Amazon Seller Analytics: Free Tools and Numbers That Matter

Most sellers shopping for Amazon seller analytics are trying to answer a question their own account already answers. Sales are flat, or down, and the instinct is to buy a dashboard. The dashboard arrives, it is prettier than Seller Central, and it reports the same numbers with the same limitations, because it is pulling them from the same reports through the same API.

The two reports that actually diagnose a stalling listing are free and sitting in your account right now: the Detail Page Sales and Traffic report, which gives you sessions and conversion per child ASIN, and the advertising Search Term report, which shows exactly which queries you paid for and what they returned. Almost nobody opens them weekly.

So this post does the diagnosis first. Learn to read six numbers, work out which of three problems you have, and only then decide whether a tool is worth money. TopTut sells none of the software named here.

What Amazon seller analytics should tell you

Good Amazon seller analytics answers four questions in order: is anyone finding the listing, do they buy when they find it, what did that traffic cost, and what is left after fees and cost of goods. Anything that does not move you toward one of those four is decoration.

Amazon covers the first three itself, for free, at ASIN level. The fourth is the genuine hole. Seller Central will happily show you ordered product sales; it will not net out your landed unit cost, your inbound freight, your returns, your ad spend and your storage, and call the result profit. That is the gap real tools fill, and it is a narrower gap than the marketing suggests.

The second real hole is memory. Amazon’s reporting windows are finite, and once a period rolls out of range you cannot get it back. A spreadsheet you export into every month is worth more than most subscriptions, and costs nothing.

The metric hierarchy: six numbers that carry the weight

Six numbers do almost all the diagnostic work, and they sit in a hierarchy: traffic first, conversion second, revenue third, advertising last. Reading them out of order is how sellers end up rewriting bullet points when the problem was that nobody was seeing the listing at all.

Metric	What it counts	What it rules in or out
Sessions	Visits to your detail page, de-duplicated over a 24-hour window per visitor	Discovery. Falling sessions means a ranking, indexing or ad-delivery problem, never a copy problem.
Page views	Raw detail page loads, not de-duplicated	Little on its own. Page views far above sessions means repeat visits, often comparison shopping or a confusing variation family.
Unit session percentage	Units ordered divided by sessions	Everything about persuasion: images, price, reviews, availability, badge status. This is Amazon’s conversion rate.
Units ordered	Units, not orders	Volume. Compare against sessions, never in isolation. Multi-unit orders inflate it relative to buyer count.
Ordered product sales	Gross revenue at the price ordered	Nothing about profit. It is before referral fees, fulfilment fees, returns, ads and cost of goods.
Buy Box percentage	Share of page views where you held the featured offer	Whether your conversion problem is actually a competition problem. A low figure caps conversion no matter how good the listing is.

Unit session percentage is the single most diagnostic number a seller has. It is the only metric that holds traffic constant, so it isolates whether the listing itself is persuading people. Sessions can double because of a deal or a competitor stock-out; revenue can rise on a price increase alone. Unit session percentage moves only when the relationship between visitor and offer changes, which is exactly the relationship you control.

On the advertising side the ladder is impressions, clicks, click-through rate, spend, ACoS and TACoS. Impressions tell you whether your bids and targets are winning placements at all. Click-through rate tells you whether your main image and price are competitive in the search grid, which is a different question from whether the detail page converts. Spend and ACoS come last, and TACoS last of all, because it only makes sense once you have a few months of data.

Which of the three problems do you actually have?

An underperforming listing has one of exactly three problems, and the ratio of sessions to unit session percentage tells you which. Discovery, persuasion, or ceiling. Work out which one before you touch anything, because the fixes are unrelated and doing the wrong one costs you weeks.

Pull the Detail Page Sales and Traffic by Child Item report for the last 30 days and the 30 days before that. Compare sessions and unit session percentage across the two periods for the same ASIN. Then read the table.

