Search & Rank Intelligence

Understanding Search Query Performance

Every other advertising report in TrackIQ tells you what your ads did. Search Query Performance tells you what the whole market did, and how much of it you took. That difference is what makes it the most strategically useful report in the product. Watch the Search Query Performance overview Prefer to watch? Fifty-four seconds on what makes this report different. What makes it different Your Search Terms report shows the searches your ads actually ran on. If you were not bidding, it does not appear. Search Query Performance covers all search activity for your products, whether or not you were advertising. It shows the market total for each query — every impression, click, cart-add and purchase across all sellers — and your share of it. Search terms show what you ran. SQP shows what was there to be won. The gap between those two is your opportunity, and it is invisible in every other report. Your share of the market The SQP headline cards The headline cards give your share, with the raw counts underneath — your clicks of the market's clicks. Read Click Share beside Purchase Share. They are the diagnostic pair: | Pattern | What it means | |---|---| | Purchase share below click share | you win the click and lose the sale — a listing or price problem | | Purchase share above click share | you convert better than the market — you are under-bidding | | The two roughly equal | you convert at market rate | That second row is a buying signal. Converting better than the market on a query you barely appear for is the clearest case for spending more. Change figures are in pp — percentage points. A move from 1.96% to 2.12% is +0.16 pp, not +8%. The funnel, week by week Market share chart and weekly table Bars are total market search volume; lines are your four funnel shares — impression, click, cart-add and purchase. Read the lines against each other. The gaps between them are your funnel: - Impression share above click share → shown but not clicked. Image, title or price. - Click share above purchase share → clicked but not bought. The detail page. - All four moving together → a visibility change, not a conversion change. It is weekly, and that matters Amazon publishes SQP in weekly cohorts, so the date control offers a span of weeks rather than free dates. Every table shows one row per week plus a blended total. The share percentages are never added across weeks — 20% one week and 20% the next is not 40%. See Search Query Performance Is Weekly for how to read the Total row correctly. Why your data may look sparse Amazon suppresses low-volume query rows, so a brand with modest search volume will have many missing. It also lags — the most recent complete week is not available the moment the week ends. Sparse data is a reporting threshold, not a fault. See SQP Data Is Sparse or Missing. What to do with it Three things, in order of value: 1. Find demand you are not buying. Queries you convert on with no keyword against them — see Finding Queries You Convert On But Do Not Bid On. 2. Work the Recommended actions list. The screen does this analysis for you and ranks it by money at stake. 3. Watch branded share. Low share on your own brand name means someone is outbidding you on demand you created. For a full tour of the screen itself, see The Search Query Performance Dashboard, Section by Section. Related - Search Query Performance Is Weekly: Reading Week Cohorts Correctly — the cadence that trips people up. - How Much to Spend on Branded Keywords — branded market share, built from this report. - Outlier and Trending Keywords — finding terms before the category does. - Scoring Product Opportunities — search volume and growth as scoring inputs. - Keyword Research from Search Query Performance — using this report to pick launch keywords.

Search Query Performance Is Weekly: Reading Week Cohorts Correctly

Search Query Performance behaves differently from every other report in TrackIQ, and knowing why will save you from misreading it. Watch the weekly-cohorts walkthrough Prefer to watch? Fifty-nine seconds on reading week cohorts without doing the arithmetic wrong. The data arrives one week at a time Amazon publishes Search Query Performance in weekly cohorts, and TrackIQ shows those weeks as they are published. That is why the date control on this screen offers a span of weeks — Last 6 weeks — rather than the free date range you get everywhere else. The Market share table, one row per week with a blended total Every table on the screen is built the same way: one row per week, then a Total row. What the Total row is, and is not The Total row sums the raw counts — market impressions, brand impressions, clicks, carts, purchases — across the whole window, and then works out your share of those combined totals. That is a legitimate figure, but it is worth being precise about what it means: - It is your share of the window, calculated once from the summed counts. - It is not the average of the weekly shares. A week with far more market volume pulls the total towards its own share. - The share percentages themselves are never added together. A 20% share one week and a 20% share the next is not 40%. So read the Total for “how did we do overall across these weeks”, and read the weekly rows for “which way is it going”. They answer different questions. Comparing weeks is the point Because the data is cohorted, week-over-week is the natural comparison, and the dashboard is built for it. The Market share chart draws your impression, click, cart-add and purchase share as four lines across the weeks, with market volume as bars behind them. A query that appears in one week and not the next has not necessarily collapsed. It may simply have fallen below Amazon's reporting threshold for that week. “New” means new to the latest week The Keywords Tracked card counts how many queries are in the report and flags how many are new. Its own footnote is precise about this: new = first seen in the latest week. It does not mean newly created on Amazon, and it does not mean new to your catalogue. A query can appear as new, drop out, and appear as new again later. It is still a useful signal. A query that has just started producing impressions for your products is worth looking at before your competitors do. The headline cards, where every change figure is in pp Percentage points, not percent The change figures on the summary cards are in pp. A click share moving from 1.96% to 2.12% shows as +0.16 pp. That is the arithmetic difference between two percentages, not a percentage change — as a percentage change it would be over 8%. Mixing these up is the easiest way to over-report a result. Why your data may look sparse The weekly analysis table Search Query Performance is a separate ingestion from your sales and advertising data, and it depends on Amazon publishing it. - Amazon suppresses low-volume rows. A brand with modest search volume will have many missing. - It lags. The most recent complete week is not available the moment the week ends. - It is not tied to your advertising. SQP covers all search activity for your products, whether or not you were bidding. That last point is what makes the report valuable. It is the basis of the single most useful thing you can do with it: finding queries you convert on but do not bid on. In short Read the weekly rows for direction, the Total row for the window, and never add two share percentages together. Related - SQP Data Is Sparse or Missing — the most common consequence. - Spotting New Search Queries Week Over Week — what the latest cohort adds. - Date Ranges, Granularity and Comparing Against the Previous Period — how every other screen behaves. - Narrowing Fifteen Ideas to Seven — reading seasonality without misreading the cohorts.

Reading Organic and Sponsored Rank

The Rank Tracker records, for every keyword and product in a project, where that product appeared in Amazon's search results on a given day. Watch the rank walkthrough Prefer to watch? Fifty-seven seconds on reading rank correctly. The Rank Tracker headline figures Two ranks, recorded separately Organic rank is your position in the unpaid results. It reflects relevance, sales history and conversion — the things you influence slowly. Sponsored rank is your position among the sponsored placements for that keyword. It reflects your bidding, and it can change within a day. They are tracked independently, and one does not imply the other. A product can rank well organically for a keyword you are not bidding on at all, and vice versa. Lower is better Rank is a position, not a score. Rank 1 is the top of the results. A product moving from rank 12 to rank 4 has improved. This catches people out when sorting: sorting a rank column ascending puts your best positions first. A blank rank is not rank zero If a cell is empty, the product was not found within the depth that was checked that day. It does not mean the product ranked zero, and it does not mean it was delisted. The practical reading of a blank is “not visible for this keyword”. That is meaningful information — a product that used to appear and now does not has lost visibility — but it is different from a numeric drop. A product may also have a blank sponsored rank simply because you were not running ads against that keyword that day. Sponsored data is naturally sparser than organic. The rank project list Only tracked keywords are recorded The Rank Tracker records what you have asked it to record. A keyword that is not part of a rank project has no history, and adding it today does not backfill yesterday. This is worth planning: decide the keywords that actually matter for a product and add them early, because the value of rank data is in its history. Your products and your competitors' A rank project can track ASINs that are not yours. Each tracked product is marked as either your own listing or an external one added purely for comparison. That makes a project a genuine head-to-head: the same keywords, the same day, your position and theirs side by side. When you are losing ground on a keyword, this is what tells you whether it is because a specific competitor gained. Rank tells you position, not demand The Rank Tracker does not tell you how many people search a keyword. Ranking first for something nobody searches is worth nothing. For search volume, use Search Query Performance, which reports market-wide impressions per query. The two together are the useful pair: SQP tells you which keywords are worth winning, Rank Tracker tells you whether you are winning them. Related - Reading Visibility, Average Position and Top 10 — the tiles above the split. - Rank Tracker: Creating Your First Project — setting tracking up. - Rank Data Shows Blank — when one side is empty.

