspects of the SEO strategy that Oneflare would follow. There was certainly no express term of that kind.
Further, and as to the possible existence of a common law duty of care to warn, the expert evidence was unanimous in opining that an SEO specialist was not obliged, in all circumstances, to warn the “client” of the risks attaching to particular SEO strategies. The experts agreed that it would be reasonable for an SEO expert to take into account, in deciding whether to advise of risks, the experience of the client and its understanding of SEO strategies. In that context, the findings of fact at [118] are relevant.
The discussion in the two preceding paragraphs is confined (as I said at [142]) to “a general duty to warn”. Specifically, where (as I find) Oneflare, through Messrs Dong and Lim, made it clear to Mr Chernih that it wished to pursue a strategy involving linkbuilding, and where Mr Chernih understood (rightly) that Oneflare, through Mr Dong at least, understood both what was involved in linkbuilding and the possible risks attaching to it, there is no basis for imposing a general duty of care to warn of the risks of linkbuilding. In any event, as noted at [115] above, Mr Chernih did proffer such advice, in the meeting of 27 April 2012, where he said, among other things:

I can build you a million links in an hour if you want, but that is more likely to get you penalised than help your rankings.

Indeed, Mr Chernih went further, and gave an example of how competitors could use “negative SEO”:

Negative SEO is when someone else, like a competitor, uses SEO as a weapon against you. They attack your site to get you penalised by Google. I’ve seen it happen. A competitor can build a million links to your site in an hour and there’s nothing you can do to stop them. If you can penalised, you then have to get Google to restore you. If that’s an algorithmic penalty, you don’t get restored until the algorithm updates again.

Those words must have informed Messrs Dong and Lim (if they did not already understand) that over-enthusiastic indulgence in linkbuilding carried the risk, and probably the likelihood, of adverse action by Google. Specifically, it must have been clear to them that there was a real risk of algorithmic demotion. Neither of them questioned the advice given by Mr Chernih. Yet, having that understanding, Mr Lim stated in the meeting of 3 May 2012 that Oneflare “will need to adopt an aggressive SEO strategy if we want to replicate Thumbtack’s strategy”, and stated a “traffic goal” of “100,000 thousand [hits] per month”. Mr Chernih replied, saying that it was “a very large increase, given present traffic volumes” and that because Google changed its algorithm, “sometimes 1 to 2 times a day”, that could affect the operation of an SEO strategy.
I do accept that if Mr Chernih had recommended a new and risky strategy to Oneflare, it may have been incumbent on him to warn Messrs Dong and Lim of the risks attending that strategy (unless well satisfied, on reasonable grounds, that they understood them). But that is not this case. And it does not follow, taking into account the facts as I have found them, that Mr Chernih was bound to warn further than he did of the risks attending the strategy that Oneflare desired, and instructed him to pursue.
The precise mechanism by which Linkbuild became the supplier of SEO services in place of (but through) Mr Chernih does not require detailed analysis. Everyone agreed that, at some stage, Mr Chernih stopped supplying those services on his own account, and started to supply them on account of Linkbuild. Everyone agreed that Oneflare was aware of, and accepted, the change. Were it necessary to do so, the process by which this happened could be described as a novation.
It is not contentious that the duties owed by Linkbuild to Oneflare, whether pursuant to the contract or at common law, were the same as, and limited to, those owed by Mr Chernih.
The remaining question for consideration, under the rubric of the first issue, is whether an understanding of the scope or terms of the retainer is to be assisted by a consideration of what was done after the retainer was formed, and in pursuance of it. In other words, what inferences, as to the scope or terms of the retainer, can be drawn from its performance? The facts as to performance are the subject of the second issue.
At this point, I will simply state my conclusion. The relevant events are described in the following section of these reasons. For the reasons that I there give, I conclude that there is nothing in the conduct of the parties following 31 May 2012 that would in any way add to the scope or terms of the retainer, as I have set out at [141], [142] and [147] above, or qualify, restrict or negate any of those terms.

Second issue: performance of the retainer

The parties’ characterisation of the activities between May 2012 (when work started) and May 2013 (when the algorithmic demotion was imposed) varied widely. Mr Lucarelli undertook a detailed examination of pages and pages of emails which, he said, showed “the collaborative implementation of the linkbuilding scheme recommended by Mr Chernih”, or “the collaborative approach to implementing Mr Chernih’s linkbuilding SEO strategy”.
Mr Robertson, on the other hand, discerned from the emails and dealings four recurrent practices or patterns of behaviour:

first, that it was Oneflare, rather than Mr Chernih, that decided the SEO strategy to be followed by Oneflare;

secondly, that Messrs Dong and Lim had a good understanding of, and often implemented without any input from Mr Chernih, various black hat SEO strategies;

thirdly, that Messrs Dong and Lim otherwise implemented SEO strategies without seeking comment or advice from Mr Chernih; and

fourthly, that on the “few occasions” that Oneflare sought advice or comment from Chernih, Mr Chernih’s response “was usually in very general and broad terms and was often equivocal”.

