Zurück zu den Erkenntnissen

E-Commerce-Analytics Aktualisiert 2026-11-15 10 Min. Lesezeit

Amazon Brand Analytics and ad software: build the evidence room before budget moves

A practical Advertentie Software guide for brand owners using Amazon Brand Analytics, Search Query Performance and self-service ad rules to decide which queries deserve budget, fixes or caps.

Von Lisa van Broekhoven Dashboards, Reporting und Marketplace Intelligence für datengetriebenen Commerce.

E-Commerce-Analytics-Zusammenfassung

Kurzantwort

Eine praktische FiveX-Perspektive auf E-Commerce-Analytics für Marketplace-Seller, E-Commerce-Marken und Agenturen. Ziel ist es, Marketplace-Teams dabei zu helfen, fragmentierte Signale in klarere Entscheidungen zu Wachstum, Profitabilität und Operations zu übersetzen.

Definition

Was dieser Artikel abdeckt

E-Commerce-Analytics behandelt Entscheidungen, Daten und operative Routinen, mit denen Marketplace-Teams profitables Wachstum verbessern.

Amazon Sponsored Products Buy Box ROAS Deckungsbeitrag Repricing Marketplace-Seller E-Commerce-Marken Bestandsmanagement Marketplace-Gebühren

Amazon Brand Analytics is dangerous in a very specific way: it makes missed demand visible.

Open Search Query Performance or Top Search Terms and you can suddenly see shoppers searching, clicking, adding to basket and buying around your category. You see terms where your brand barely appears. You see queries where competitors win the click. You see products that get added to cart but do not turn into purchases. For a brand owner managing Amazon ads in-house, that feels like a clear instruction: spend more.

Not so fast.

The named mistake I see is turning Brand Analytics into a keyword export. A team downloads search queries, sorts by volume, pushes the biggest gaps into Sponsored Products, and asks the bid automation to “capture share”. The data was useful. The decision was incomplete. Amazon Brand Analytics can show where the shopper journey leaks. It cannot tell you whether the next euro of ad spend has enough contribution margin, stock cover, price strength or campaign permission to chase that leak.

My stance: self-service brands spending from roughly €1.5K per month on marketplace ads need an ad evidence room around Amazon Brand Analytics. Not a prettier dashboard. A daily decision layer that turns Search Query Performance, Top Search Terms and Search Catalog Performance into five actions: scale, fix, learn, cap or quarantine.

That distinction matters because Brand Analytics is not Amazon Ads reporting. It is search-funnel evidence. Advertising software should use it to decide where spend is allowed to move, where listing work should happen first, and where apparent opportunity is actually a profit trap.

What the existing advice gets right

The public guides are genuinely helpful. Amazon explains Brand Analytics as a set of dashboards for registered brands: Search Query Performance, Search Catalog Performance, Top Search Terms, Repeat Purchase Behavior, Market Basket Analysis and other shopper-behavior views. That is the right starting point. Brand owners need to understand how customers search and purchase, not only how campaigns spend.

Helium 10 does a strong job explaining Search Query Performance at the query and ASIN level. Their framing is practical: look at impressions, clicks, cart adds, purchase rates and share metrics to see whether your product wins attention and converts against the market. Perpetua goes deeper on using SQP as a baseline before advertising changes, then comparing performance after A/B tests, deals, listing optimization or campaign moves. That is exactly how operators should think: evidence before and after the intervention.

Ad Badger usefully separates a query from a keyword. The shopper types the query. The advertiser bids on the keyword. They overlap, but they are not the same thing. m19, BidX, SellerApp and SalesDuo all explain the core metrics: search frequency, click share, conversion share, impressions, cart adds and purchases. Reddit threads add the honest operator frustration: SQP numbers often do not match PPC reports, and many sellers do not know what to do with the data after the first download.

The gap is consistent: most advice explains what the reports show, then jumps to keyword expansion or listing optimization. The missing step is permission. Should this query receive ad budget today? If yes, how much? If no, what needs to change first?