The Search Query Performance Dashboard, Section by Section

Search Query Performance is the densest screen in TrackIQ, and the most valuable. It shows what shoppers actually typed, how the whole market performed on those searches, and how much of it you captured. It has four sections, stacked down the page. This article walks through each one. Watch the Search Query Performance walkthrough Prefer to watch? A ninety-second tour of all four sections. A tour of the four sections of the SQP dashboard The four sections in order: Market share, SQP Analysis, SQP & Keywords, and Recommended actions. 1. The headline cards The five summary cards at the top of the SQP dashboard Five figures, all calculated across the week range you have selected. | Card | What it tells you | |---|---| | Click Share | Your share of every click in the market for your tracked queries. The small figure beneath is the raw count — your clicks of the market's clicks. | | Purchase Share | The same, for purchases. Compare it against Click Share: if purchase share is lower, you are winning the click and losing the sale. | | Keywords Tracked | How many queries are in the report, and how many are new — meaning first seen in the latest week. | | PPC on Tracked Queries | What you spent on Sponsored Products and Sponsored Brands search terms across the same window, with its ACoS. This is the money side of the report. | | Actions | How many recommendations are waiting, broken down by type. | The change figures are in pp — percentage points, not percent. A share moving from 1.96% to 2.12% is +0.16 pp, not +8%. 2. Market share The Market share chart and its weekly table The chart carries two axes. Bars are total market search volume for your tracked queries, on the left axis. Lines are your funnel shares on the right axis — impression, click, cart-add and purchase. Read the lines against each other rather than in isolation. The gap between them is your funnel: - Impression share above click share → you are being shown but not clicked. A listing image, title or price problem. - Click share above purchase share → you are being clicked but not bought. A detail page problem. - All four moving together → a visibility change, not a conversion change. Underneath, the same thing as a table, one row per week: market and brand figures side by side for impressions, clicks, carts and purchases, with your share of each. The Total row at the bottom sums the raw counts across the whole window and works out your share of that combined total. That is a legitimate blended figure — but note it is not the average of the weekly shares, and a single heavy week will pull it. 3. SQP Analysis The SQP Analysis table, ranked by SQ Score The query-level detail, ranked by SQ Score. Two tabs sit at the top: - SQP & Search Terms — queries matched against the search terms your ads actually ran on. - SQP & Keywords — queries matched against the keywords you are bidding on. This is where you find demand you are not buying. Every query gets one row per week plus a Total row, so you can see a trend without leaving the table. SQ Score ranks the queries by how much they matter to you. Funnel Score is colour-coded, and the colour is the instruction: green beats the market, so scale it. Red trails the market, so fix it. The term-type filters — Generic, Branded, Competitor — are worth using deliberately: - Generic is your acquisition pool. People describing a product, not a brand. - Branded is demand you have already earned. High share here is expected; low share means someone is outbidding you on your own name. - Competitor is queries naming another brand. Small numbers are normal. Clicking a query opens its funnel — brand against market — and the split across your ASINs. 4. Recommended actions The Recommended actions table This is the section to start with, and the one most people scroll past. It filters everything above down to the queries that need a decision, and states the decision. Five action types, each with a count: | Action | Meaning | |---|---| | Cut bids | You are paying too much for this query. | | Fix listing | You are shown but not clicked. | | Fix conversion | You are clicked but not bought. | | Scale up | You beat the market here and are under-invested. | | Add as keyword | Real demand you are not bidding on at all. | Filtering the recommendations by action type Choosing an action type filters the table to just those recommendations — here, the three queries with real demand you are not bidding on. Each row carries an Impact in money, labelled as wasted spend, recoverable sales or spend to save — so you can work in value order rather than alphabetically. The two columns that make it trustworthy are Why and Suggestion. Why states the evidence: “You're shown (3.9% impression share) but shoppers click competitors: your CTR 0.24% vs theirs 1.80%.” Suggestion states what to do about it. You do not have to agree. But you should not dismiss a row without reading the Why. Where to start Open Recommended actions, sort by Impact, and work down. Everything above it is the evidence for what it is telling you. Related - Account vs Products: Two Ways to Read SQP — the two tabs, and what changes between them. - Impression, Click, Cart-Add and Purchase Share — the four shares, read as one funnel. - SQ Score and Funnel Score — the two scores beside every query. - Recommended Actions: The Queries That Need a Decision — the queries that need a decision. - SQP & Search Terms: Matching Demand to What You Actually Ran — market demand against what you actually ran. - Spotting New Search Queries Week Over Week — demand that did not exist last week. - Search Query Performance Is Weekly: Reading Week Cohorts Correctly — why the latest week is often missing.

Finding Queries You Convert On But Do Not Bid On

There is a specific gap worth hunting in every Amazon account: search queries where your products already convert well, but which you are not bidding on at all. You are winning those sales organically. Adding a keyword lets you win more of them — and if a competitor finds the query first, you start losing the ones you had. TrackIQ has a view built for exactly this. Watch the walkthrough Prefer to watch? A minute on finding and claiming unbid demand. Where it is Analytics → Search Query Performance, then scroll to SQP Analysis and choose the SQP & Keywords tab. The SQP Analysis table with the SQP & Keywords tab The other tab, SQP & Search Terms, matches queries against the search terms your ads actually served on. SQP & Keywords matches them against the keywords you are bidding on. That difference is the whole point: a query can produce sales through a broad match or organically without ever appearing as a keyword you chose. What you are looking for A query with all four of these: 1. Meaningful search volume — enough demand to be worth the effort. 2. A green Funnel Score — you beat the market on conversion for this query. 3. Purchase share at or above click share — people who click you, buy. 4. No keyword bidding on it. The first three say the query converts for your catalogue. The fourth says you are not paying to be there. Set the term filter to Generic while you do this. Branded queries you already win, and competitor queries are a different decision entirely. The shortcut The Recommended actions section at the bottom of the same screen does this analysis for you and files the results under Add as keyword. Recommended actions, filtered by action type Each row states the evidence in the Why column and the action in Suggestion. Start there, then use the Analysis table to check anything you want to see the week-by-week trend for. Ask Aria instead If you would rather not read a table, ask: Which search queries have high purchase share but no keyword bidding on them, over the last six weeks? Aria queries the same Search Query Performance data. Give it a date range, because SQP is weekly and the answer changes by week. What to do with what you find Add the query as an exact-match keyword in a campaign that already contains the product that converts for it. Exact rather than broad, because you know the specific query works — you do not yet know its neighbours do. Start the bid modestly. You already rank organically for this query. The aim is to capture more of it, not to pay top of search for a position you partly hold. Check back in two weeks. SQP publishes weekly, so two weeks is the first honest read. Look for click share rising without purchase share falling. The reverse case, worth the same attention While you are in this view, look for the opposite: queries where you are spending but the Funnel Score is red. You are buying clicks on searches your catalogue does not convert. Those appear under Cut bids in Recommended actions. Finding demand you are missing and stopping spend you are wasting are the same exercise, done in the same table. Related - SQP & Search Terms: Matching Demand to What You Actually Ran — the sibling tab, matched against what you ran. - Harvest Winning Search Terms into Exact-Match Campaigns — turning a finding into a keyword. - Generic, Branded and Competitor Terms — the chips that filter the table. - Keyword Research from Search Query Performance — the same data, read for a relaunch.

