The SEO Heist – A Cautionary Tale About AI

You would be forgiven for not knowing the name Jake Ward or the SEO Heist that made Jake infamous in the SEO industry. 

However the lessons learned at the end of this tale of digital deception are important for anyone who has a website and wants it to be found in search results, which, let’s be honest, is pretty much everyone.

The founder of an SEO company, Ward managed to divert 3.6 million views in 18 months from a competitor to his target website. It received 489,509 sessions in October alone, doubling the amount of just six months previous. 

How did he do it? Clever writing? Insightful tactics and marketing? Well no. He pulled it off by using an AI to game the system.

Unless you’ve been off grid for the last year, you will have seen the explosion in the use of AI tools and how popular ChatGPT has become since its launch in November 2022. The use of AI has already changed the marketing industry and since the launch of ChatGPT there has been a lot of debate around the use of AI-written content and Google’s stance on the use of it. 

Ward figured out he could use an AI bot to generate content quickly and effortlessly. His next move was to export the competitor’s sitemap, and turn the list of URLs into article titles. From there it was quick and easy to churn out 1,800 articles from those titles using AI, which in turn dragged Google’s attention from the original site, to his.

Now here’s the science bit…

In the beginning Google’s guidelines could be interpreted that AI content = spam. Google’s definition of spam is “content that is generated for the primary purpose of manipulating search rankings”. So it could be argued that if you’re using AI to write content aimed at helping your website rank better, then you are spamming. 

At the same time, if you are not doing this at scale, is it really to manipulate the search results? This left a grey area on Google’s position on AI content which led them to give new guidance in February 2023.

In their blog post guidance about AI generated content, they reiterate that Google’s ranking system aims to reward original, high quality content that demonstrates the qualities of what they call EEAT: expertise, experience, authoritativeness and trustworthiness. Their focus is on the quality of the content and not how the content is produced. But can AI really produce content that falls into EEAT? 

Take ChatGPT. When it first launched in November 2022 it had crawled information up to September 2021. This has now been updated to January 2022 (at the time of writing this). It means that if you asked something simple, such as ‘Who is the Prime Minister of the UK?’, it would tell you Boris Johnson is still in charge, when we have had two others since then (correct at time of publication). With this in mind, can ChatGPT really produce content that shows expertise, experience, authoritativeness and trustworthiness? 

The other part of Google’s guidelines to consider is original content. How original will your content be when the output is actually predictive text that could be supplied in the same form to an unlimited number of people? And let’s not forget your brand, does AI content fit within your tone of voice or will you sound the same as everyone else?

Beware the wrath of Google

Clearly from the data provided, the SEO heist worked to the extent that it generated a lot of traffic. No doubt this increased traffic would have resulted in more business for the company whose website was suddenly receiving double their monthly traffic.

However, dubious SEO practices like this (dubbed Black Hat) is very much a cat and mouse game. People find ways to game the system until Google finds ways to combat it. 

Google released their helpful content update in August 2022. This update rewards content created for humans and demotes content that seems to have been primarily created for ranking well in search engines rather than to help or inform people. 

Similar to the way a good movie villain tells the hero all the details of their evil masterplan before being foiled twenty minutes later, Ward couldn’t resist boasting about his achievements, and shortly after he wrote a post on X bragging about what he’d done Google sat up and took notice. His site’s organic traffic has now decreased by 42%

So, the big question is, was it worth it? Ward claims it was, because for that period of time, the extra traffic bought ‘conversions’. 

But we have no context as to what these conversions are. And what about the long term damage to the brand? How many people clicked on content that wasn’t actually helpful and came away with a negative opinion of the company? If the traffic levels have returned to the same pre-Heist level, were the extra ‘conversions’ over a small period of time worth the investment and the long term damage?

Black Hat SEO tactics nearly always work, for a period of time. They often work quickly too but Google soon catches up, and breaking Google’s guidelines can result in severe penalties for your website,

Risking long term damage, both to your web presence and to your brand reputation, just for a quick hit rarely pays off. 

It’s far better to use AI as a tool, to help you deliver a dedicated long term SEO strategy. It’s useful for rough idea generation and inspiration, but it will never replace the human brain’s ability to embody your brand’s values and tone of voice. And if you need help with creating that – you know where we are.