Pattern in the numbers	Diagnosis	Check first	Action
Sessions down materially, unit session percentage flat or up	Discovery problem. The listing still converts; fewer people arrive.	Indexing for your main terms, ad impressions and budget caps, competitor entries, a category or search-suppression flag.	Fix indexing and ad delivery. Do not touch copy, images or price.
Sessions flat or up, unit session percentage down	Persuasion problem: listing, price, reviews or offer.	Recent price change, a new one-star review, a lost Buy Box, an image or A+ change, shipping speed on the offer.	Change one variable at a time and give each at least two weeks before judging.
Sessions flat, unit session percentage healthy and stable, units flat	Traffic ceiling. The listing is working; demand at your current reach is capped.	Share of voice on your head terms, ad budget exhaustion before end of day, keyword coverage breadth.	Buy more reach or add adjacent keywords and placements. Optimising the listing further returns very little.
Sessions up sharply, unit session percentage down sharply, units flat	Traffic quality problem, usually ad targeting.	Search Term report for irrelevant queries taking spend and clicks.	Add negatives, tighten match types, split broad campaigns.
Both down together	Account or offer level, not listing level.	Buy Box loss, stock-out or low inventory suppressing rank, an account health issue, seasonality.	Resolve the offer or account issue first; listing work is wasted until it is fixed.

What Seller Central gives you free

Seller Central includes four free reporting surfaces, and between them they cover traffic, conversion, keyword-level ad performance and, for brand-registered sellers, query-level market share. None of them costs anything beyond your existing selling account.

Business Reports

Business Reports is the sales and traffic hub, and the Detail Page Sales and Traffic by Child Item report is the one that matters. It gives sessions, page views, Buy Box percentage, units ordered, ordered product sales and unit session percentage, per ASIN, per day. The main limitation is that it is Amazon’s view of Amazon’s own funnel: it starts at the detail page. It cannot tell you what happened in search before the click, and it has no concept of your costs.

Advertising reports, especially the Search Term report

The advertising console’s report suite covers campaigns, ad groups, targeting, placements, advertised products, purchased products and search terms, documented in the Amazon Ads reporting reference. The Search Term report is the one with teeth, because it shows the actual customer queries that triggered your ads rather than the keywords you bid on. That is where wasted spend is visible and nowhere else. Its limitation is that it only ever describes paid traffic, so it tells you nothing about the organic half of your sales. If the report is mostly telling you your keyword set is wrong, that is a research job rather than an analytics job, and our guide to free Amazon keyword research tools is the right next stop.

Brand Analytics and Search Query Performance

Sellers enrolled in Amazon Brand Registry get Brand Analytics, and inside it the Search Query Performance report, which is the closest thing Amazon offers to a search console. It shows, per query, how many impressions and clicks the query generated across the market and what share of them your ASIN took, through to purchase share. That converts a vague ranking question into a measurable one. Two limitations: it is gated behind brand registration, so unbranded resellers cannot see it, and it aggregates by week and quarter rather than giving you clean daily numbers.

Free tools that fill the gaps Amazon leaves

Four gaps are worth paying attention to: profit per SKU after all costs, history beyond Amazon’s retention windows, competitor and market movement, and alerting when a listing changes without you. Some of those have credible free options, and one of them does not really.

Tool	Free tier, as published	Gap it fills	Where it stops
Seller Central Business Reports	Included with any selling plan, no separate cost	Sessions, conversion, Buy Box, units, revenue per ASIN	No costs, no profit, no pre-click data, finite history
Amazon advertising reports	Included with an advertising account	Query-level paid performance, placement and targeting detail	Paid traffic only, and a shorter data window than most sellers expect
Brand Analytics and Search Query Performance	Free, but requires Brand Registry enrolment	Query-level impression, click and purchase share against the market	Brand-registered sellers only, weekly and quarterly aggregation
Keepa browser extension	Free extension adds price and sales-rank history charts to Amazon product pages; per-account limits on tracking and data export are not published as specific numbers	Long-run price and rank history on competitor ASINs, beyond your own account’s memory	Product Finder, bulk data and API access sit behind the paid subscription
Helium 10 free plan	A free tier exists; the pricing page lists paid Platinum at 99 USD per month billed yearly or 129 USD monthly, Diamond at 279 or 359, Enterprise from 1,499. Per-tool free limits are not stated on that page	Profit dashboard, keyword tracking, listing alerts in one place	Free usage caps are low enough that ongoing weekly use pushes you to a paid tier
Sellerboard	Free trial rather than a permanent free tier; trial length and plan prices were not retrievable from the pricing page at the time of writing, so check before relying on it	Profit by SKU after fees, ads, returns and cost of goods	Becomes a monthly subscription once the trial ends
A spreadsheet plus monthly CSV exports	Free, permanently	Unlimited history, your own cost model, any comparison you want	Requires you to remember to export, every month, without fail

The pattern worth noticing: every genuinely free option is either Amazon’s own or is a history archive. Profit tools are a trial-then-subscribe category, and if a page tells you otherwise it is usually calling a trial a free plan. Fee modelling, inventory planning and reimbursement chasing belong to a different toolset, covered in our roundup of free Amazon FBA software.