Recommended Actions: The Queries That Need a Decision

Most reports tell you what happened. The Recommended actions section at the bottom of the Search Query Performance dashboard tells you what to do, why, and what it is worth. It is the fastest route to value in the product, and it sits below three sections of dense analysis that people rarely scroll past. Watch the Recommended Actions walkthrough Prefer to watch? A minute on working the list by impact. The Recommended actions table The five action types Every recommendation is filed under one of five types, and the filter chips along the top carry live counts. | Action | The situation | What it usually means | |---|---|---| | Cut bids | High ACoS on a query | Either it does not convert for your catalogue, or it converts fine and the clicks are simply overpriced. The Why column tells you which. | | Fix listing | Shown, but not clicked | Your impression share is healthy and your CTR trails competitors. The listing is losing the click, not the ranking. | | Fix conversion | Clicked, but not bought | Traffic is fine; the detail page is not converting. | | Scale up | You beat the market and are under-invested | The rare pleasant one. | | Add as keyword | Real demand you are not bidding on | You are winning these organically and not buying any of them. | Impact is money, not a score The Impact column is a currency figure, and its label changes with the action type: - wasted spend — what this query is currently costing you for nothing. - recoverable sales — what fixing it could reasonably return. - spend to save — spend you could stop without losing the position. Sort by Impact and work down. A £1,592 wasted spend row is worth more of your morning than a £46 one, whatever the percentage next to it says. Why and Suggestion These two columns are what make the table worth trusting. Why states the evidence in full. Not “low CTR” but “You're shown (3.9% impression share) but shoppers click competitors: your CTR 0.24% vs theirs 1.80%. Matching it would add ~269 clicks this window.” You can check every part of that. Suggestion states the action concretely, including how many keywords it affects: “Negative-match this term (exact) or pause the 25 keywords on it — after a quick check the listing really cannot serve this search.” Note the hedge in that example. The recommendation tells you what it thinks and what to verify first. Read both halves. The supporting columns Volume is market-wide search volume for the query — the size of the prize. Click % is your click share. ACoS is your advertising cost of sale on that query, or no ads if you are not bidding on it at all. no ads combined with a high-volume query and a Fix listing recommendation is a particularly clear signal: real demand, you are visible, and the listing is the thing standing in the way. How to work it 1. Filter to one action type. Mixing Fix listing and Cut bids mixes two different jobs — one is a listing task, the other is a bidding task. 2. Sort by Impact. 3. Read the Why before acting. Every row is an inference from a specific window of data. 4. Do the bidding actions first. Cutting wasted spend takes minutes and the saving starts immediately. Listing fixes take longer to make and longer to show. A caution on windows Every recommendation is calculated across the week range selected at the top of the screen, and Search Query Performance is weekly. A query with one unusual week can produce a confident-looking recommendation built on very little. Before acting on a large-impact row, click through to the query in SQP Analysis and look at its week-by-week rows. A pattern across five weeks is a decision. One spike is a coincidence. Related - The Search Query Performance Dashboard, Section by Section — the dashboard this panel sits on. - SQ Score and Funnel Score — the scores that order the queue. - Generic, Branded and Competitor Terms — why competitor queries are skipped. - Keyword Research from Search Query Performance — the comparisons behind the recommendation.

Rank Tracker: Creating Your First Project

Analytics → Rank Tracker watches where your products actually appear in Amazon search — organic rank, sponsored rank and Best Seller Rank — and records it every day. Nothing is tracked until you create a project. A project is a set of products plus the keywords you want to watch them for, and that pairing is the unit everything else in Rank Tracker reports on. Watch the walkthrough Prefer to watch? Seventy-five seconds on the form, the two limits, and why history starts today. The limits, up front Two numbers shape how you should organise this, and they are printed on the screen itself: | Limit | Value | |---|---| | Projects per account | 50 | | Keywords per project | 25 | Twenty-five keywords is a real constraint. It is not enough for "every term we care about", and it is not meant to be — it is enough for one coherent question. Projects are cheap; vague projects are not. Creating one + Add project opens the full-page form. The Create project form — name, then products from your catalog Project name is the only required field. Description is optional and worth using — in six months, "launch terms we're defending after the Q3 price change" is the difference between a project you trust and one you delete. Products The Products panel moves items from your catalog into the project. Search on the left, add to move them across, and they appear under Selected products on the right. Your products on the left, selected products on the right These are your live catalog listings, so each carries its ASIN and seller SKU — useful when you have several near-identical variants and need to be sure which one you are tracking. Track other ASINs Below the picker sits the panel that makes Rank Tracker more than a self-audit: Track other ASINs, keywords one per line, and the tracking toggle Track a competitor's or vendor's ASIN alongside your own products — no catalog listing required. One ASIN per line, optional. Anything you paste here is tracked on the same keywords, on the same days, and appears as another column everywhere the project reports. See Tracking Competitor ASINs Alongside Your Own. Keywords One per line, with a live 0 / 25 counter. Pick them the way you would pick a watchlist, not a wish list: - Terms you already rank for, so you can see erosion before it costs you. - Terms you are paying for, so you can tell whether the spend is buying position or replacing it. - The one or two you are actively trying to win, and nothing else aspirational. A project stuffed with twenty-five terms you rank #90 for produces a wall of amber and tells you nothing. See Reading Visibility, Average Position and Top 10. Enable tracking Enable tracking — when enabled, keywords are tracked daily The toggle is on by default. Tracking runs daily once enabled, and history accumulates from that day forward — there is no backfill, so a project created today has a chart that starts today. That is the practical argument for creating projects before you need them. The launch you want a rank history for in March is one you had to start tracking in January. Turning the toggle off keeps the project and its history but stops collecting. It is the right move for a discontinued line you may revisit, and it frees a slot against the 50-project limit. How to divide 50 projects Each project is answering one question. The groupings that hold up: | Grouping | The question it answers | |---|---| | Branded terms | are we defending our own name? | | One product line | is this line visible for its category terms? | | Competitor set | where do we sit against the ASINs we lose to? | | A launch | is the new listing climbing? | What does not hold up is one project per ASIN. You get 50 disconnected charts and no comparison — and comparison across products is exactly what the Compare tab exists for. What you get back The Rank Tracker portfolio, one row per project Saved projects land on the Rank Tracker portfolio with Products, Keywords, Visibility, Avg Position, Top 10, Amazon's Choice and Status, plus a totals row across everything you track. From there, opening a project gives you the Products and Compare tabs, and opening a product gives you its Rank Heatmap, Keywords and BSR Tracker. Related - Reading Visibility, Average Position and Top 10 — what the columns mean. - Tracking Competitor ASINs Alongside Your Own — the other half of the picker. - Best Seller Rank Tracking — the third rank this screen records. - Organic vs Sponsored Rank — why one keyword has two positions.

Reading Visibility, Average Position and Top 10

Rank Tracker reports the same four measures at three levels — portfolio, project, product — and they mean the same thing at each. Learning them once is most of learning the screen. The one to be careful with is Visibility, because it is an average of averages, and averages hide the thing you most want to find. Watch the walkthrough Prefer to watch? Seventy-nine seconds on the four columns, and the zero that drags a project down. The four measures The Rank Tracker portfolio — the same columns at every level | Measure | What it is | Direction | |---|---|---| | Visibility | share of voice across the keywords you track | higher is better | | Avg Position | mean organic rank across those keywords | lower is better | | Top 10 | how many tracked keywords rank in the first ten | higher is better | | Amazon's Choice | badges held across the set | higher is better | Two of the four move in opposite directions, which is the single most common misreading of this screen. A project whose Avg Position went from 14.4 to 20.5 got worse. See Tracking the Amazon's Choice Badge for the fourth. Visibility averages upward A product's visibility is its share of voice across the keywords tracked for it. A project's visibility is the plain mean of its products'. The portfolio figure is the plain mean of the projects'. That is worth stating plainly, because it means a single dead product silently drags a project down: | Product | Visibility | |---|---| | MGO 80+ Squeezy | 76.00% | | Raw Manuka Honey MGO 30+ | 96.00% | | Manuka Honey Squeezy MGO 182+ | 68.00% | | MGO 80+ Squeezy (second listing) | 0.00% | | Project | 60.00% | A project whose average is carrying a zero The project reads 60% and looks mediocre. It is actually three healthy listings and one that returns nothing — a duplicate listing, a suppressed ASIN, or a product that simply does not rank for these terms. So always open the project before believing its headline. A number in the sixties is either "everything is middling" or "most things are fine and one is broken", and those need completely different responses. A 0.00% row is covered in Rank Data Shows Blank. Avg Position is a mean, so it hides its own shape Avg Position tells you the centre. It does not tell you whether you have a tight cluster around #16 or two clusters at #4 and #40 — and those are different businesses. Rank distribution beside average position The product view solves this directly with Rank Distribution, a stacked bar splitting the tracked keywords into the standard bands: 1–3 · 4–10 · 11–20 · 21–50 · 51–100 An average of 16.5 that is mostly 4–10 with a tail at 21–50 is a strong product with a few terms it should stop chasing. The same 16.5 spread evenly across every band is a product with no clear position at all. Read the distribution first, then the average. Top 10 is the honest one Top 10 is a count, so it cannot be dragged by outliers. Nobody argues about whether a keyword is in the first ten. Use it as the check on the other two. A project whose Avg Position improved while Top 10 stayed flat did not win anything that matters — it moved terms from #60 to #40, which changes the mean and changes no customer's behaviour. Moving a keyword from #11 to #9 is worth more than moving five keywords from #60 to #40. Top 10 is the column that says so. Keywords Covered, on the Compare tab The Compare tab — keywords covered and average best position The Compare tab swaps in two measures that answer a different question — not "how is this product doing" but "does the project cover its keywords at all": - Keywords Covered — how many keywords have at least one product ranking in the top 50, shown as a count and a percentage. - Avg Best Position — the mean of the best rank per keyword, across every product in the project. These are catalog-level questions. Keywords Covered at 96% means one product or another shows up for nearly every term; the gap is the term nothing you sell ranks for. Avg Best Position of 1.8 on a branded project means your own name reliably returns you first. Reading the three levels in order Portfolio — which project moved? Sort by Visibility, then check Avg Position moved the same way. Project — is it every product, or one? Look for a row far from its siblings. Product — which keywords? Open the Rank Heatmap and read the columns right to left. Three clicks, narrowing each time. Doing it in the other order — starting from a keyword you have a hunch about — is how you spend an afternoon confirming something that was never the problem. What not to conclude Visibility is not traffic. It is a rank-derived measure across the keywords you chose. Adding twenty-five easy terms will raise it without selling a single extra unit. Averages across different intents are meaningless. A project mixing branded terms with broad category terms produces an Avg Position that describes neither. A single day is noise. Amazon search results are personalised and volatile. The heatmap exists because the trend is the signal. Related - Rank Tracker: Creating Your First Project — how the keyword set gets chosen. - Rank Data Shows Blank — the 0.00% rows. - Best Seller Rank Tracking — the rank that is not a search rank. - Organic vs Sponsored Rank — the second position on every keyword.