As to Mr Lucarelli’s characterisation, it may be accepted that there was some collaboration between Oneflare (Messrs Dong and Lim) on the one hand and Mr Chernih on the other. After all, Oneflare had retained Mr Chernih for (putting it neutrally) assistance in relation to SEO strategies. However, to suggest that the pattern and course of dealings over the year in question showed “a collaborative approach in implementing Mr Chernih’s SEO strategy” is, in my view, a mischaracterisation of the evidentiary material on which that characterisation is said to be based. It is, moreover, a mischaracterisation that does not assist in identifying the core issue: who devised and implemented Oneflare’s SEO strategy.
On the other hand, I think, Mr Robertson’s characterisation (or derivation of patterns of behaviour), although closer to the mark as a summary description of the way in which Oneflare and Mr Chernih conducted their business relationship over that year, tends to minimise the role that Mr Chernih played.
The starting point, in terms of analysis although not chronologically, may be found in Mr Lim’s email of 1 May 2012 to Mr Chernih (see at [121]) above. In that email, Mr Lim said that he had “another question regarding the link building process”. He then asked whether, if Mr Chernih could “come up with the strategy and process”, Oneflare could assist in its implementation. That email, Mr Lucarelli submitted, set out the essence of the retainer.
However, to look at the email in the way that Mr Lucarelli did is to ignore the factual background against which that email was written. That background includes specifically the meeting of 27 April 2012, in which Messrs Dong, Lim and Chernih discussed what it was that Oneflare wanted and what it was that Mr Chernih could provide. I will not repeat in detail what I have found as to that meeting. The key features of it, for present purposes, are that:

Mr Dong at least, and probably Mr Lim as well, had a good understanding of SEO strategies, including in particular linkbuilding;

Oneflare wanted assistance in adapting the Thumbtack model, including its “very aggressive SEO strategy” that had linkbuilding as an element;

Mr Chernih suggested that the initial focus should be on “on-page” content (i.e., on the Oneflare website) and thereafter, on “off-page, in particular linkbuilding” should be pursued;

Messrs Dong and Lim indicated both that they were aware of what was involved in linkbuilding and that they were happy to “build more and more links”;

Mr Chernih warned them in clear and serious terms against over-excessive indulgence in linkbuilding, and of the risk that it could lead to Oneflare’s being “penalised by Google”; and

the meeting concluded, as Mr Chernih said, with his indicating that he would do some research “and then we can work towards a link-building strategy in the coming weeks”.

The quotations, and the substance other than the quotations, set out in the preceding paragraph come from Mr Chernih’s account of the meeting. For the reasons I have given, I accept his version of the meeting, in particular to the extent that it differs from Mr Lim’s account (or from Mr Dong’s account, as it was explored in cross-examination).
Against that background, the email of 1 May 2012 was at most a request for Mr Chernih “to come up with a strategy and process” for implementing the two SEO activities outlined at the meeting: improving on-page content, and working on off-page activity, or linkbuilding. And he was to do so in circumstances where it had been made clear that Oneflare wanted, or at least was prepared, to emulate Thumbtack’s “very aggressive SEO strategy”.
In short, the essential features of the SEO strategy were laid down by Oneflare at the meeting of 27 April 2012. The request (if that is what it is) made in the email of 1 May 2012, that Mr Chernih “come up with the strategy and process”, must be, in context, a request for him to design a strategy or process for the implementation of the two-pronged approach (on-page and off-page SEO) that had been outlined at the meeting four days earlier.
I add, in the context of the reference to “very aggressive SEO strategy”, that on the expert evidence, this is not an appropriate way of describing on-page optimisation. That, in Google’s eyes, is and in April 2012 was a desirable way of improving SERP ranking. It was not “aggressive”. The strategy that is “aggressive” is linkbuilding; and aggressive linkbuilding could have the adverse consequences of which Mr Chernih warned in the meeting.
I pause at this stage to note that the email of 1 May 2012 does indicate that Oneflare was proposing, in one sense of the word, to collaborate with Mr Chernih. The email proposed, in effect, that if Mr Chernih told Oneflare what to do, Oneflare would undertake the labour of doing it.
The next relevant event is the meeting of 3 May 2012. For the reasons I have given, I accept the substance of Mr Chernih’s account of that meeting. I have set out the relevant passage at [122] above. Of present relevance, Mr Lim said that Oneflare would “need to adopt an aggressive SEO strategy” if it wished “to replicate Thumbtack’s strategy”. There is little doubt, from what had been said at the earlier meeting, that Oneflare did indeed want to replicate Thumbtack’s strategy. In context, that “aggressive strategy” could only refer to linkbuilding.
On the day following the meeting, Mr Lim sent an email to Mr Chernih. Somewhat surprisingly, neither Mr Dong nor Mr Lim saw fit to refer to that email in his principal affidavit. The email asked Mr Chernih to indicate “how traffic would grow if we were to insert more money into the campaign”, or “how traffic can grow subject to increase in spend”. The purpose was said to be a presentation “to our investors”. The email then said:

As you understand we are looking to implement an aggressive SEO growth strategy e.g. we are going to import close to 100k business profiles in the next 2 weeks … We are looking to hire content writers … to build on relevance online and most probably would outsource this to eLance …

