Most Amazon shopping happens on a phone, and almost every Amazon mobile listing was written in a spreadsheet and approved on a 27 inch monitor. That gap does not show in your advertising reports. It shows in conversion rate. Seven things to check this afternoon.
Listings and Creative
Your Amazon mobile listing is the one that matters, because the large majority of Amazon shopping happens on a phone. Almost every listing is written in a spreadsheet, approved on a 27 inch monitor, and never opened on the device it will actually be read on. That gap does not show up in your advertising reports. It shows up in unit session percentage, sitting in your business report where nobody looks.
This is the first thing we check on any account we take over, before campaigns, before bids, before structure. Not because listing work is glamorous, but because it is the multiplier underneath every other number. Fix the page and every pound already in the account starts working harder.
What is on this page
Why the listing comes before the campaign
Every click you buy lands on the same page. If that page converts at 8% instead of 16%, you are paying exactly twice as much for every sale, and no bid adjustment anywhere in the account will fix it.
That is the whole argument. Bidding decides what a click costs. Conversion decides what a click is worth. One of those is a lever you pull and the other is a multiplier that sits underneath every number in the account, including your ACOS, your total advertising cost of sale and your organic rank.
The ranking part matters as much as the maths. Amazon promotes listings that turn visits into sales, so a page that converts badly gets less organic traffic, which means more of your sales have to be bought, which pushes costs up again. The mechanism is set out in how Amazon organic rank actually works, and conversion rate sits close to the middle of it.
Seven things to check on your own phone
Open your own listing on your phone, held at normal arm’s length, in the Amazon app rather than a desktop browser at a narrow width. Then work through these.
One. Your title after eighty characters. That is roughly where mobile truncates. Everything that has to land has to land before that point: brand, what the thing is, and the one qualifier that separates you from the listing beside you.
Two. Whether the bullets are visible at all. On many mobile layouts they sit below the fold and behind a tap. If your only real differentiator is in bullet four, most shoppers never reach it.
Three. Text inside your images. Fourteen pixel copy in a desktop mock up is unreadable on a phone. If you cannot read it at arm’s length, neither can a buyer, and there is no hover to zoom.
Four. Your comparison table. On a phone a three column table becomes a horizontal scroll almost nobody performs. Stacked comparisons convert. Scrolling ones do not.
Five. The main image at thumbnail size. It has to work at roughly the size of a postage stamp in search results, not at full width on a detail page. Shrink it on screen and see whether you can still tell what the product is.
Six. Whether A+ repeats the bullets. If it does, it is decoration you paid for. It should be answering the objections the bullets could not carry, which we went through in why your A+ content is not converting.
Seven. Brand Story. It appears above A+ on every product you own and it is the highest traffic cross sell space you control. Most brands fill it with a founder paragraph rather than a shop window for the range.
What an Amazon mobile listing actually shows
The test that settles it. Hand your phone to somebody who has never seen the product, open the listing, and give them fifteen seconds. Then ask what it is, what size it is, and why it costs what it costs. If they cannot answer all three, the listing is not doing its job, and every click you buy is paying for that gap.
Want us to check your listings before you spend another pound on ads?
We audit the page as well as the campaigns, and tell you plainly which one is costing you money.
What good looks like in the numbers
Unit session percentage is Amazon’s conversion rate: orders divided by sessions, per product, in the business report. It is the cleanest single measure of whether a listing works, because it strips out how much traffic you bought.
There is no universal target, since a £15 consumable and a £400 considered purchase behave completely differently. What matters is the comparison against yourself over time and against the products ranking around you. Two readings are worth doing every month.
| What you see | What it usually means | Where to look first |
|---|---|---|
| Sessions up, orders flat | Traffic is arriving but the page is not closing | Main image, price, reviews |
| Conversion fell, nothing changed | Something outside the listing moved | Competitor price, featured offer, ratings |
| Mobile far below desktop | The page was built for the wrong screen | Title truncation, image text, tables |
| Good conversion, poor rank | Not enough traffic to prove itself yet | Advertising and keyword coverage |
The third row is the one this article exists for, and it is easy to spot: your business report lets you compare mobile and browser sessions. If mobile converts noticeably worse, the design is the reason, not the audience.
A real account, thirteen months
One account in our book went from roughly $23,000 a month to past $140,000 across thirteen months. Advertising went in properly in September, and it worked because the listings could take the traffic, which had been dealt with first.