Tracking the Amazon's Choice Badge

Amazon's Choice is a badge Amazon attaches to a product for a specific search term — not to the product itself. Rank Tracker counts those badges everywhere it counts rank, which makes it the one place you can see the badge behave like the keyword-level thing it actually is. Watch the walkthrough Prefer to watch? Sixty-eight seconds on why the badge belongs to the keyword, not the product. The badge is per keyword, not per product This is the fact everything else follows from. The same ASIN can hold Amazon's Choice on one term and hold nothing on the term next to it. Amazon's Choice counted across the portfolio So the number in the Amazon's Choice column is a count of badges, not a yes/no. At portfolio level it is badges across everything you track; at project level, "Choice badges across products"; at product level, badges out of the keywords tracked for that product. A product showing 15 in a 25-keyword project holds the badge on fifteen of those terms and not on the other ten. Where it appears | Level | What the count means | |---|---| | Portfolio | badges across every tracked project | | Project | badges across the products in it | | Product | badges out of that product's tracked keywords | | Compare grid | a marker on the individual keyword × product cell | Per-product badge counts inside a project The per-product column is where the badge stops being a vanity metric. Three products in the same project holding 15, 3 and 0 is not a story about brand strength — it is a story about which listing Amazon currently considers the default answer. Why you should care about the count The badge is worth watching for a reason that has nothing to do with prestige: It moves conversion on terms you may not be bidding on. Amazon's Choice sits in the organic result, and it disproportionately affects the shopper who is not comparing — the one typing a generic term and buying the first credible option. That is the traffic hardest to buy and easiest to lose quietly. It signals what Amazon thinks the default is. Amazon awards it to a well-rated, well-priced, reliably in-stock product that ranks highly for the term. Losing it is therefore a symptom: something in that list changed before the badge did. It is the cheapest early warning you have. A badge count that drops from 15 to 9 over a fortnight is usually the first visible consequence of a stock-out, a price move, or a review-rating slide — all of which show up here days before they show up in sales. Reading a change A badge count that falls is worth a specific sequence of checks, in this order: 1. Did you go out of stock? Even briefly. This is the most common cause by a distance, and the badge does not always come straight back. Check Inventory. 2. Did the price move? Yours or a competitor's. The badge is sensitive to price competitiveness within the term. 3. Did the rating slip? A drop below the neighbours on the same term is enough. 4. Did organic rank fall first? Open the Rank Heatmap for the product. If rank moved before the badge did, the badge is the consequence, not the cause — fix the rank. 5. Did a competitor take it? The badge is awarded per term, so somebody else generally now has it. A competitor project on the same terms answers this directly — see Tracking Competitor ASINs Alongside Your Own. In the heatmap and the grid The rank bands and the Amazon's Choice marker Both the Compare grid and the product Rank Heatmap carry an Amazon's Choice marker in the legend alongside the organic rank bands, so a badged cell is identifiable at the level that matters: this product, this keyword. That is the view to use when the count changed and you need to know which term you lost. What not to conclude More badges is not automatically better. Badges on twenty-five low-volume long-tail terms are worth less than one badge on the term that carries your category. Read the count beside Search Vol. The badge does not replace rank. A product can hold Amazon's Choice on a term and still be beaten on the page by sponsored placements above it. See Organic vs Sponsored Rank. You cannot apply for it. There is no lever. Everything you can do is upstream — availability, price, rating, rank — which is exactly why the count is useful as a readout rather than a target. It moves on Amazon's schedule. Day-to-day flicker is normal; a sustained fall over a week or more is the signal. Related - Reading Visibility, Average Position and Top 10 — the other three columns. - Rank Tracker: Creating Your First Project — choosing terms worth holding a badge on. - Best Seller Rank Tracking — the other Amazon-assigned marker on the same screen. - Tracking Competitor ASINs Alongside Your Own — who has it now.

Tracking Competitor ASINs Alongside Your Own

Rank Tracker will track any ASIN on Amazon, not only the ones in your catalog. Paste a competitor's ASIN into a project and it is tracked on the same keywords, on the same days, and appears as another column everywhere the project reports. That turns "are we doing well?" into the question that actually has an answer: doing well compared to whom? Watch the walkthrough Prefer to watch? Eighty-six seconds on the ASINs field, the compare grid, and who to track. Adding them On the Create project form, below the catalog picker: Track other ASINs — one per line, no catalog listing required Track a competitor's or vendor's ASIN alongside your own products — no catalog listing required. One ASIN per line, and the field is optional. There is no verification step and no relationship required — if the ASIN exists, it gets tracked. The vendor case is worth calling out separately: if you sell through a vendor relationship as well as your own seller account, the vendor-listed ASIN is not in your seller catalog. This field is how you watch it. What you get Competitor ASINs behave like your own products throughout the project. They get organic rank per keyword, Best Seller Rank, and Amazon's Choice badges — recorded daily. Every product in the project as a column, keyword by keyword The Compare tab is where this pays off. Keywords down the side, Search Vol beside them, one column per tracked ASIN, and a colour-coded rank in every cell. Your products and theirs, on the same terms, on the same day. Read it across a row, not down a column. A row where you sit at #16 and two competitors sit at #5 and #7 is a term you are losing, and the search volume beside it tells you whether losing it matters. Choosing which ASINs to track The instinct is to add the market leader. That is usually the least useful choice — a dominant ASIN sits at #1 on everything and its row never moves, so it teaches you nothing week to week. Track the ones you actually trade positions with: | Add | Why | |---|---| | The ASIN directly above you on your best terms | the gap you can realistically close | | A recent entrant climbing fast | you find out how fast before it reaches you | | The cheap substitute | tells you whether you are losing on price or on relevance | | Your own vendor-listed ASIN | often competing with your seller listing | Remember the ceiling: 25 keywords per project, and competitor ASINs share those keywords rather than getting their own. Adding four competitors does not cost you keywords — it costs you columns to read. Three or four is usually the point where the grid stops being scannable. The four readings worth doing Head-to-head on a term. One row, your rank against theirs. This is the only rank comparison that is genuinely like-for-like, because it is the same query on the same day. Their BSR trend. You cannot see a competitor's sales, but BSR is a sales-velocity rank and it is public. A competitor whose BSR climbs steadily while yours is flat is taking share right now — see Best Seller Rank Tracking. Who holds the badge. When your Amazon's Choice count falls, the badge went somewhere. A competitor column on the same term names them. See Tracking the Amazon's Choice Badge. Coverage gaps. The Keywords Covered tile counts terms where at least one tracked product ranks in the top 50. A keyword where every one of your columns is blank and theirs is not is a term you are absent from — the most actionable cell on the screen. Turning it into action A competitor beating you on a term has exactly three explanations, and the grid plus your other screens separate them: They rank better organically. Check Search Query Performance — if your click share is low on a term you rank near them for, it is a listing problem, not a rank problem. They are buying the position. Check the Sponsored column on your product's Keywords tab. If you rank #8 organically and they sit above you in paid, the fight is a bidding one. See Organic vs Sponsored Rank. They are simply selling more. Their BSR will show it. This is the slow one, and the one advertising cannot fix on its own. What not to conclude Rank is not share. A competitor at #3 to your #6 is ahead on that term. It says nothing about total revenue, breadth of catalog, or margin. Their absence is not weakness. An ASIN that ranks nowhere on your terms may be winning on terms you did not think to track. One day is noise. Amazon results are personalised and volatile. Compare medians and trends in the heatmap, not yesterday's cell. This is observation, not intelligence. You see public rank and public BSR. You do not see their spend, their margin, or their inventory — so do not build a pricing strategy on an inferred cost base. Related - Rank Tracker: Creating Your First Project — where the ASINs field lives. - Best Seller Rank Tracking — the competitor signal with the most in it. - Tracking the Amazon's Choice Badge — finding who took it. - Generic, Branded and Competitor Terms — choosing the keywords to compare on.