Why Brand Analytics and PPC reports will not match perfectly

Before you build rules, accept one reality: Amazon Brand Analytics and Amazon Ads reports answer different questions.

Brand Analytics is shopper-search evidence. It looks at what happened in the search funnel: impressions, clicks, cart adds, purchases and your share of those moments for a query. Amazon Ads reporting is advertising-attribution evidence. It looks at what your campaigns spent and which sales Amazon attributes to those ads inside the reporting window.

That is why a Reddit seller can see one purchase in Search Query Performance for “red t-shirt” while an exact-match PPC campaign reports seven attributed purchases. The reports are not necessarily “wrong”. They are measuring different systems with different definitions, time windows and attribution rules.

The operator mistake is trying to force them to reconcile row by row. The better move is to give each report a job:

  • Brand Analytics: where are we gaining or losing shopper momentum in search?
  • Amazon Ads: what did we pay for, what did it return, and which targets deserve bid changes?
  • FiveX or your operating layer: is the SKU allowed to receive more spend after margin, stock, price, returns and campaign role are checked?

Once those jobs are clear, disagreement becomes useful. If SQP shows high market demand but PPC has weak sales, you may have a targeting, bid or campaign-structure issue. If PPC looks efficient but SQP shows tiny query share, the campaign may be harvesting easy demand without building category presence. If both look good but SKU margin is thin, the query still may not deserve scale.

The ad evidence room: four checks before budget moves

An ad evidence room is a simple table, refreshed daily or weekly, where every meaningful query is scored before advertising software acts. It does not need to be complicated. It needs to stop Brand Analytics from becoming an uncontrolled spend trigger.

CheckQuestionTypical action
Search demandIs the query large enough and relevant enough to matter?Ignore, monitor or add to learning queue
Click permissionDo shoppers choose our offer when they see it?Scale, improve image/title/price, or cap bids
Conversion permissionDo clicks become cart adds and purchases?Fix listing, reviews, delivery or promotion before scaling
Profit permissionCan the SKU afford the traffic after fees, returns and stock constraints?Raise bids, hold, pause or route to manual review

This is where FiveX fits naturally. FiveX connects marketplace analytics, advertising performance, SKU profitability and inventory signals in one cockpit. Brand Analytics can tell you that a query is attractive. FiveX helps decide whether the SKU has enough contribution margin, stock runway and ad-performance evidence to deserve spend.

Give every query a role before it enters automation

Search Query Performance becomes much easier to use once queries stop being treated as one big list. Assign one of five roles:

  • Defend: branded and close product-line terms where losing the click would damage profitable demand.
  • Win: non-brand terms with strong click and purchase evidence, enough margin and enough stock.
  • Fix before spend: terms with demand but weak click share or conversion share. The listing or offer needs work before bids rise.
  • Learn: relevant but uncertain terms that deserve a small controlled test.
  • Quarantine: queries that look attractive but fail margin, stock, return-rate or strategic-fit checks.

The named trade-off is simple: the highest-volume query is often not the next best query. The next best query is the one where demand, conversion and profit permission overlap.

Scenario 1: high demand, weak click permission

Imagine NorthSea Nutrition, a supplement brand selling electrolyte powder on Amazon Germany. Search Query Performance shows the query “electrolyte powder sugar free” had 42,000 impressions last week. The brand’s ASIN appeared often enough to matter, but its click share was only 4.8%. Purchase share was 1.1%.

The PPC reflex is to bid harder. The evidence-room answer is more cautious.

The product sells for €18.90. After referral fee, fulfilment, landed cost and average promo cost, contribution margin is €6.20. Break-even ACOS is about 32.8%. The current Sponsored Products exact campaign spends at 41% ACOS on this query. Stock cover is 38 days, so inventory is not the issue. The click problem is likely offer quality: the hero image looks clinical, the title hides “sugar free” near the end, and the leading competitor shows a 4.6-star rating with 2,800 reviews while NorthSea has 420 reviews.