A.    Yes.
Q.   You had formulated an SEO strategy that had as its centrepiece the creation of backlinks. Correct?
A.   Yes.
Q.   You had yourself executed the strategy of creating backlinks by both manual link building and automated link building. Correct?
A.   Yes.
Q.   You had sent to the client the email we have just seen telling it how it could buy Ultimate Demon, where to get the content for the automated links, and containing a do-it-yourself, step-by-step guide on how to use Ultimate Demon. Correct?
A.   Yes.
Q.   You had been asked by the client whether the acquisition of aged domains was a good idea, and you advised that it was. Correct?
A.   Yes.
Q.    You had participated, I think you've told us, almost on a daily basis in that process of selecting appropriate age domains [sic] for the clients to acquire. Correct?
A.    At some points it was a daily, yes.
Q.    You had indeed also, to some extent, assisted Mr James Martin in the development of a program called Backlink Panda, the purpose of which was to try and harvest and present data relevant to the valuation of available aged domains. Correct?
A.    Again, I don't like to say that I helped assist because he took what I had done and then built it himself. So I didn't assist him in - since the creation of his - the idea in his head and the going live of Backlink Panda, I had no input as to Backlink Panda in its inception to its completion.
Q.    But you did speak to him about the types of - I forgot the word - the types of--
A.    Metrics.
Q.    --metrics that a useful program would contain. Correct?
A.    No.
Q.    In any event, you have told us that you used the Backlink Panda program in order to assist you in attributing a value to available aged domains. Correct?
A.    Yes.
Q.    You had, with the client, discussed the notion of scraping content and using the scraped content for SEO purposes. Correct?
A.    Rather they had discussed with me.
Q.    Fair enough. But that was discussed?
A.    Yes.
Q.    I think you've agreed with me that it was in fact done by the client to your knowledge, that is, the scraping of content for SEO purposes. Correct? To your knowledge?
A.    I'm not - I'm not convinced - I'm not sure how to say this. I'm not sure it was for SEO benefit.
Q.    Just going back a step. The aged domains had either been used for redirects to OneFlare's website or alternatively to host links contained in articles to OneFlare's website. Correct?
A.    Mostly for the latter. Yes.
Q.    Each one of those things we've just discussed, each single one of them was something that either you recommended to the client, or alternatively you discussed with the client prior to it being pursued. Correct?
A.    I wouldn't say about the scraping.
Q.    Let us scrape away the scraping issue and leave the rest. The rest of it, would you agree, was undertaken either at your recommendation or after your approval had been given to that particular strategy?
A.    Yes.

As is apparent from the first question in that passage, it was in effect a summary of all the matters that had been covered in the previous 76 pages of Mr Chernih’s cross-examination. To my mind, it is not appropriate to consider the passage stripped of its context.
Further, it will be noted, the passage starts by referring to “certain high level strategic advice and guidance” that Mr Chernih had been asked to give, and had given, in response to requests from Oneflare. The nature of the request cannot be put to one side in considering the significance to be attached to the evidence that followed. As I have said more than once, Oneflare retained Mr Chernih to devise and implement an aggressive SEO strategy that included linkbuilding, similar to the one that, as Messrs Dong and Lim knew, Thumbtack had employed in America. That explains why, as Mr Chernih also agreed, the strategy that he formulated “had as its centrepiece the creation of backlinks”, and why, as he also agreed, he both implemented and assisted Oneflare to implement that strategy.
No doubt, Mr Chernih did so knowing of the risks. That he was aware of the risks is clear from the passages to which I have referred at [204] above (and from other passages of his evidence in cross-examination). But it does not follow, simply from the fact that Mr Chernih was aware of the risks in the strategy that Oneflare retained him to implement, that he had a duty to warn in respect of those risks.

Impact of those matters on the scope and terms of the retainer

Considering the question as one of the content of an implied contractual duty to use reasonable care and skill, appropriate to one professing expertise as an SEO practitioner, the critical point is whether, in the whole context of the particular contractual relationship, that implied duty should be held to extend to a duty to warn of risks.
In the case of a medical practitioner (in particular, a specialist practitioner advising a patient to undertake a particular course of treatment), the reason why the duty owed will include a duty to warn of the risks is clear. The medical practitioner knows of those risks. In all likelihood, the patient does not. And the patient’s decision, whether to undertake the treatment, is one that cannot be made rationally unless the patient understands the risks and benefits of the treatment that is proposed.
Further, in the case of medical treatment, the patient goes to the medical practitioner for advice. The usual situation is (or, before the days of “Dr Google”, was) that the patient is not aware of the range of treatment options available. One of the purposes of consulting the practitioner is to find out what can be done. In this case, however, Oneflare did not consult Mr Chernih to find out what could be done. It retained Mr Chernih to devise and implement an SEO strategy that had as its centrepiece the creation of backlinks. Why should Mr Chernih assume that he was required to warn Oneflare of the risks inherent in such a strategy? Why should the law impose on him, either through the mechanism of an implied term in the contract of retainer or an equivalent common law duty of care, such an obligation?
If the question is whether a common law duty of care should be imposed, it could only be a duty in respect of pure economic loss. Mr Lucarelli accepted this. There are at least two fundamental points:

what is there, in all the circumstances, to make good the conclusion that Oneflare was relevantly vulnerable, and

What is the impact of Div 4 of Pt 1A of the Civil Liability Act?

The submissions left those matters substantially unexplored.
Further, the suggestion that the scope of the contractual or common law duty of care should extend to an obligation to warn flies in the face of the facts. Specifically, it ignores the warnings that, I have found (see at [115] to [118] above), Mr Chernih gave to Messrs Dong and Lim on 27 April 2012.
Mr Chernih was correct to admit, in response to the second question in the passage set out at [206] above, that he “had formulated an SEO strategy that had as its centrepiece the creation of backlinks”. That was precisely what he was retained to do; and he was retained to do it after he had warned, in what in my view were clear terms, of the risks attending that strategy.
The subsequent steps referred to in that lengthy passage of evidence (to the extent that Mr Chernih agreed with the questions put to him) were all steps undertaken in the execution of that strategy. Thus, they were all steps undertaken in the execution of the strategy that Mr Chernih had devised in response to the specific instructions that he received. No doubt, Mr Chernih was aware of the risks inherent in that strategy. It does not follow that he was obliged to continue to warn Oneflare of those risks.