Generic, Branded and Competitor Terms

"Is our branded ACoS good?" is a question you cannot answer until something has decided which queries count as branded. In TrackIQ, you decide — and the same setting quietly governs several screens at once. Watch the walkthrough Prefer to watch? Eighty-two seconds on the three buckets, and the one word that ruins the split. Where it lives On the SQP dashboard, the Brand & competitor terms control opens the editor. How a query gets classified A query is Branded if it contains one of your terms, Competitor if it contains a rival's, otherwise Generic (matched as whole words). Three buckets, one rule, applied to every query the dashboard loads. The three buckets | Bucket | Means | Reading | |---|---|---| | Branded | contains one of your brand terms | people looking for you | | Competitor | contains a rival's brand term | people looking for them | | Generic | neither | people looking for the thing | They behave completely differently, which is the entire reason for splitting them: Branded queries convert well and prove nothing. A high branded conversion rate is not a compliment to your listing; the shopper had already decided. Branded ACoS is normally excellent, and averaging it in with everything else flatters your account. Generic queries are where growth lives. Somebody typing the category is choosing between you and everyone else, and that is the fight worth measuring. Competitor queries are the expensive ones. You are interrupting somebody who wants a different brand. Sometimes worth it, always dear. Setting your own terms Your brand terms, one per line One term per line, in each of two boxes — yours and your rivals'. And the instruction in the panel is the one thing to get right: Use full brand phrases shoppers type — not single generic words like honey or drops, which mark whole categories as branded. This is the mistake that ruins the whole split. If your brand is Northwind Goods and you add goods as a term, every query containing that word becomes branded — and your generic bucket, the one that measures whether you can win new customers, empties into it. Add full phrases. The brand as people type it, plus the misspellings you actually see. Add sub-brands and line names, if shoppers search them by name. Do not add category words, however strongly you associate them with yourself. For competitors, add the rival brand names that turn up in your search terms. You will find them in Search Terms and Aggregated Search Terms. Check the split before you save The live classification count The panel classifies your loaded queries as you type: With these terms, your 301 loaded queries classify as 15 Branded (5%) · 11 Competitor · 275 Generic. Use it as a sanity check, because the failure modes are visible immediately: | Split looks like | Probably means | |---|---| | Branded share far above expectation | a generic word is in your term list | | Branded near zero on a known brand | the phrase does not match how people type it | | Competitor zero | you have not added any rivals yet | A branded share in the low single digits is normal for a category-led business. A well-known brand will sit far higher. What matters is that the number matches what you would have guessed. The two consequences people miss It changes your Amazon Ads reporting too. The panel says so: Brand terms are shared with the Amazon Ads brand vs non-brand split. One list, two places. Edit it here and your advertising screens reclassify as well. That is convenient and it means a careless edit has a wider blast radius than the screen suggests. It changes which queries get recommended actions. Competitor-brand queries are deliberately excluded from the Fix listing, Fix conversion and Add-keyword recommendations, for a sound reason the panel states plainly: A better page can't win a shopper who wants their brand. So classifying a query as competitor does not just label it — it removes it from the action queue. A rival's name missing from your competitor list means TrackIQ will keep recommending you improve a page that was never going to win that search. See Recommended Actions: The Queries That Need a Decision. How to use the split Read metrics per bucket, never blended. Set separate ACoS expectations for branded, generic and competitor. One target across all three is meaningless. Track generic share over time. Growing generic volume is growing reach. Growing branded volume is your existing customers coming back — good, but not the same thing. Judge campaigns by which bucket they serve. A campaign living on branded terms should be cheap and efficient. One fighting for generic terms should not be held to the same standard. Revisit after a launch or rebrand. New line names become new brand terms, and until they are in the list they count as generic. What not to conclude This is your classification, not Amazon's. Two accounts can label the same query differently. It is not a standard. Whole-word matching is literal. A misspelling shoppers actually type will not match unless you add it. A high branded share is not automatically good. It can mean strong brand demand, or it can mean you have stopped competing for the category. Related - Recommended Actions: The Queries That Need a Decision — what the competitor exclusion changes. - Search Terms and Aggregated Search Terms — where to find rival names to add. - Brand Keywords: Proving Upper-Funnel Spend Works — the AMC view built on the same list. - Understanding Search Query Performance — the dashboard this control sits on.

Search Query Cart Analysis: Why a Query Doesn't Convert

A query can be doing well at every stage except the one that matters. Plenty of impressions, a healthy click rate, and then people put it in the basket and never buy — or never reach the basket at all. Click a query on the SQP Analysis table and TrackIQ opens its funnel, stage by stage, with the market beside you at every step. Watch the walkthrough Prefer to watch? Eighty-five seconds on the four funnel stages, and the classic misdiagnosis. Opening it Search Query Performance → SQP Analysis, then click for details under any query. The note above the table says what you get: Click a query for its funnel (brand vs market) and ASIN split. The table is sorted by SQ Score and shows the top 200 queries, five weeks each, with chips filtering to Generic, Branded and Competitor. Those three come from your own term list — see Generic, Branded and Competitor Terms. The window summary Window summary for a single query Eight tiles across a five-week window: | Tile | Reads | |---|---| | Search Volume | total market searches for the query | | Click Share | your clicks as a share of the market's | | Purchase Share | your purchases as a share of the market's | | SQ Score | the query's rank in your list — 1 is top | | Funnel Score | 5.0 = at market, uncapped | | CVR vs Market | 1.0× = at market | | SP+SB Spend | what you paid on this query, with sales beside it | | ACoS | for the query, and whether an exact target exists | Funnel Score and CVR vs Market are the two to read first, because both are calibrated against the market rather than against a target you invented. A funnel score above 5.0 and a CVR multiple above 1.0× mean you are beating the average seller on this query. Below means you are not, and the funnel below tells you where. The ACoS tile also flags "exact target: exists" — whether you are already bidding on this query with an exact-match keyword. A query performing well with no exact target is a harvest candidate. See Harvest Winning Search Terms into Exact-Match Campaigns. The funnel, brand against market Impressions, clicks, cart adds and purchases, each against the market Four stages, each showing your share, your absolute figure beside the market's, and from the second stage on, your rate against the market's rate: | Stage | Shows | Rate compared | |---|---|---| | Impressions | share of all impressions for the query | — | | Clicks | share of all clicks | CTR vs market CTR | | Cart Adds | share of all cart adds | Cart rate vs market | | Purchases | share of all purchases | CVR vs market | Rates compare your brand with the full market (all sellers) at each funnel step. That comparison is the whole point. A 2% CTR means nothing on its own; a 2% CTR against a market at 1.5% means your listing is winning the click. The same number against a market at 4% means it is losing badly. Reading where it breaks Walk the four stages and find the first one where your share drops relative to the stage before it. That drop is the problem, and each one has a different fix: | Pattern | What it means | Where the fix is | |---|---|---| | Impression share low | you are barely shown | rank and bids | | Impression share fine, click share drops | shown and not chosen | main image, title, price, rating | | Click share fine, cart share drops | they arrive and leave | the detail page — content, A+, reviews, price | | Cart share fine, purchase share drops | they abandon at checkout | price, delivery, availability, Buy Box | The last row is the rarest and the most annoying, because it is furthest from anything advertising can influence. The classic misdiagnosis is treating a click-share problem as a bid problem. If you are already being shown as often as your share of the category warrants and people are not clicking, more impressions just costs more money to be ignored at a larger scale. The ASIN breakdown Which of your products the query actually reaches The same query, split by the products it landed on, with Click Share, Cart Share and Purchase Share per ASIN, and a View details link on each. This answers a question the funnel cannot: is the query underperforming, or is it reaching the wrong product? One ASIN dragging the average. If four products convert and one does not, the query is fine and the listing is not. Fix or exclude that ASIN rather than abandoning the term. Click share high, purchase share low on the same ASIN. That product is attracting the query and failing it. A detail-page problem, narrowed to one page. A product you did not expect. Sometimes a query is being served by a variant you never intended to advertise for it. That is a targeting fix, not a listing one. What to do with each finding Losing the click — the listing needs work before any bid change. Main image first; it does most of the work in a search result. Losing the cart — the detail page. Compare your page against the sellers winning the query. Losing the purchase — check price, delivery promise, stock and Buy Box status. See Catch a stockout before it costs you rank. Winning everywhere but small share — this is the good problem. You convert better than the market and are not shown enough. Bid up, and check rank. See Organic vs Sponsored Rank. What not to conclude Share is not quality. A small brand will have small shares on every stage. Read the rates against market, not the shares. SQP is weekly. The window is five weekly cohorts, not an arbitrary range. Compare week to week, not day to day. See SQP Is Weekly: Reading Week Cohorts Correctly. Market includes everybody. The comparison is against all sellers on that query, including ones who are not your competitors in any meaningful sense. A single week is thin on a low-volume query. The five-week window exists for a reason. Related - The SQP Dashboard: Market Share, Analysis and Recommended Actions — the screen this sits inside. - Generic, Branded and Competitor Terms — the chips above the table. - Recommended Actions: The Queries That Need a Decision — the same findings, pre-sorted by money. - Work out which queries lose you the sale at the cart — the playbook version.

