Over the last year, Google has inaugurated several large scale algorithm changes that have sent many in and outside the affiliate industry into a flurry of conversation and confusion. While aiming to increase quality, these shifts have created a strange mess. Chief among these changes are:
- The presence of 3-5 sponsored posts at the top of almost all search results
- The deranking of smaller and then larger content sites
- The introduction of AI-generated results at the top of the feed
In this post, we’re setting the stage for a conversation with Declan Dunn, host of the podcast, The AI Optimist, in a couple of weeks. As always, we tailor this content to those in the affiliate space and unpack what these changes might mean for you.
The Deranking of Content Sites
One of the largest shifts Google initiated in the past year was aimed at improving search results for quality content over sites pruned and groomed for maximum SEO. Some companies have even gone out of business, like the Canadian affiliate fintech company, Hardbacon, whose CEO offered:
“They killed really bad websites, and they killed a bunch of good websites as well.” –Julien Brault
These changes initially tended to privilege large corporate retailers over smaller content sites that had previously thrived. Sites like HouseFresh.com were suddenly down 97% their normal traffic. After years of top ranking, many of the companies hurt by these changes were legitimate coupon and content publishers providing honest and useful content.
I spoke with Tricia Meyer of the Performance Marketing Association about all the changes Google has made and how they’re affecting the affiliate space. She offered that, when many complained, Google started going after larger media and affiliate sites who had been left unscathed.
Some were taking advantage of their name brand to promote affiliate content that was not actually being evaluated by people on their team. Instead, they were outsourcing to SEO companies that would market affiliate products under their respected brand name. For example, while Forbes initially escaped notice, Google seems poised to penalize them now because many complained about the omission. While still unconfirmed, it appears likely that Forbes has now been targeted by Google’s site reputation abuse policy.
As a helpful tip for merchants and managers in the coming months, Meyer urges folks to review what revenue is coming in month over month right now for each of their content publishers. She suggested that they might need to make some quick decisions about where to put that money this year.
Sponsored Posts
Sponsored posts were harmful to a lot of affiliates because, instead of genuine SEO, which allows Google to evaluate sites based on qualities like trustworthiness and user-friendliness, companies are now paying for that prime real estate at the top of the feed.
Users have to scroll down an entire page to get a non-sponsored result on Google these days. You are also likely to click on the wrong link, wasting your time and giving Google unearned revenue. For example, the site HelloPartner.com is the fourth result down when you directly search for it, under four paid posts:
Despite its policing of SEO externally, Google itself has become more of a marketing platform than a search engine.
Google AI: A Beta Jockeying as an Alpha
Google AI has changed affiliate forever because of its AI-generated search results, which started popping up around May 2024. This article from SearchEngineLand.com suggests that Google AI is hurting the click through rate for every website on the internet by 18-64%.
“Now, no one has to click on the content sites their query generates because Google AI will just give them the answer.” Tricia Meyer, PMA
One of the few things everyone agrees on concerning AI is that we are still in the technology’s infancy. However, Google seems to just be coming to terms with its AI’s inaccuracy. One reason for this is that Google AI was trained on a large amount of Reddit content.
Reddit is a useful site to humans who know how to read sarcasm, untrue answers, and sift for the right or helpful answer. It is not a site that many of us would assume as a great training ground for AI. Some industries are finding that 43% of their industry-specific responses in Google AI are factually inaccurate!
A Strange Mess
Interestingly, in the last several days Google made a change to how you respond to its AI’s accuracy. Initially, they only offered the feedback option of “report a problem.” When you finished reporting a mistake, Google shows you an image of two girls dancing up some stairs. This belies the seriousness of the liability the inaccuracy of Google AI presents.
Now, there is a thumbs up and down option, where you can provide more information. This may be a small step in the right direction because it invites engagement more actively, and thus may help us move toward the improvement of the technology.
In my conversation with Declan Dunn, we’ll be diving deeper into where we are with AI. I’ll ask for his thoughts on where we are with implementing the ethics and guardrails we need to keep us safe. Until then, I’ll hope for a future where algorithm changes at companies as mammoth as Google are done in conversation with experts in various fields who can help save good companies from going out of business.
Featured image by Brett Jordan.