Conclusions on the performance of the retainer

I return to the topic of the second issue: the performance of the retainer; specifically, who it was that designed and implemented Oneflare’s SEO strategy. The essential elements of the strategy that Oneflare employed were those laid down by Messrs Dong and Lim in their meeting with Mr Chernih on 27 April 2012. I have referred to those elements of the strategy more than once. Nothing that happened thereafter, either in the period between 27 April 2012 and 17 May 2012 or in the performance of the retainer afterwards, requires any modification of that conclusion.
My conclusion as to performance of the retainer is that Mr Chernih assisted Oneflare in the execution of the SEO strategy that it had laid down. He did this both by providing assistance to Oneflare’s employees in SEO work and by carrying out aspects of that SEO work himself. And from time to time, when he was asked for advice, he responded; although in the less than enthusiastic terms that I have recorded.
I do not accept that the steps taken by Oneflare and its employees were (as Messrs Dong and Lim sought to argue during their cross-examinations) steps undertaken at the direction or on the instruction of Mr Chernih. On the contrary, such direction or instruction as Mr Chernih gave was limited to working with Oneflare and its employees, in the way that I have described.
No doubt, Mr Chernih from time to time provided software and other tools to enable Oneflare’s employees to continue with their SEO work. But it does not follow from this that he thereby assumed direction or control of the strategy. In those instances where this happened, Mr Chernih was doing no more than assisting in the execution of the strategy whose basic and essential elements had been laid down by Oneflare.

Third issue: breach of any duty owed

On my view as to the scope of the retainer (including the scope of the duty to exercise reasonable care and skill), no question of breach arises. Mr Chernih was retained to devise and implement an aggressive SEO strategy based on linkbuilding. Mr Lucarelli did not submit that, if this were the nature of the retainer, the work done by Mr Chernih nonetheless breached his implied contractual duty or his equivalent common law duty. Nor do I see how such a submission (if made) could be sustained: at least, if it be accepted that (as I have concluded) the duty owed did not have, as a component, a duty to warn of the risks inherent in that strategy.
Of course, if the retainer were one simply to advise on and implement an “appropriate” SEO strategy, entirely different questions would arise. But that was not the scope of the retainer, and the question of breach of some hypothetical retainer is not one that need be pursued.

Fourth issue: damages

On the basis of the conclusions that I have reached so far, the question of damages does not arise. However, in case others take a different view, I shall deal with the competing contentions.

Outline of the parties’ positions

Oneflare’s case on damages, as Mr Lucarelli articulated it, depends on the following propositions:

at the relevant time, there was a direct and essentially linear relationship between traffic to Oneflare’s website and revenue generated;

the effect of the algorithmic demotion in May 2013 and of the manual spam action in July 2013 was to produce a significant drop in traffic;

as a result, there was a significant drop in revenue;

in consequence, the rate of growth that Oneflare had enjoyed (and that, on its case, it would have enjoyed had Mr Chernih devised and implemented an appropriate white hat SEO strategy) was interrupted.

Mr Cavanagh, the accounting expert who gave evidence for Oneflare on the question of damages, quantified Oneflare’s damages by reference to what he said was revenue lost as a result of the decline in both traffic and what he called “sales” following the events in question. Mr Cavanagh also considered revenue lost as a result of the second manual spam action in November 2013, but since I have said that Oneflare should not be allowed to rely on this, that aspect of his report can be put to one side.
Mr Robertson submitted that:

there was no acceptable evidence of a linear (or any relevant) relationship between traffic and sales;

Mr Cavanagh’s evidence on this point should not be accepted as expert evidence, because it relied simply on a visual comparison of the graphic representations of traffic and sales, and that involved no application of expertise;

further, Mr Cavanagh had not employed any appropriate form of statistical analysis, such as regression analysis, to establish a statistically significant relationship between traffic and sales;

there was a further problem, which he described quaintly as an “endogeneity problem”, because Mr Cavanagh’s analysis did not allow for the beneficial influence, on traffic and sales in the later period that Mr Cavanagh chose as his benchmarking period, of the SEO strategies employed in the earlier period; and

Mr Cavanagh had calculated lost sales from 22 May 2013; but the level of sales at that date of necessity reflected (on Oneflare’s case, which Mr Robertson did not accept) the impact of the increased traffic generated by Mr Chernih’s SEO activities. To discount the effect of those activities, a much lower starting point should have been selected.

Mr Lucarelli submitted in response that:

the absence of statistical analysis was irrelevant, because the Court itself could look at the figures as depicted graphically by Mr Cavanagh, and reach the conclusion that there was a direct relationship between traffic and sales;

in any event, this was a matter of common sense;

further, it was something that had been recognised by Mr Chernih himself, in his budget email of 9 May 2012 (see at [123] to [127] above);

if there were an endogeneity problem, it could be cured by allowing an appropriate discount; and

Mr Robertson’s last point was invalid because, on Mr Cavanagh’s analysis, white hat SEO strategies would have generated traffic and sales, as at 22 May 2013, of approximately the same level as in fact had been generated by the SEO strategies actually employed.