That order is the whole point. The same advertising, on the same budget, spent against the earlier version of those listings would have produced a fraction of the result and a much worse ACOS, and the obvious conclusion would have been that the ads were not working. They would have been working perfectly, against a page that could not close.
It is also why we are cautious when somebody asks us to cut their ACOS in the first month. Sometimes the honest answer is that the campaigns are fine and the listing is the problem, which is a less satisfying answer to deliver but a far cheaper one to act on. The other side of that argument, where the campaigns genuinely are the problem, is in cutting ACOS without losing rank.
Every click you buy lands on this page
We rebuild listings around what actually converts on a phone, not what looks good in a mock up.
Fixing them in the right order
Not everything is worth doing at once. This order gets the most conversion back for the least work.
First, the main image. It decides your click through rate in search and your first impression on the page. It is also usually the cheapest thing to improve, because it is one image rather than a whole suite.
Second, the first eighty characters of the title. Rearranging what you already have costs nothing and is often the single biggest available gain.
Third, bullet one. Assume it is the only bullet anyone reads and make it carry the main reason to buy rather than a specification.
Fourth, image text sizes. Anything a buyer needs to read has to be legible at arm’s length on a phone. This usually means fewer words, larger.
Fifth, the comparison module. Stack it rather than scroll it, and keep it to three columns at most.
Sixth, Brand Story. Turn it into a shop window for the range and it starts earning its place on every listing you own.
Notice that four of those six cost nothing but attention. The expensive work, reshooting photography, comes last and often turns out to be unnecessary once the free fixes have landed.
Bids change what a click costs. The listing changes what a click is worth. Only one of those improves the money you already spent.
How to test a change without fooling yourself
Change one thing. Leave everything else alone for a fortnight. Compare unit session percentage on that product before and after. That is it, and the discipline is harder than it sounds because the temptation to fix six things in one afternoon is strong.
Brand registered sellers can use Manage Your Experiments on eligible listings, which runs a proper split test and removes most of the guesswork. Where that is not available, the before and after comparison works provided you protect it from three things: a price change in the same window, a stock or featured offer interruption, and a big shift in advertising mix. Any of those will move conversion more than your edit did, and you will draw the wrong conclusion with complete confidence.
Keep a simple log of what you changed and when. Listing edits are the most common untracked change in Amazon accounts, and six months later, when conversion has drifted, that log is the only way to work out what happened. It takes thirty seconds per change and it is worth more than most reporting dashboards.
Send us the reports and we will tell you if the listing or the campaign is the problem
Free, no card, written answer in three working days, and yours to keep either way.
Frequently asked questions
The large majority, and in most consumer categories it is the overwhelming majority. The exact share varies by category and price point, with higher value considered purchases skewing more towards desktop, but no category is desktop first any more. Your own business report will show you the split for your products.
Front load everything that matters into the first eighty characters or so, because that is roughly where mobile truncates. The full title still matters for search relevance, so do not cut it short. Just make sure the brand, the product and the single most important qualifier all land before the cut off.
Often not without a tap. On many mobile layouts the bullets sit below other content and behind a expand action, so a differentiator hidden in bullet four may never be read at all. Assume only the first bullet gets seen and write accordingly.
It is Amazon’s conversion rate: the share of sessions on your listing that turned into orders, shown per product in your business report. It is the number to watch when judging a listing change, because it isolates whether the page converts rather than how much traffic it received.
Check them first, then decide. The two failures worth fixing immediately are text too small to read at arm’s length and a main image that does not work at thumbnail size in search results. Both are common and both are cheap to correct compared with a full reshoot.
Yes, and heavily. Amazon promotes listings that turn visits into sales, so conversion rate feeds ranking, which feeds traffic, which feeds more sales. A listing that converts poorly is fighting the algorithm every day rather than being carried by it.
Fix the listing, almost always. A bid change alters what you pay for a click. A conversion improvement changes what every click in the account is worth, including the ones you already paid for. One is a lever, the other is a multiplier.
Change one thing, leave everything else alone for a fortnight, and compare unit session percentage before and after on the same product. Brand registered sellers can use Manage Your Experiments on eligible listings for a cleaner test. Avoid changing price in the same window, since price moves conversion more than anything else.
Send the right reports. Get a straight answer back.
We read the account properly and write back with where the money is going, what it is buying, and the three things we would change first. You keep it either way.
No obligation · no card · a person replies, usually the same working day