Amazon Conversion Rate Too Low: Nine Things To Check Before You Touch PPC

A supplements seller we work with had one SKU quietly converting below everything else in the catalogue. It did not arrive as a crisis. It showed up in a weekly report, as a line where the ad spend kept going out and the units did not follow, an Amazon conversion rate too low to earn back what the clicks cost.

Pausing that spend saved more than $800 that was buying nothing, and once the listing itself was improved, returns on that product came down 12%.

What makes that case worth telling is how ordinary it is. The instinct when sales fall is to add budget, because budget is the one lever that responds immediately.

It is also the most expensive instinct in this business, because advertising amplifies a page and cannot repair one. If the page converts at 4% and the category converts at 12%, more traffic buys more of the same 4%, at whatever the click costs that week.

So this is the list our team works through before touching a campaign, in the order it costs to fix. Nine checks, cheapest first, and most accounts find their answer in the first four.

The metric that actually tells you something

Amazon does not label a column "conversion rate". The number that does that job sits in Business Reports and is called unit session percentage, and it is simply units ordered divided by sessions for the period, expressed as a percentage. The path is Reports, then Business Reports, then Detail Page Sales and Traffic by Child Item.

Two details in that definition matter more than they look.

It counts sessions, not page views, so a shopper who visits your page four times in a day is one session. It also counts units, not orders, so a two-pack sale reads differently from a single. That is why unit session percentage and any conversion figure from an advertising dashboard rarely agree, and why comparing them against each other produces an argument rather than an answer.

The other reason it matters is child level reporting. A parent listing averages its children, so one strong variation can carry four weak ones for months while the ad budget quietly funds the weak ones. Child level is where the truth lives.

How to tell which one you have

1. Put sessions and unit session percentage side by side for 90 days, per child ASIN. Those two columns answer the question on their own, and almost nothing else needs to be opened first.
2. Check whether the drop has a date or a slope. A clean step down on a specific day is usually an event: a price change, a stock-out, losing the Featured Offer, an image swap. A slow slope is usually competitive drift.
3. Look at whether it is one child or the whole family. One child pointing down while its siblings hold steady rules out most account-wide explanations immediately.
4. Compare against the same weeks last year rather than last month. Plenty of categories have a seasonal shape, and a January comparison against November flatters or damns a product for no reason.
5. Open your own listing on a phone, logged out. Not on a desktop with three tabs of Seller Central open. The logged-out mobile page is what most of your buyers are looking at, and it is regularly not what the seller thinks it is.

What to check before you change anything

There is a strong urge to fix all nine things this afternoon. It is worth resisting, because a page with five simultaneous changes teaches you nothing about which one worked, and Amazon needs time to settle after each edit anyway.

1. Write down the current numbers. Sessions, unit session percentage, units, price and the review count and rating, per child ASIN, with today's date next to them. Ten minutes now saves an argument in six weeks.
2. Rule out the boring explanations. A stock-out inside the period, a coupon that expired, a competitor's deal week, a price change nobody logged. These explain more conversion drops than anything on the nine item list.
3. Check whether you can test rather than guess. Amazon's Manage Your Experiments covers titles, main images and A+ content for brand registered sellers on ASINs with enough traffic to reach a result.
4. Decide the order and stick to it. Cheapest and most reversible first, which is the order the next section is written in.