Mr Cavanagh’s methodology

Before I deal with the first, and in some ways principal, issue as to damages, I will outline briefly the methodology employed by Mr Cavanagh. The simplest way to do that is to set out paras 5.5 to 5.10 of his report dated 7 April 2015:

5.5   The total economic loss suffered by Oneflare to 17 March 2015 as a result of the imposition of the penalties can be calculated by:
a)   Estimating the sales that would otherwise have been achieved by Oneflare during the period 22 May 2013 to 17 March 2015 had it not been for the imposition of the penalties;
b)   Subtracting the sales actually achieved by Oneflare during the period 22 May 2013 to 17 March 2015; and
c)   Subtracting the additional direct costs that Oneflare would have likely incurred in achieving sales at the level described in paragraph 5.5(a) over the direct costs actually incurred.
5.6   The economic loss suffered by Oneflare for the periods 26 July 2013 to 17 March 2015 and 21 November 2013 to 17 March 2015 can be calculated using the same methodology as noted at paragraph 5.5, but substituting the appropriate periods.
5.7   The economic loss suffered by Oneflare as a result of Google applying an algorithmic penalty on 22 May 2013 will be calculated as the difference between the economic loss calculated for the period 22 May 2013 to 17 March 2015 and the economic loss calculated for the period 26 July 2013 to 17 March 2015.
5.8   The economic loss suffered by Oneflare as a result of Google applying a manual penalty on 26 July 2013 will be calculated as the difference between the economic loss calculated for the period 26 July 2013 to 17 March 2015 and the economic loss calculated for the period 21 November 2013 to 17 March 2015.
5.9   The economic loss suffered by Oneflare as a result of Google applying a manual penalty on 21 November 2013 will be calculated directly using the methodology noted at paragraph 5.5 but substituting the period 21 November 2013 to 17 March 2015.
5.10   The sales that would otherwise have been achieved by Oneflare during a particular period had it not been for the imposition of the penalties can be calculated by:
a)   Establishing the maintainable level of sales that Oneflare was achieving immediately prior to the relevant period;
b)   Determining the rate of growth in sales that Oneflare would likely have achieved during the relevant period had it not been for the imposition of the penalties;
c)   Determining, if applicable, a maximum limit on sales that could be achieved by Oneflare with respect to the size of market and other factors;
d)   Calculating estimated sales by taking the maintainable level of sales achieved by Oneflare immediately prior to the relevant period (paragraph 5.10(a)) as a starting point and uplifting it by the determined rate of growth (paragraph 5.10(b)). The results should be limited to the maximum achievable sales (paragraph 5.10(c)).

There are three matters that need to be noted:

the first is that, for the reasons I have given, para 5.9 can be put to one side;

the second is that Mr Cavanagh’s determination of “the sales that would otherwise would have been achieved” was made by reference to the actual growth rate (in traffic and sales) after 28 February 2014 (when in Mr Cavanagh’s view the likely effects of the algorithmic demotion and the two manual penalties had dissipated); and

as one would expect, Mr Cavanagh made allowance for what he estimated were additional direct costs that Oneflare would have incurred in deriving that projected higher growth rate.

The only real criticism that Mr Robertson offered, of the comparison period (referred to at [227(4)] above) is that it did not allow for the beneficial influence, on the endogeneity problem. Mr Robertson did not criticise Mr Cavanagh’s methodology in deriving and quantifying additional direct costs of sale.

Is there a demonstrated relationship between traffic and sales?

The first question is whether there is any causal link between the hypothetical breach of duty and the loss that is alleged. In essence, Oneflare’s case is that the strategy that Mr Chernih devised and implemented, having as its centrepiece linkbuilding, was inherently likely to be detected. If detected, the likely result (as Mr Chernih recognised) would be either algorithmic demotion or the imposition of a manual penalty (or, perhaps, both).
Mr Lucarelli submitted that this is exactly what happened in the present case. He relied on the evidence of the SEO experts Messrs Weyher and Hamann, and on the evidence of the accountant, Mr Cavanagh.
Messrs Weyher and Hamann agreed, in substance, that:

Google at all material times regarded “black hat” SEO strategies, including artificial linkbuilding, as undesirable, and continually changed its search engine algorithm to improve its ability to detect such practices;

In 2012 and 2013, experienced SEO practitioners should have known of this; and

the use of such black hat strategies was inherently risky, and (if detected) would result in algorithmic demotion or the imposition of a manual penalty.

Mr Chernih in fact knew, at all material times, that Google had warned webmasters not to engage in link schemes because they were in breach of its Guidelines, and that Oneflare’s strategy was to break those Guidelines (I have given the references at [205] above).
Mr Chernih must have realised that if those practices were detected, then the consequences – algorithmic demotion, or manual spam action, or both – were almost certain to follow.
Mr Cavanagh collated, from Oneflare’s records, movements in traffic to the Oneflare website from about February 2012 to August 2014. He depicted that graphically. Mr Cavanagh also collated sales levels over that period. He depicted, in another graph, the traffic and sales data over that period.
In very broad outline (and as one might perhaps expect) sales tracked traffic: that is to say, as traffic increased, so did sales.