ACoS versus TACoS, precisely

ACoS is ad spend divided by ad-attributed sales; TACoS is ad spend divided by total sales, organic included. The difference sounds trivial and is not, because ACoS is blind to the entire reason most sellers advertise.

ACoS  = ad spend / sales attributed to ads
TACoS = ad spend / total ordered product sales

Break-even ACoS = contribution per unit / selling price
  where contribution = price - COGS - referral fee - fulfilment fee

Advertising on Amazon does two jobs at once. It sells units directly, and the velocity from those units feeds organic rank, which sells more units that the ad report never claims credit for. ACoS only ever sees the first job. So a campaign sitting at an ACoS above break-even can still be the most profitable thing in the account, and cutting it on ACoS alone removes the velocity that was holding your organic position up. The damage shows up three or four weeks later as falling sessions, and by then the cause looks unrelated.

TACoS catches that, because organic sales sit in its denominator. Read it as a direction of travel, not a target. TACoS falling while total sales rise means advertising is buying rank that then works for free, which is the outcome you want. TACoS flat while sales rise means ads are scaling proportionally and you are renting your volume. TACoS rising while sales are flat means you are paying more for the same business and something needs to change. Anyone quoting you a universal healthy percentage is guessing, because the honest answer depends on your margin, your category and how old the listing is.

Worked example: where the money is actually going

This is an illustrative calculation with invented but stated assumptions, not a measured result from any real account. One ASIN, 30 days. Price 34.99. Sessions 4,000. Unit session percentage 8 percent, so 320 units ordered. Ordered product sales 320 x 34.99 = 11,196.80. Landed cost of goods 9.00 per unit. Referral fee assumed at 15 percent, which is the rate in many but not all categories, so 5.25. Fulfilment fee assumed at 5.50. Ad spend 2,100, of which ad-attributed sales are 4,800.

• Contribution per unit: 34.99 – 9.00 – 5.25 – 5.50 = 15.24
• Total contribution before ads: 320 x 15.24 = 4,876.80
• After ad spend: 4,876.80 – 2,100 = 2,776.80
• ACoS: 2,100 / 4,800 = 43.8 percent. TACoS: 2,100 / 11,196.80 = 18.8 percent
• Break-even ACoS: 15.24 / 34.99 = 43.6 percent

Read that carefully. ACoS of 43.8 percent against a break-even of 43.6 percent means the advertised units are making almost nothing. Ad-attributed sales of 4,800 at 34.99 is roughly 137 units, so about 183 units came organically and those units carry the entire 2,776.80. The ads are not a profit centre; they are a rank subsidy costing about 12 in net contribution for the month.

Now the lever. Moving unit session percentage from 8 to 9 on the same 4,000 sessions produces 360 units instead of 320: forty extra units at 15.24, roughly 610 more contribution, with no additional spend. That is why conversion sits above advertising in the hierarchy. A point of conversion is cheaper than a point of ACoS every time, and if you cannot find that point, the problem may be the product rather than the listing, which is a product research question.

A reporting cadence that survives contact with reality

Look at operational risk daily, performance weekly, and trends monthly. The common failure is inverting that: checking sales every morning and reviewing the Search Term report never.

Daily is for things that break: inventory about to run out, an account health notification, a suppressed or stranded listing, a lost Buy Box, a pricing error. Those are binary and urgent. Sales are neither.