The nine checks, in order of what they cost

1. Price and Featured Offer status. This is first because it is the cheapest to check and the most brutal when it is wrong. If the Featured Offer is not yours, most of your traffic lands on a page where the buy button belongs to someone else, and conversion collapses without a single word of the listing changing. Amazon says featured offers are commonly priced at or below the lowest alternatives, that offers with fast free shipping and higher certainty delivery dates are more likely to be featured, and that an out of stock offer cannot be featured at all. Sources: Amazon and Amazon, both Amazon primary. Our Buy Box loss page covers the diagnosis when it is not obviously price.
2. The main image. Second because it decides whether the click happens at all, and because it is one image file. Amazon's product image guide requires a pure white background on the main image, the product filling the frame, and no text, logos or borders over it. Source: Amazon, Amazon primary. Beyond compliance, the question is whether the product is identifiable at thumbnail size on a phone, because that is the size at which the decision is made. Our main image CTR page goes into the testing side.
3. Title clarity. Third because it is free to change and because a confusing title costs you the shopper who was already interested. The character limit is 200 and Amazon's own recommendation is 80 or fewer, which is a useful discipline. Source: Amazon, Amazon primary. What tends to work is brand, then what the product is, then the variant and the quantity, in the order a buyer checks them. A title that leads with eleven keyword phrases makes the shopper do the sorting, and some of them will not.
4. Review count and rating. Fourth because it is real and it is slow, which is an uncomfortable combination. A page sitting under a handful of reviews next to competitors with hundreds is not converting badly because of its copy. Amazon's Vine programme exists for products with fewer than 30 reviews and is the legitimate route. Nothing outside Amazon's own programmes is worth the risk to the listing, and this is one place where a slow honest answer beats a fast dangerous one.
5. A+ content. Fifth because it takes design time. Amazon puts the average sales increase from adding A+ content at 5.6%, which is a modest, honest number rather than a transformation, and it is worth reading it that way. Source: Amazon, Amazon primary. Modules that answer objections, show scale, and compare variants do more than modules that restate the brand story, because the shopper who scrolls that far is looking for a reason to be reassured.
6. The delivery promise. Sixth because it is partly operational. The date the shopper sees is part of the offer, and a page promising delivery in eight days converts differently from the same page promising two, especially against a Prime competitor. Amazon's own description of the Featured Offer makes the same point from the other direction, with fast free shipping and higher certainty delivery dates listed among the things that help an offer get featured. This is where fulfilment choices, handling time settings and stock levels stop being an operations topic and become a conversion topic.
7. Variation structure. Seventh because it is a catalogue job. A family that pools reviews and traffic correctly gives every child a fair start, and a broken or wrongly themed family leaves children stranded with no proof and no traffic. It also distorts every number above: a parent that looks fine can contain one child that has been converting at half the rate of its siblings for months.
8. Mobile presentation. Eighth because it is a rework of what already exists rather than something new. The app reorders the page, truncates the title, shows fewer bullets before a tap, and puts A+ content in a different position than the desktop layout does. A page built and approved on a large monitor can read completely differently on the surface where most of the decision happens.
9. Attribute completeness. Ninth in cost order, and worth doing early anyway because it is mechanical. Blank attribute fields keep a product out of the filters buyers use to narrow a category, so the sessions never arrive to convert. Amazon's own generative listing tool is built to populate more than 70% of required attributes, which tells you how commonly those fields sit empty. Source: Amazon, Amazon primary.

Notice what is not on that list. Nothing here is a bid, a match type or a placement modifier. Those are worth attention once the page converts, because at that point spend compounds instead of leaking.

Two of the nine account for most of what we find. The Featured Offer, because it fails silently and takes the buy button with it, and reviews, because a page with almost none is being asked to win an argument it has no evidence for.

Both are unglamorous, and neither is what a seller hopes the answer will be, since the hoped-for answer is usually a copy rewrite that can be finished in an afternoon.

Matching the symptom to the check

What happens when the page is actually fixed

Our team rebuilt 12 SKUs for a haircare brand across the US and UK, with the item builds done properly and the copy built around use cases, material and quantity rather than a keyword list. Those listings indexed for more than 25 keywords within two weeks, conversion moved from 8% to 17%, and two variants picked up the Amazon's Choice badge.

The reason that matters in a post about ad spend is the arithmetic underneath it. Doubling conversion halves the effective cost of every click that was already being paid for. No bid adjustment does that, because bids change what you pay to arrive and conversion changes what arriving is worth. If the whole catalogue needs this rather than one page, our listing optimization page explains how we work through it.

Keeping conversion from drifting again

Most conversion problems are not events. They are drift, noticed late, which is why the supplements case earlier was caught by a report rather than by a person feeling that something was wrong.

1. Run one weekly report with sessions, unit session percentage and units per child ASIN. Not a dashboard nobody opens. One page, once a week, with last week beside it.
2. Set a floor per ASIN and treat crossing it as a task. The number matters less than having one, because a threshold turns a vague feeling into something that gets assigned.
3. Photograph the page quarterly. A screenshot of the live listing, dated. When conversion moves six months later, that is how anyone reconstructs what the page looked like before.
4. Check the Featured Offer weekly on your top ASINs. It moves silently, and it is the single fastest way to lose conversion without anything on the page changing.
5. Review returns reasons alongside conversion. Rising returns and falling conversion often share one cause, usually an expectation the page sets that the product does not meet.

What we would do first if this were our account

Pull sessions and unit session percentage per child ASIN for 90 days and find the three worst performers by revenue at risk rather than by percentage, because a 2% drop on your hero ASIN outweighs a 9% drop on something that sells twice a month.

Then check the Featured Offer and the price on those three, since that is fifteen minutes of work and it explains a surprising share of sudden drops.

Then take one element on the worst page, change it, and leave the rest alone long enough to read the result. The main image is usually the one worth taking first, because it is a single file and it decides whether the click happens at all, which is where our main image and CTR optimization work tends to start.

We cannot promise a conversion number, because conversion depends on your price, your category, your competition and your reviews, and anyone quoting you a target percentage before looking at the account is selling something.

What we can tell you, before you spend anything, is which of the nine is costing you the most and whether your ad spend is currently amplifying a page that is ready for it. That is what the free, no-obligation audit covers, and if the honest answer is that your pages are fine and it really is a traffic problem, we will say so.

Related guides

• The three-margin pricing framework
• Optimising listings for Rufus and COSMO
• Ads getting clicks but no sales

Common questions about low Amazon conversion rates