An AI assistant can now compare thirty products on specification, price and review sentiment in a few seconds, and produce a competent recommendation.
That capability has removed the value from a specific and very large category of affiliate content: the spec-comparison roundup. Best twelve of something, assembled from manufacturer pages, no first-hand use.
Which is uncomfortable, because that category was most of the affiliate internet.
But it has taken nothing from a different category, and understanding the boundary tells you exactly what to write. Here is where the line falls.
Quick Info
What They Do Well
Being clear about this makes the rest of the analysis usable.
Every one of those was previously the value proposition of a roundup article. That is why roundup traffic has fallen, and it is not coming back.
What They Cannot Do
This is the list to build content against.
They own the specification. You own the experience.
Notice that all six of these require ownership and time, which is precisely why they are defensible.
What This Means for Content
Concretely, what to stop and what to start.
Stop: roundups assembled from manufacturer pages, spec comparison tables you did not generate from testing, best of lists covering products you have not used, and anything whose value is aggregation.
Start: long-term use reports, single-product deep reviews with your own measurements, comparisons between two products you have both used properly, and the professional's view of what to buy.
And one format that has become more valuable rather than less: the specific situation. Not best laptops, but which laptop for someone doing this particular job in these particular conditions. Assistants handle constraints stated in the query; they cannot supply constraints the buyer did not know to state, and that is where an experienced person adds value.
Where the Commission Goes
A practical question people are right to worry about.
When an assistant recommends a product and links to it, the affiliate is generally not in that chain. Some assistants have commercial arrangements of their own. That share of the market is not addressable by affiliate content at all.
But the researched purchase has not disappeared. What happens in practice, and I watch this in my own behaviour, is that people use an assistant to shortlist and then search for a real review of the shortlist before spending meaningful money.
That second step is your traffic, and it is higher intent than the traffic you lost. Fewer visitors, much closer to purchase.
So the content to own is not the shortlist. It is the verification of the shortlist, which is a smaller market with a much better conversion rate.
Being the Source They Cite
The other side of this, which is the more optimistic one.
Assistants have to draw their judgement from somewhere, and for anything the specification does not capture, they draw it from reviews written by people who used the product.
Which means a review containing your own measurements, your own long-term observations and your own named assessment is the only available source for that information. It gets cited, and your name goes into the answer.
How to make that likely: real numbers from your own testing, a clear statement of how long you used it, named authorship with relevant standing, and a self-contained passage per decision factor. The mechanics are the same as in my citation article.
This is a better position than being one of forty roundups competing on the same aggregated information.
Categories Most and Least Affected
Worth knowing where you sit.
Most affected: standardised products where the specification genuinely tells you most of what you need. Consumer electronics accessories, commodity goods, anything where the buying decision is largely price and features.
Least affected: anything where use over time matters, anything with a learning curve, anything bought by professionals with opinions, services, software with workflow implications, and anything where support quality is part of the purchase.
If you are in a most-affected category and the content was aggregation, the honest advice is to change either the content or the category. If you are in a least-affected one, this shift is largely an opportunity, because your competitors' aggregation content is losing ground and yours never depended on it.
The Practical Plan
That is a smaller content library doing more work, which is the pattern across everything in search right now.
Frequently Asked Questions
They are removing the value from aggregation content, particularly spec-based roundups. Content based on first-hand use and original measurement is largely unaffected and often benefits.
Long-term use reports, original measurements, what specifications fail to capture, real-environment demonstration, and what a professional in the field would actually buy.
Generally not. That share of the market is not addressable by affiliate content, which is why the verification step after the shortlist matters more.
Standardised goods where specification and price largely decide the purchase. Least affected are products where use over time, learning curve or support quality matter.
Yes, and this is the opportunity. For anything specifications do not capture, assistants draw on reviews by people who used the product. Original measurements make you the only source.
Stop maintaining the pure aggregation pieces and rebuild the highest-value ones as first-hand reviews with your own measurements.
Before You Go
The easy half of affiliate content is gone. The half that required buying the product and using it for six months is now the whole business, and it is more defensible than it has ever been.
My testing checklist and the review template with the measurement prompts are in the resources section.
If your affiliate traffic has fallen and you want to know which half it was, get in touch.
They own the specs, you own the use, Tariq
I'm a marketing consultant, entrepreneur and content creator. I help businesses grow through practical marketing, websites, SEO, content and AI.
More About Tariq →