Mr Cavanagh’s report shows a steady increase in traffic from about July 2012 to 22 May 2013 (the date of the algorithmic demotion). There was a slight drop thereafter until 26 July 2013 (the date of imposition of the manual penalty). There was then a very sharp drop immediately after that date. However, in the period from August 2013 to 21 November 2013 (the date of imposition of the second manual penalty), traffic recovered to almost the same level as it had been at 22 May 2013. After 21 November 2013, the phenomenon of rapid drop and then recovery occurred once more. By March 2014, traffic was back at the 21 May 2013 level.
Although I have said that there appears to be a visual correlation between the level of traffic and the level of sales, it should be noted that sales did not drop as precipitately as traffic after 26 July or 21 November 2013.
Mr Cavanagh described the traffic data that he had collated and graphed as follows (para 4.7 of his report):

4.7 The key trends that are shown in the chart at paragraph 4.3 are:
A consistent increase in the level of website traffic from 1 January 2012 until the algorithmic penalty was imposed by Google on 22 May 2013;
An sharp [sic] decline in the level of website traffic immediately following the imposition of the algorithmic penalty on 22 May 2013;
A sharp decline in the level of website traffic immediately following the imposition of the manual penalty on 26 July 2013;
A sharp decline in the level of website traffic immediately following the imposition of the manual penalty on 21 November 2013;
A consistent increase in the level of website traffic from 28 February 2014 when Oneflare’s website was transferred back to and all penalties had been removed.

As to para 4.7(a), although the graph does show an increase in the level of website traffic for the period referred to, that is not the whole story. The level of increase appears to have been minimal – that is to say, traffic was “flat” – until about August 2012. It increased thereafter up until December 2012, and dipped in January 2013. Thereafter, up until 22 May 2013, it increased.
Further, the seven day average traffic line for the period from August 2012 to 22 May 2013 does not appear to increase at as steep a rate as does the equivalent line for the period from 28 February 2014 to August 2014. That is of some significance, because Mr Cavanagh used the later period as a model for what might be called the “white hat growth rate” that Oneflare should have enjoyed in the earlier, had Mr Chernih (on Oneflare’s case) performed his retainer with appropriate skill and competence.
Mr Cavanagh described the sales data that he had collated and graphed as follows (para 4.14):

4.14 The key trends that are shown in the chart at paragraph 4.10 are:
A relatively flat level of sales from 1 January 2012 until 1 December 2012;
A consistent increase in the level of sales from 1 December 2012 until 15 June 2013 with the exception of a marked peak in sales around 15 March 2013 and trough in sales around 15 April 2013;
A general decline in the level of sales between 15 June 2013 and 28 February 2014;
A consistent increase in the level of sales from 28 February 2014 until 17 March 2015.

Mr Cavanagh then dealt with what appeared to be the causation question by saying, at para 4.18, that the trendlines in his graphs showed an apparent “direct correlation between the level of turnover and the level of website traffic”. As a result, he said at para 4.19 that “the decline in traffic to Oneflare’s website resulted in a decline in the level of turnover”. He gave four reasons, the first three of which basically stated what was apparent from the graphs and the fourth of which restated his opinion as to “a direct correlation between the level of website traffic and the level of turnover”.
Those are the opinions as to causation that Mr Robertson attacked, on the basis that they involved no application of relevant expertise, and that they had not been validated by statistical analysis. Mr Lucarelli sought to explain and justify the relevant paragraphs in various ways. Since I did not understand the submissions that he put, and because in any event this seems to me to be a distraction from the fundamental point, I shall not set them out.
Mr Cavanagh accepted that he had not performed any statistical analysis. I accept Mr Robertson’s submission that a visual comparison of trendlines on a graph does not require any relevant application of expertise. However, that does not seem to me to have any particular significance.
First of all, causation is a matter of common sense, not a matter to be established only by statistical analysis. Of course, statistical analysis may help to prove causation in a particular case. But it does not follow, in the absence of statistical analysis, that causation has not been proved.
Secondly, Oneflare’s business model worked because people went to its website to look for a service provider and, in some cases at least, gave work to that service provider. The revenue that Oneflare derived was earned substantially through its website “introduction” of the two parties. As a matter of common sense, and on the balance of probabilities, it is reasonable to infer a causal relationship between traffic and revenue (or sales).
Visual observation of Mr Cavanagh’s graphs is consistent with that inference. In broad but sufficiently accurate terms, they do show sales tracking traffic.
Thirdly, that is precisely what Mr Chernih represented to the defendants would happen. Specifically, his “Campaign” box (see at [124] above) demonstrated a conversion rate between inquiries (which is another way of describing traffic to the website) and sales. And his “Budget” box (see at [125] above) suggested that expenditure on SEO, being directed at increasing traffic to the website, would produce an increasing return. It seems a little inconsistent for Mr Chernih now to argue that there is no link between traffic and sales.
I conclude, on the evidence as a whole, and on the balance of probabilities, that there is a reasonably direct link between the volume of traffic to Oneflare’s website and the sales revenue generated. The latter, I think, is in a reasonably direct sense a function of the former.
Once that link is established, the obvious conclusions from Mr Cavanagh’s evidence are that:

the algorithmic demotion caused an immediate decline in traffic;

the first manual penalty caused an immediate, and very sharp, decline in traffic;

the decline in sales following the algorithmic demotion was a direct result of that demotion; and

the decline in sales following the manual penalty was, likewise, a direct result of that manual penalty.

