With Amazon advertising, poor efficiency and a listings problem look identical from the outside – here’s how to tell them apart…
When the efficiency of your Amazon advertising starts to slide – TACOS creeping up, ROAS softening, cost per click rising – the instinct is to look at the media plan, changing bids, adjusting targeting and reviewing whole campaign structures.
And it’s understandable. Ads are visible and adjustable.
But, as our expert panel discussed in our latest Beyond the Listings session, all too frequently the problem lies entirely elsewhere.
What Amazon advertising can and can’t do
As explained directly in the session: an advertising problem is a failure to drive impressions or clicks.
And if that’s what’s happening then yes, your media plan deserves scrutiny.
But if clicks are coming through and conversion isn’t following, that’s a different problem; and one which no amount of bid adjustment will solve.
Emma Pickard, our Head of Client Services, framed it simply: you can drive as much traffic as you want to a product, but if the listing isn’t doing its job – if it’s not showing or telling the customer what they need to see – the traffic converts to nothing.
The media investment is wasted not because the media is wrong, but because the destination wasn’t ready for it.
So the practical question to ask before touching any budget is: is this an advertising problem, or a listing and conversion problem?
Because the fixes are completely different, and confusing one for the other is how brands end up spending more to get less.
The retail readiness question (aka Buy Box ownership)
Mike Godsiff, our Head of Advertising and Media, used the example of Buy Box ownership, one of the most common (and most commonly overlooked) retail readiness issues.
You can’t effectively advertise a product that doesn’t hold the Buy Box.
If a meaningful portion of your catalogue is losing the Buy Box, and it doesn’t matter if that’s to other sellers, to pricing misalignment or to operational issues, the advertising efficiency problem is a symptom, not the cause.
If you fix the Buy Box position first, suddenly a portion of the catalogue that wasn’t advertising-ready becomes available to drive orders – and then your media budget works harder because it’s being pointed at products that can actually convert.
It’s the same story for content.
A product with weak imagery, unclear differentiation, an insufficient review profile or a price point that the listing can’t justify will underperform in organic and paid search alike, and sending more paid traffic to that listing just accelerates the loss.
Emma’s recommendation? A quick review of your own listings versus competitors at this point in the year; specifically looking at whether positive signals in your reviews are being reflected in your content, and whether there are recurring negatives that the listing isn’t addressing.
Watch the session on demand here
Amazon Advertising costs are rising regardless
The broader cost context is worth understanding, because it means the pressure to have listings working at full efficiency is higher than it’s ever been.
Mike cited the trend in average cost-per-click on Amazon – up meaningfully over the past five years, with increased competition across most categories continuing to push bids in one direction.
In that environment, a listing that converts at 10% and a listing that converts at 7% aren’t just slightly different; the profitability gap between them, at scale, is significant.
Amazon’s own push is toward DSP as a higher-funnel offset to rising search-based CPCs, and the logic is sound.
DSP allows you to bid on an audience rather than a keyword, which means reaching a shopper who is showing all the signals of intent without competing at the most expensive end of the paid search auction.
For brands with the scale and catalogue readiness to support it, DSP offers a meaningful way to improve blended efficiency without just accepting higher costs because competition is fierce.
But DSP doesn’t fix a listing problem either. If the product page isn’t converting organic traffic efficiently, DSP-driven traffic will convert just as poorly.
The AI layer that’s already changing discoverability
And there is – of course – another dimension that makes listing quality more commercially important now than it was even twelve months ago.
Andy’s observation in the session was that everything Venture Forge is seeing around AI-driven discoverability on Amazon points to product attributes becoming the critical input – not keywords alone, and not copy.
It’s now in the depth and completeness of the structured data sitting behind a listing, and these are the attributes that many brands have historically left partially filled because Amazon’s systems didn’t visibly penalise the gaps.
Now, those gaps matter.
As Rufus and Amazon’s AI systems determine more and more which products get surfaced for exploratory, conversational and gifting-led searches, AI is reading attribute data to understand what a product is and who it’s for.
And a listing with rich, complete attribute data will be understood and matched correctly, but one with missing or generic attribute information?
That listing might not be surfaced at all, regardless of how strong the keyword optimisation is.
And it’s also worth remembering, as Emma flagged, that there’s a real risk in relying on Amazon’s own AI to fill those gaps, with plenty of cases where Amazon’s systems have rewritten product titles based on attributes it’s inferred, but produced descriptions that bore very little relation to the real product.
That makes having human eyes on how your listing data is structured is crucial – a true commercial protection measure.
Our take
If you want to get your advertising efficiency right in the second half of this year, start with the listing, not the bid.
That means retail readiness – Buy Box health, content quality, review profile, attribute completeness – coming before media investment, not alongside it or after it.
And the commercial logic is straightforward: every pound of advertising spend is working against a conversion rate.
So if that conversion rate is low because the listing isn’t ready, increasing the spend doesn’t improve the rate – you need to improve the listing.
This is an environment where CPCs are rising and AI discoverability is increasingly driven by structural data.
Now, the quality of what’s behind the ad matters more than the ad itself.
The Bottom Line
Advertising efficiency is so rarely a pure media problem – think of it instead as a signal that something upstream needs attention.
For most established brands, that something is a combination of retail readiness, content quality and structural data completeness that the advertising is making visible rather than causing, so fix those first and then scale the media.
For the complete conversation, including how to distinguish between genuine advertising inefficiency and listing-level problems, Mike’s take on DSP strategy through peak and what AI-driven discoverability means for how you structure your listing data, watch the session on demand today.
And if you’d like a practical tool to help you pinpoint exactly where you stand now – while there’s still enough time to fix things before peak – our new Amazon Foundations Scorecard will give you the lowdown on what’s working, what isn’t and where to focus your attention before Q4 arrives.
…..