SQ Score and Funnel Score

Two scores appear beside every query on the SQP Analysis table, they are colour-coded, and neither is explained anywhere obvious. They are worth understanding, because between them they answer "where should I look first?" faster than any column of raw numbers. Watch the walkthrough Prefer to watch? Seventy-nine seconds on a rank, a market-calibrated grade, and the grid they make together. Where they appear Search Query Performance → SQP Analysis. The analysis table, sorted by SQ Score The note above the table sets out the convention: Sorted by SQ Score. Green Funnel/CVR = beats market (scale), red = trails (fix). Click a query for its funnel (brand vs market) and ASIN split. That single line contains most of what you need. The colour tells you what to do; the score tells you how much. SQ Score — the priority rank SQ Score is a rank, and 1 is top. It orders your queries by how much they matter to you, so the table opens with the query most worth your attention rather than the one with the biggest search volume. That distinction matters: the highest-volume term in your category may be one you have no realistic chance on, while a mid-volume term you already convert well on is where the next hour is best spent. Read it as a queue, not a grade. SQ Score 1 does not mean "excellent"; it means "start here". The table shows the top 200 queries by SQ Score, five weeks each. Funnel Score — how you convert, against the market Funnel Score is calibrated so that 5.0 is at market, and it is uncapped. The score in the query's window summary This is the part that makes it useful. Most scores in most tools are graded against a target somebody invented. This one is graded against what every other seller on that query actually achieves. | Funnel Score | Means | |---|---| | above 5.0 | you move shoppers through the funnel better than the market | | 5.0 | exactly average for this query | | below 5.0 | the market converts this query better than you do | Uncapped matters too: a score of 9 is genuinely twice as far above market as a 7, so the differences between strong queries stay visible instead of everything bunching at the top of a 1–10 scale. CVR vs Market — the same idea, as a multiple Beside it sits CVR vs Market, where 1.0× is at market. A query at 1.10× converts ten percent better than the average seller on that search. Funnel Score covers the whole journey — impressions through to purchase. CVR vs Market is just the final step. When they disagree, that is informative: a strong funnel score with a weak CVR multiple means you win the click and lose the checkout. Reading the colours The colour coding is a two-word instruction: Green — scale. You beat the market on this query. The constraint is visibility, not quality, so bidding up and improving rank will buy more of something that already works. Red — fix. The market converts this better than you. Bidding up buys more of a problem. Open the funnel and find the stage that breaks. See Search Query Cart Analysis: Why a Query Doesn't Convert. That is the whole decision, and it is the opposite of what most people do with an underperforming term, which is spend more on it. Using the two together The scores are most useful as a grid rather than individually: | SQ Score | Funnel Score | What it is | Do | |---|---|---|---| | high priority | green | a proven winner you under-serve | scale — bid, rank, harvest it | | high priority | red | an important query you are losing | fix — listing work, then revisit | | low priority | green | works, but small | leave it; note it for later | | low priority | red | not worth the effort | ignore, or negative-match | The top-left cell is where money is. A query already near the front of your queue that you convert better than anybody else on, and are not shown enough for, is the cheapest growth available — and it is exactly the case where raising a bid is the right move. What not to conclude Green does not mean profitable. It means you convert better than the market. A query can beat the market and still cost more per sale than your margin allows — check ACoS beside it. A low SQ Score is not a bad query. It is a lower priority for you, given your products and performance. The market includes everybody. All sellers on that query, not just your real competitors. On a broad term that pool is wide. Five-week window. Both scores are computed across the window, not for a single week. SQP reports weekly cohorts — see SQP Is Weekly: Reading Week Cohorts Correctly. Related - Search Query Cart Analysis: Why a Query Doesn't Convert — what to do when a score is red. - The SQP Dashboard: Market Share, Analysis and Recommended Actions — the screen these sit on. - Recommended Actions: The Queries That Need a Decision — the same judgement, pre-sorted by money. - Generic, Branded and Competitor Terms — the chips that filter the table.

SQP & Search Terms: Matching Demand to What You Actually Ran

SQP tells you what the whole market searched for. Your Search Terms report tells you what your ads actually ran on. They are two views of the same shopper behaviour, and the gap between them is where most missed opportunity lives. The SQP & Search Terms tab puts them side by side so you do not have to reconcile two reports by hand. Watch the walkthrough Prefer to watch? A hundred and one seconds on the demand you never reached, and why it cannot appear in your ad reports. Where it lives Search Query Performance → SQP Analysis. The two tabs on the analysis table Two tabs sit above the table: SQP & Search Terms — market demand matched against the customer searches your ads were actually served on. SQP & Keywords — market demand matched against the keywords you bid on. See SQP & Keywords: Finding Queries You Convert On But Don't Bid On. The difference matters more than it sounds. A keyword is what you chose to bid on. A search term is what a customer actually typed. Broad and auto matching means the two are frequently not the same phrase at all. The four situations Matching demand against what you ran produces four cases, and each has a different action: | Market demand | You ran on it | Meaning | Do | |---|---|---|---| | high | yes, converting | working | scale — bid and rank | | high | yes, not converting | you are present and losing | fix — listing or page | | high | no | demand you never reached | add it | | low | yes, spending | you are paying for a thin query | cut it | The third row is why this tab exists. A high-volume query where you have no presence at all will never appear in your Search Terms report — because your ads never ran on it, there is nothing to report. It is invisible in advertising reporting by construction, and visible here. The gap worth hunting Work in this order: 1. Sort by search volume. Start where the demand is. 2. Find rows with volume and no click share. The market searches this; you are not being clicked. Either you are not shown, or you are shown and ignored. 3. Check whether you are shown at all. Impression share separates the two: - No impression share — you are absent. This is a targeting gap: add the term. - Impression share, no clicks — you are shown and passed over. This is a listing problem, and adding keywords will not fix it. See Search Query Cart Analysis: Why a Query Doesn't Convert. That distinction determines whether you spend money or spend effort, and getting it backwards is the common mistake. The other direction — what you ran that nobody wanted The same tab exposes the reverse case: terms your ads served on that carry very little market demand. Broad and auto matching drifts. Some of that drift is discovery and worth keeping; some of it is spend on phrases almost nobody searches. Sort by your spend against market volume, and the thin-demand, real-spend rows are negative-match candidates. This overlaps with the aggregated Search Terms view in Amazon Ads, which is the better screen for the spend side. See Search Terms and Aggregated Search Terms. Read the classification chips The Generic / Branded / Competitor chips above the table filter the population, and they change what a gap means: A generic query you are absent from is a genuine growth opportunity. A branded query you are absent from is unusual and worth checking — it is your own name. A competitor query you are absent from is often a deliberate choice, and expensive to change. Note that competitor-brand queries are deliberately excluded from Fix listing and Add-keyword recommendations, because a better page cannot win a shopper who wants somebody else's brand. See Generic, Branded and Competitor Terms. Turning findings into keywords A gap you decide to close becomes an exact-match keyword, in a campaign where you can see it, at a bid informed by what the query already costs the market. Then negative-match it in whichever broad or auto campaign might otherwise pick it up, so you are not bidding against yourself. The full routine is in Harvest Winning Search Terms into Exact-Match Campaigns. What not to conclude Absence is not always a mistake. Some high-volume queries are wrong for your product, and the right answer is to stay out. Market volume is not your opportunity. You will win a share of it, not all of it. Estimate with your existing share on similar terms. Weekly cohorts. SQP reports whole weeks; your Search Terms report is daily. Compare like with like. See SQP Is Weekly: Reading Week Cohorts Correctly. SQP is not real-time. It lags, so a term you added this week will not appear here immediately. Related - SQP & Keywords: Finding Queries You Convert On But Don't Bid On — the sibling tab. - Search Terms and Aggregated Search Terms — the advertising-side report. - Generic, Branded and Competitor Terms — the chips that filter the table. - Harvest Winning Search Terms into Exact-Match Campaigns — acting on the gap.