In each case, the drop in sales following the algorithmic demotion and the manual penalty was not as sharp as the drop in traffic. The reasons for that are unclear; and it is apparent from Mr Cavanagh’s graphs that although the drops in sales were sharp, they recovered very quickly. It may be more accurate to describe the impact on sales as a drop in the rate of growth of sales. However, I do not think that this semantic distinction is of any present relevance. I note that Mr Cavanagh’s exercise was, effectively, an attempt to capture in money terms the growth in sales revenue that, in his view, Oneflare had lost as a result of those two events.
It follows, in my view, that Mr Cavanagh’s evidence and common sense both support the conclusion, on the balance of probabilities, that each of the events in question – the algorithmic demotion and the first manual penalty – caused a loss of revenue.
Once that conclusion is reached, Mr Cavanagh’s method of calculating loss is in principle sound. He sought to establish a rate of growth in sales that, in all likelihood, Oneflare would have enjoyed from 22 May 2013, and to project that forward to 17 March 2015. By comparing the revenues that would have been earned on that basis with the revenues actually earned, and making allowances for increased direct costs and the like, Mr Cavanagh quantified the monetary value of that lost growth.

The endogeneity problem

In essence, Mr Robertson submitted, there was no reason to think that the beneficial impact (on growth in traffic) of the SEO strategies that Oneflare had implemented over the period of the defendants’ retainers was entirely lost. As Mr Robertson submitted, customers who had been directed to the website over that period as a result of those SEO strategies and who wanted to utilise Oneflare’s services again would not need to make a Google search. They could go straight to the website. And to the extent that such people recommended Oneflare to their families, friends and associates, the same would happen. Thus, Mr Robertson submitted, it was likely that the traffic figures for Mr Cavanagh’s comparison period (March to August 2014) reflected, at least in part, the SEO strategies employed prior to 22 May 2013.
I accept this aspect of Mr Robertson’s submissions. Mr Lucarelli did not directly challenge the proposition. As I have noted, he submitted that if the Court were to conclude that there was an endogeneity problem, it could be addressed by applying an appropriate discount to Mr Cavanagh’s calculated figure of loss. In my view, that is the correct approach. I shall return to the question of what is an appropriate discount.

The starting point sales figures

Mr Robertson submitted that it was wrong in principle to calculate loss based on (and moving forward from) the sales figures as at 22 May 2013. He submitted, undoubtedly correctly, that the traffic figures as at that date had been heavily influenced by the SEO strategies employed during the period of the defendants’ retainers. If, contrary to his primary submission, there was a demonstrated link between traffic and sales, it follows that sales, too, must have been boosted by those strategies.
I accept that traffic as at 22 May 2013 reflected in large part the SEO strategies employed to that time. I accept, further, that given my finding as to the direct relationship between traffic and sales, sales, too, would have been heavily affected by those strategies. However, Mr Robertson’s fundamental point is in my view invalid.
When one analyses Mr Cavanagh’s graph of sales during his comparison period (March to August 2014), the rate of growth is similar to the rate of growth in sales in the period from December 2012 to 22 May 2013. (I have chosen December 2012 as the starting point because sales were relatively flat up until them). Thus, the rate of growth in sales due to white hat strategies in a period unaffected by the algorithmic demotion or the manual penalties was similar to the rate of growth in sales in the earlier period. The obvious inference is that if those white hat strategies had been employed in the earlier period, the rate of growth in sales would have been much the same, and the starting point (in value of sales) for the calculation of loss from 22 May 2013 would have been much where it actually was.
I noted at [243] above one possible qualification to this: the rate of growth in sales in the later (comparison) period appears to be slightly higher than the rate of growth of sales in the earlier period. On that basis, use of the later period would somewhat overstate the likely starting point. If that were so, the exercise becomes a little more impressionistic.
On one view, the choice of a lower starting point would in fact inflate the loss figure. On another view, it may be that the rate of growth over the extended period would have flattened out somewhat. There may be other possible outcomes, or other possible explanations as to how these matters may have played out. Since the matter has been left at this level of uncertainty by the party whose onus it is to prove loss, I think the best way to deal with the uncertainty is to factor it into the level of discount to be applied.

Other issues relevant to loss

Mr Robertson referred to the daily sales figures that were buried in the appendices to Mr Cavanagh’s report. Based on those figures, Mr Robertson submitted that at the lowest points recorded following the algorithmic demotion and first manual penalty, daily traffic was still higher than it had been before Mr Chernih started work. Thus, he submitted, Oneflare had suffered no loss.
No doubt, Mr Robertson’s analysis of the figures is correct. Mr Lucarelli did not suggest otherwise. However, it does not address the way in which Mr Cavanagh sought to calculate the loss of revenue. As I have pointed out, Mr Cavanagh’s calculation was one that attempted to establish the revenue that would have been earned but for (as he saw it) the effect of the algorithmic demotion and the first manual penalty, and to compare that with the revenue actually earned.
Once Oneflare’s case on damages is properly understood, it matters not that the SEO strategies employed up to 22 May 2013 produced some increase in traffic. The question is, rather, what is likely to have happened but for those strategies.
Mr Cavanagh considered the raw data very carefully. He took considerable care to ensure that the sales figures were accurate, and to identify and quantify all increased direct costs. Mr Robertson did not criticise those elements of Mr Cavanagh’s work.
Mr Cavanagh also considered, and sought to factor in, the likelihood that growth would flatten out in the future. That has been addressed to some extent by the way in which Oneflare puts its claim for damages. That claim is now based on Mr Cavanagh’s analysis up until 17 May 2015, and there is no claim made for future economic loss.