The minimum viable weekly review

1. Pull Detail Page Sales and Traffic by Child Item for the last 7 days and the 7 before it. Sessions and unit session percentage only.
2. Flag any ASIN where either number moved more than you would expect from normal variation, and classify it against the decision table above.
3. Open the Search Term report for the same period. Sort by spend descending.
4. Negate any term with meaningful spend and zero orders across a run long enough to be more than bad luck, and note any converting term you are not bidding on directly.
5. Check Buy Box percentage per ASIN. Anything materially below 100 is a competition problem, not a listing problem.
6. Record total ad spend and total ordered product sales in your own sheet, so TACoS builds into a line you can read over months.
7. Make at most one change per ASIN, and write down the date you made it.

Monthly is for the slower questions: the trend line on TACoS, Search Query Performance if you have it, fee or price changes that reset your break-even, and the archive export that protects you from Amazon’s retention limits.

Attribution, retention and why the numbers never reconcile

Your advertising reports and your Business Reports will not tie out, and no tool will make them. They are built on different rules, and understanding the differences is more useful than chasing them.

Advertising sales are attributed on a lookback window: a click today can be credited with a purchase made days later, and that purchase is credited back to the click date, not the purchase date. So an ad report you read on Monday will keep changing for days afterward as late conversions land. Business Reports, by contrast, record orders on the day they were placed. Two different clocks, one of which runs backwards.

Add the ordinary reporting lag, where the most recent day or two is provisional, plus returns and cancellations that reduce Business Report figures later but do not retroactively remove an ad-attributed sale, and exact reconciliation stops being achievable. It is not a bug you can configure away, and any dashboard promising a single reconciled truth is picking one source and hiding the other.

Retention is the harder constraint. Both surfaces hold detailed data for a limited window rather than forever, and the advertising side is generally shorter than the sales side. When a period ages out it is gone, which means your year-over-year comparison for next spring depends on whether you exported last spring. Set a monthly calendar reminder to dump the CSVs somewhere permanent. This is also why off-Amazon numbers matter to sellers with their own storefront, and why our comparison of website traffic estimators is worth a look if you are trying to see the whole picture.

Five mistakes that cost real money

Every expensive analytics mistake is a reading error rather than a tooling error. These five account for most of them.

• Judging a listing on a week of data. A week of a low-volume ASIN is a handful of events. You will confidently identify a trend that does not exist, act on it, and destroy your ability to measure the previous change.
• Optimising ACoS toward zero. The lowest-ACoS account is one with no advertising and shrinking organic rank. ACoS is a constraint, not a goal. The goal is total contribution after ad spend.
• Never opening the Search Term report. This is where money leaks quietly: a broad match term with steady spend and no orders can run for months without ever appearing in a campaign-level view, because the campaign average looks acceptable.
• Treating ordered product sales as profit. Referral fees, fulfilment fees, storage, returns, ad spend and cost of goods all sit between that number and your bank account. Sellers scale a SKU that is losing money per unit precisely because the revenue line looks healthy.
• Comparing this year to last without noting what changed. A price rise, a main image swap, a variation added to the parent or a category fee change all break the comparison. If you did not log the change, the comparison is not evidence.

Verdict: what to actually use, by seller size

Most sellers should run Seller Central plus a spreadsheet and nothing else until a specific question forces an upgrade. Here is where I would draw the lines.

A handful of SKUs, low volume. Business Reports, the Search Term report, and a monthly CSV archive. Add the free Keepa extension for competitor price and rank history. Do not buy anything. At this volume a paid dashboard is a monthly fee for a nicer view of numbers you can already see, and the weekly review above takes twenty minutes.

Growing, ten to a few dozen SKUs, meaningful ad spend. The profit-by-SKU gap becomes real, because at this point you genuinely cannot hold each SKU’s contribution in your head. Trial one profit tool, keep Business Reports as the source of truth for sessions and conversion, and pick the tool on how well it models your actual costs rather than on dashboard design.

Brand-registered with a catalogue. Search Query Performance is the highest-value thing you have and most sellers in this group under-use it. Build the weekly review around it and the Search Term report together, since one shows what the market searched and the other what you paid for. If a meaningful share of your revenue is also moving through your own site, keep that measurement separate rather than trying to merge it, and treat your own storefront as a second, independently measured channel.

The one thing I would avoid at every size: paying for a tool because sales are down and you want something to look at. The report that will tell you why is free, it is called Detail Page Sales and Traffic, and it is two clicks from your Seller Central homepage.

Frequently asked questions