Account vs Products: Two Ways to Read SQP

The SQP dashboard opens on two tabs, Account and Products, and they answer different questions with the same numbers. Reading the wrong one is how people conclude their share is fine when one product is quietly losing its category. Watch the walkthrough Prefer to watch? Eighty-three seconds on what changes when you switch tabs, and why the denominator is the whole story. The difference in one line Account measures your brand against the market across every tracked query. Products measures one ASIN at a time against the market on the queries that reach it. The chart caption changes with the tab, which is the clearest tell: brand funnel share lines on Account, product funnel share lines on Products. Account — the portfolio view Account-level shares across every tracked query Five tiles, all computed across your whole tracked query set for the window: | Tile | Reads | |---|---| | Click Share | your clicks as a share of the market's, with the change in pp | | Purchase Share | your purchases as a share of the market's | | Keywords Tracked | how many queries, and how many are new | | PPC on Tracked Queries | SP + SB search-term spend, with its ACoS | | Actions | how many recommendations are waiting, split by type | This is the tab for "how are we doing?" — the one you read weekly, and the one that belongs in a report. Watch the units. Share changes are in percentage points, not percent. A move from 2.12% to 2.28% is +0.16 pp, not +7.5%. Reporting it the second way overstates it dramatically. Products — the diagnostic view The same tiles, recomputed for one ASIN Switch to Products and a product pager appears — prev, next, a dropdown, and a position counter like 1/33 — with the number of queries in the window beside it. Every tile then recomputes for that one ASIN. A product with a 1.52% click share sits inside an account at 2.12%: the account figure is an average, and this is one of the things being averaged. This is the tab for "why?" You come here after Account tells you something moved. Why the two disagree, and what that means The account figure is a blend. Divergence between the tabs is the finding, not an error: | Pattern | Reading | |---|---| | Account flat, one product falling sharply | a product problem hidden by the average | | Account falling, all products falling similarly | something account-wide — brand, price, category | | Account rising, driven by one product | concentration risk — check what happens if it stocks out | | Product strong, account weak | the rest of the catalog is dragging | The first row is the one that costs money, because nothing on the Account tab will ever show it. A catalog of thirty products can absorb one falling badly without the headline moving at all. How to use them together Weekly, read Account. Click share and purchase share against last week. That is your five-minute check. When something moves, open Products. Page through the ASINs and find the ones that moved most. The pager makes this quicker than it sounds — you are scanning for outliers, not reading each one. When you find the product, open the query. Clicking a query gives its funnel against the market and the ASIN split. That is where the cause lives. See Search Query Cart Analysis: Why a Query Doesn't Convert. Account → product → query is the same narrowing discipline as the rank triage playbook, and it works for the same reason: averages hide, specifics explain. The Actions tile is a shortcut On both tabs, Actions counts the waiting recommendations and breaks them down by type — Fix listing, Cut bids, Scale up, Fix conversion, Add as keyword. On the Products tab this is unusually useful, because it tells you what kind of problem a specific ASIN has before you read a single chart. A product whose actions are all Cut bids has an efficiency problem; one whose actions are all Fix listing has a conversion problem. See Recommended Actions: The Queries That Need a Decision. What not to conclude A low share is not automatically bad. In a large category, single-digit share is normal. Read the direction and the rates, not the absolute. Products is not your whole catalog. It covers ASINs with tracked query data. A product with no SQP data will not appear. The market is everybody. All sellers on those queries, not just your competitors. SQP is weekly. Both tabs report weekly cohorts across the window, so compare week to week. See SQP Is Weekly: Reading Week Cohorts Correctly. Related - Understanding Search Query Performance — the dashboard overall. - Impression, Click, Cart-Add and Purchase Share — what the four shares mean. - Search Query Cart Analysis: Why a Query Doesn't Convert — the level below Products. - Triage a Rank Drop Across Your Top Keywords — the same narrowing discipline.

Impression, Click, Cart-Add and Purchase Share

Four shares run through every SQP screen, and they are the same four at every level — account, product, and individual query. Together they describe a funnel; individually each one answers a different question. Once you can read the four as a sequence, most SQP diagnosis is a matter of finding where the sequence breaks. Watch the walkthrough Prefer to watch? A hundred and six seconds on reading four shares as one funnel, and the two traps in reporting them. The four Market share, with all four funnel lines Every share is yours as a proportion of the whole market's for the same queries: | Share | Counts | Answers | |---|---|---| | Impression share | how often you were shown | are you visible? | | Click share | how often you were clicked | are you chosen? | | Cart-add share | how often you reached the basket | does the page work? | | Purchase share | how often you were bought | does the sale close? | The Market share panel plots all four as lines against search-volume bars, and the table beneath gives the raw figures. The weekly table: market and brand at each stage, with the share between The table is the clearest statement of what a share actually is. Each stage gets three columns — the market's number, your number, and the share between them. Impressions, clicks, cart-adds, purchases, week by week, with a total row. The middle column is named for the scope you are on. On Account it reads BRAND; on Products it reads PRODUCT. Same arithmetic, different denominator population. See Account vs Products: Two Ways to Read SQP. Read them as a sequence, not four numbers The four are cumulative. Each one starts from the population the previous stage produced, so a weakness early carries forward whether or not the later stages are healthy. So find the first stage where your share drops relative to the one before it. That step is the problem; everything after it is a symptom. | Where the drop is | What it means | The fix lives in | |---|---|---| | impression share already low | you are not being shown | rank, bids, budget | | impression → click | shown and not chosen | image, title, price, rating | | click → cart | visited and not added | detail page, content, reviews | | cart → purchase | added and not bought | price, delivery, stock, Buy Box | Shares that hold flat across all four are good news, even if the absolute number is small. It means you convert like the market and simply need more visibility — which is the one case where spending more is the right response. Percentage points, not percent The single most common reporting error on these screens. Share changes are shown in pp — percentage points. A click share moving from 2.12% to 2.28% is +0.16 pp. Describing that as "up 7.5%" is technically a relative change but reads as a much bigger move than it is, and in a board pack it is misleading. Report shares in pp. If somebody wants the relative figure, give both. Small shares are normal A single-digit share — often a low single-digit one — is ordinary for most sellers on most queries. The market includes every seller on that search, including large ones you do not compete with in any meaningful sense. So do not read the absolute number as a grade. Read: Direction. Is it moving, and which way? Consistency across the four. Where does it break? Rates against market, which the query-level funnel gives you. A 2% click-through rate against a market at 1.5% is a listing that wins the click, whatever your share of the category. See Search Query Cart Analysis: Why a Query Doesn't Convert. Cart-add share is the underused one Impression, click and purchase get the attention. Cart-add sits between click and purchase and isolates a stage most reporting skips entirely. A healthy cart share with a poor purchase share is a narrow, tractable problem: people wanted it, put it in the basket, and something at checkout stopped them. Price, delivery promise, availability, Buy Box. A poor cart share with a healthy click share is the detail page. They arrived, looked, and left without ever intending to buy. Those two get treated as the same problem constantly, and they are not. Where the same four appear | Screen | Scope | |---|---| | SQP → Account | your brand, all tracked queries | | SQP → Products | one ASIN, its queries | | Query funnel (click a query) | one query, with rates against market | | Market share table | week by week, market and brand side by side | The query funnel is the most useful of the four because it adds the market's rate at each step, not just your share. See Account vs Products: Two Ways to Read SQP. What not to conclude Share is not conversion rate. Share compares you to the market; a rate compares outcomes to inputs. You can hold share while your conversion falls, if the whole market falls with you. Impression share is not rank. It is how often you appeared, paid and organic combined. Weekly cohorts. SQP reports whole weeks. Comparing an arbitrary date range against another is how phantom movements appear. See SQP Is Weekly: Reading Week Cohorts Correctly. Purchase share does not include everything you sold. Only purchases tied to the tracked queries. Related - Understanding Search Query Performance — the dashboard overall. - Account vs Products: Two Ways to Read SQP — the same shares at two scopes. - Search Query Cart Analysis: Why a Query Doesn't Convert — the four with market rates attached. - Work out which queries lose you the sale at the cart — the playbook.