Calculation of and conclusions on damages

Mr Cavanagh’s calculations included what he called “low estimates” of loss based on each of the events that he had been asked to address: the algorithmic demotion and the first and second manual penalties. In final submissions (written submissions in reply dated 18 August 2016, para 87) Mr Lucarelli submitted that the Court should accept those low estimates as the measure of loss. For present purposes, the estimates are $464,516.00 relating to the algorithmic demotion and $497,713.00 relating to the first manual penalty. In each case, those figures are calculated to 17 March 2015. There is no claim now pursued for future economic loss.
In my view, considering the thorough way in which Mr Cavanagh went about establishing those figures (in particular), the care taken to normalise sales figures, and to identify and strip out increased direct costs, those figures should be accepted as the starting point for any assessment of damages.
There remains the question of a discount. That discount should address not only the two matters specifically identified above – the endogeneity problem and the apparent difference in the growth trendlines – but also, as contingencies (or known unknowns) or other factors. One of those factors could be, I think, the impact of disruption. Specifically, as I see it, the establishment of a successful online business (assuming that Oneflare can be so described) leads others to emulate it. Where the market is finite, market share depends to some extent at least on the number of competitors in the marketplace. Accepting that Oneflare has a “first mover” advantage, and accepting that what I am saying is put at the level of theory because it was not dealt with in the evidence, nonetheless, I think, assumptions as to growth should be treated with some care.
A related consideration is the common phenomenon that, as a business grows, the rate of growth tends to flatten out. As I have said, Mr Cavanagh sought to allow for this. Nonetheless, it being something inherently incapable of precise estimation, it should be factored into the discount figure.
The third additional matter to be considered, in relation to the discount rate, is the possibility that Messrs Dong and Lim would seek to employ their own SEO strategies once again (as, in my view, they did with the domain “ojr.org”). It may be that they have learned their lesson. It may be that, having learned their lesson, they now feel more confident in using their own endeavours to supplement those of their current SEO consultant. Mr Dong’s representation of himself, in October 2013, as an SEO expert could suggest the latter possibility.
Putting all those matters together, and accepting that the exercise is necessarily impressionistic, I conclude that an appropriate discount rate to apply is 30%. That figure seems to me to be appropriate to reflect the various matters to which I have referred and, moreover, the inherent uncertainty that seems to attend so many internet-based businesses.
It follows that, were it necessary to do so, I would assess damages, as at 17 March 2015, at $673,560.00. If damages were to be awarded, interest should of course accrue on that sum from 17 March 2015.

Costs

The obvious costs outcome, given my conclusions, is that the defendants should have their costs. However, Mr Robertson sought to be heard on costs regardless of the outcome. In those circumstances, I shall defer making any costs order, and give directions for the parties’ positions on costs to be established through written submissions. My present view is that whatever costs disputes may emerge should be dealt with on the papers.

Conclusion and orders

Oneflare’s claim against the defendants fails.
I make the following orders:

direct entry of judgment for the defendants on the plaintiff’s claim.

Reserve costs.

Direct the parties to exchange written submissions on costs (not exceeding 10 pages in length), together with any affidavits or other evidence in support, by 30 September 2016, and to provide copies to my Associate at the same time.

Direct the parties to exchange written submission in reply on costs (not exceeding five pages in length), together with any affidavits or other material in support, by 14 October 2016, and to provide copies to my Associate at the same time.

Subject to any order otherwise, direct that the question of costs be dealt with thereafter on the papers.

Direct that the exhibits be handed out once all questions of costs have been resolved.

**********

Decision last updated: 13 September 2016

Medium Neutral Citation:	Oneflare Pty Ltd v Chernih [2016] NSWSC 1271
Hearing dates:	14/03/2016, 15/03/2016, 16/03/2016, 17/03/2016, 18/03/2016 and 19/08/2016
Date of orders:	13 September 2016
Decision date:	13 September 2016
Jurisdiction:	Equity - Commercial List
Before:	McDougall J
Decision:	Judgment for defendants on plaintiff’s claim. Reserve costs. Directions for submissions on costs.
Catchwords:	CONTRACT – contract for search engine optimisation services – where scope and terms of retainer not fully expressed – implied contractual and common law duty to use reasonable care and skill – whether duty extended to duty to warn – whether scope and terms informed by subsequent conduct and performance of retainer – whether plaintiff or defendants responsible for design of search engine optimisation strategy – whether defendants used reasonable care and skill in the performance of the retainer DAMAGES – pure economic loss – whether breach of retainer results in loss – where direct correlation between alleged breach and loss – assessment of expert evidence – whether discount should be allowed if damages awarded
Legislation Cited:	Civil Liability Act 2002 (NSW) Trade Practices Act 1974 (Cth) Fair Trading Act 1987 (NSW)
Cases Cited:	Brambles Holdings Ltd v Bathurst City Council (2001) 53 NSWLR 153 Commercial Union Assurance Company of Australia Ltd v Ferrcom Pty Ltd (1991) 22 NSWLR 389 County Securities Pty Ltd v Challenger Group Holdings Pty Ltd [2008] NSWCA 193 Liverpool City Council v Irwin [1977] AC 239 Prestia v Aknar (1996) 40 NSWLR 165 Wallace v Kam (2013) 250 CLR 375 Watson v Foxman (1995) 49 NSWLR 315
Category:	Principal judgment
Parties:	Oneflare Pty Ltd (Plaintiff) Nicholas Chernih (First Defendant) Linkbuild SEO Pty Ltd (Second Defendant)
Representation:	Counsel: G Lucarelli (Plaintiff) D W Robertson (Defendants) Solicitors: Levitt Robinson Solicitors (Plaintiff) Robertson Saxton Primrose Dunn (Defendants)
File Number(s):	2014/336363