Spotting New Search Queries Week Over Week

Amazon's search demand is not static. New phrases appear every week — a trend, a competitor's campaign, a seasonal shift, a product nobody was searching for a month ago. TrackIQ counts them for you, on a tile most people read as a total and never look at again. Watch the walkthrough Prefer to watch? Eighty-seven seconds on a tile most people read as a total, and the ten-minute weekly routine behind it. The number is already on the screen Keywords tracked, with the new count beside it Search Query Performance → Keywords Tracked. The tile shows the total, and beneath it a green figure with the definition attached: new = first seen in the latest week That is the whole feature. A count of queries that were not in your data last week and are now, refreshed every week, sitting on the dashboard you already open. It appears on both Account and Products, so you can ask the question for the brand or for one ASIN. See Account vs Products: Two Ways to Read SQP. Why new queries matter more than they look A new query is demand that did not exist for you last week. Three things follow: You are almost certainly not bidding on it. New queries arrive through broad, phrase and auto matching, at broad-match prices, if they arrive at all. Nobody has optimised for it yet. Competition on a genuinely new phrase is thinner than on an established one, which is the cheapest position you will ever get. It tells you something about the market. A cluster of new phrases around one attribute is a shift in how people are shopping your category. The weekly routine Ten minutes, once a week: 1. Read the new count. Both tabs. A number well above its usual level is worth investigating on its own — something changed. 2. Find the new queries. Sort the SQP Analysis table by week and look at the most recent cohort. Queries with data in the latest week and nothing before it are your new arrivals. 3. Sort them into three piles. | Pile | What it is | Do | |---|---|---| | Relevant, converting | real new demand you are already winning | harvest it | | Relevant, not converting yet | too new to judge | watch a week | | Irrelevant | broad match drifting | negative-match it | 4. Act on the first and third piles. The middle one keeps. Harvesting a new query A new query that converted is the best harvest candidate there is — proven, and probably cheap. Move it to an exact-match keyword, in a campaign where you can see it, and negative-match it in the campaign it came from so you are not bidding against yourself. That last step is the one people skip. See Harvest Winning Search Terms into Exact-Match Campaigns. Do not harvest on one order. A new query with a single sale is a coincidence until it repeats. Reading the count itself The trend in the number tells you something the individual queries do not: | Pattern | Reading | |---|---| | Steady low number each week | normal churn | | Sudden spike | a trend, a season, or your match types widened | | Persistent high number | your broad/auto campaigns are casting very wide | | Near zero, consistently | you have tight targeting — efficient, but you will miss shifts | A spike is not automatically good. If a bid change or a match-type change widened your net, the new queries are your own doing rather than the market's, and many will be irrelevant. Check what you changed before concluding the category moved. Consistently near-zero is a strategic choice, and it has a cost. Exact-match-only accounts are efficient and blind: they cannot discover the phrase that becomes next year's main term. Where else new demand shows up Recommended Actions → Add as keyword flags queries you have impressions and clicks on but no keyword bidding, which overlaps with this population and comes pre-sorted by money. See Recommended Actions: The Queries That Need a Decision. Brand Keywords in AMC shows new branded searches your advertising caused — a different and complementary signal. See Brand Keywords: Proving Upper-Funnel Spend Works. What not to conclude New to you, not new to Amazon. "First seen in the latest week" means first seen in your data. The phrase may be long-established in the category and simply new to your ads. A new query is not free traffic. It arrived through a match type you are paying for. Weekly cohorts. SQP reports whole weeks, so "the latest week" is a complete week, not the last seven days. See SQP Is Weekly: Reading Week Cohorts Correctly. Volume takes time. A new query with a handful of impressions cannot be judged for a fortnight. Related - Understanding Search Query Performance — the dashboard overall. - Harvest Winning Search Terms into Exact-Match Campaigns — what to do with the good ones. - Recommended Actions: The Queries That Need a Decision — the same population, money-sorted. - Search Terms and Aggregated Search Terms — the advertising-side view of the same searches.

Best Seller Rank Tracking

Best Seller Rank is the one number on this screen that has nothing to do with search. It is Amazon's ordering of every product in a category by recent sales velocity — so it moves when you sell, not when you rank. That makes it the closest thing you have to a competitor's sales figure, and the fastest read on whether something you did actually worked. Watch the walkthrough Prefer to watch? Eighty-three seconds on the inverted axis, the arrows, and what BSR is actually for. Where it lives Open a project, open a product, and take the BSR Tracker tab. The BSR Tracker — five tiles and an inverted chart | Tile | Reads | |---|---| | Current BSR | today's rank, and the category it is ranked in | | 24h change | vs yesterday | | 7-day change | vs seven days ago | | Best in range | best rank in the window, with the worst beside it | | All-time best | the high-water mark across all history | Lower is better, and the screen means it BSR is a position, so #641 beats #4,107. Two things on this screen are built around that, and both are easy to misread: The chart's axis is inverted. The y-axis runs 3,800 at the top to 4,400 at the bottom, and the panel says so — "Best Sellers Rank over time per ASIN · lower is better". A line going up is good. It is the opposite of every other chart in TrackIQ, and it is deliberate: it makes improvement look like improvement. The arrows track the outcome, not the number. A green ▲ means the rank improved — the numeric rank went down. A red ▼ means it got worse. So a green ▲ 1,695 on the 30-day column means the product climbed 1,695 places over the month. The category is part of the number Current BSR names the category it is measured in — "in Grocery & Gourmet Food". That matters more than it looks: - #4,107 in a huge category can represent far more sales than #600 in a small one. BSR is a rank within a pool, and the pools are wildly different sizes. - Comparing two products' BSRs is only meaningful in the same category. Across categories it is not a comparison at all. - Amazon can re-categorise a listing, and the rank will step change overnight with no sales change behind it. A sudden jump of thousands of places with a flat sales line is a category move, not a demand move. The tracked products table Every product in the project, with 24h, 7d and 30d movement Below the chart, Tracked Products puts every product in the project side by side with Category, Current BSR and the 24H / 7D / 30D movement columns, so you can see which listing is carrying the line. Read the three windows together rather than one at a time: | Pattern | Reading | |---|---| | 24h down, 7d and 30d up | ordinary daily noise on a rising product | | 24h and 7d down, 30d up | something changed this week — look at price, stock, reviews | | all three down | a sustained decline, not a blip | | 24h up, 7d and 30d down | a bounce inside a downtrend — do not celebrate yet | The 30-day column is the one to trust. BSR is recalculated frequently against a rolling sales window, so a single day says very little. What it is actually good for Reading a launch. A new listing's BSR falls fast and then plateaus. Where it plateaus is your real velocity, and you know it weeks before the sales reports settle. Testing a price change. Price moved on Tuesday; watch the 7-day column on Friday. BSR responds to units, so it answers the volume half of the question before revenue data does. Watching a competitor. A tracked competitor ASIN gives you their BSR too. You cannot see their sales — but a competitor whose BSR is climbing steadily while yours is flat is taking share in your category right now. See Tracking Competitor ASINs Alongside Your Own. Catching a stock-out you did not notice. BSR decays quickly with no sales. A cliff in the chart with no marketing change behind it is very often availability. What not to conclude BSR is not sales. It is a rank derived from sales relative to everyone else. Your BSR can worsen in a week you sold more, because the category grew faster. It is not linear. The gap between #100 and #200 is enormous; the gap between #20,000 and #20,100 is nothing. Treat percentage moves at high ranks with suspicion. Advertising moves it indirectly. Ads sell units, units move BSR. But the same ad spend on a different product moves BSR by a different amount, because the categories differ. Do not use BSR to compare ad efficiency — use ROAS and TACoS. All-time best is context, not a target. It may have been set during Prime Day, a promotion, or a competitor stock-out. Related - Reading Visibility, Average Position and Top 10 — the search-side ranks. - Tracking Competitor ASINs Alongside Your Own — getting their BSR too. - Tracking the Amazon's Choice Badge — the other Amazon-assigned signal. - Inventory — the availability that BSR reacts to first.