Action: label the query fix before spend. Keep a small exact campaign live with a bid cap, but do not scale. First rewrite the title, improve the first image, add “sugar free” to the visible benefit stack and test a 5% coupon for two weeks. In FiveX, the query sits in the exception queue with a release condition: click share above 7% and ad ACOS below 33% for seven days. Only then can automation raise the bid.

That is a better decision than “high volume equals high budget”.

Scenario 2: cart adds are strong, purchases are weak

Now take Oslo Baby Co., a brand selling a Wi-Fi baby monitor. The query “baby monitor with app” shows 18,500 impressions in SQP. Click share is acceptable at 6.2%. Cart-add share is surprisingly strong at 3.4%. Purchase share drops to 0.7%.

This is not a keyword problem. Shoppers are interested enough to click and add to basket. Something breaks near purchase.

The SKU has €21 contribution margin on a €129 price point, but returns are running at 14% because setup instructions are unclear. FBA stock cover is only 11 days. The current PPC campaign has a tidy 24% ACOS, which looks scaleable if you only read Campaign Manager. But scaling would accelerate a product with return pressure and low stock cover.

Action: label the query cap, not win. FiveX would connect ad performance to inventory and returns, so the software can hold daily budget at €35 until stock cover is above 25 days and the return-rate issue is fixed. The practical work is not a bid increase. It is improving setup content, adding a comparison image, tightening the FAQ and waiting for the next inbound shipment. Once purchase share improves and stock recovers, the query can move from cap to win.

Scenario 3: branded defence is not always incremental

Brand Analytics is also useful for branded search, but the trap is different. Consider Linden Home, which sells air purifier replacement filters. The branded query “linden filter pack” gets 9,800 searches per month. The brand owns 64% click share and 58% purchase share. PPC spends €420 per month defending the term at 9% ACOS.

That looks excellent. But the evidence room asks a sharper question: what would happen if spend was reduced?

Competitor presence is light. Organic rank is position one. The product has 4.7 stars and strong Subscribe & Save uptake. Instead of letting automation keep increasing bids because ACOS is low, the team runs a holdout: reduce branded exact bids by 35% for 14 days while monitoring SQP click share, organic sessions and total orders. If purchase share stays above 55% and total orders remain flat, some of that defensive spend can move to a non-brand learn queue.

This is the part many tools miss. Low ACOS does not automatically mean “more budget”. Sometimes it means “test whether we are paying for demand we already own”.

How to operationalize this in self-service ad software

A practical weekly workflow looks like this:

  1. Monday: pull Brand Analytics query movements: rising impressions, falling click share, falling purchase share, new competitor terms and branded changes.
  2. Tuesday: match those queries to campaign targets and SKU-level economics: contribution margin, break-even ACOS, stock cover, return rate and price position.
  3. Wednesday: assign roles: defend, win, fix before spend, learn or quarantine.
  4. Thursday: let advertising software apply safe changes: bid increases only on win queries, small test budgets on learn queries, caps on low-stock or weak-margin queries, negatives for quarantine.
  5. Friday: review exceptions where Brand Analytics and PPC disagree. These are usually the most valuable operator conversations.

FiveX product hook number three is the approval queue. Instead of automation silently changing every bid, FiveX can surface recommendations with the commercial reason attached: “raise bid because purchase share improved and margin permission is green”, “cap because stock cover is 9 days”, or “hold because click share is weak despite high impressions”. That gives brand owners control without dragging them back into spreadsheet babysitting.

The final checklist

Before a Brand Analytics query receives more ad spend, ask six questions:

  • Is the query relevant to the product, or only adjacent volume?
  • Do we have click permission: image, title, price and reviews strong enough to win the click?
  • Do we have conversion permission: cart adds and purchases moving in the right direction?
  • Do we have profit permission: contribution margin and break-even ACOS strong enough for the test?
  • Do we have stock permission: enough runway to support the demand we might create?
  • Do we know the campaign role: defend, win, fix, learn or quarantine?

If the answer is unclear, do not scale yet. Put the query in the evidence room, assign an owner and define the release condition.

Amazon Brand Analytics is one of the best datasets available to brand owners. But its value is not in the export. Its value is in the operating discipline that follows. The winners will not be the teams that download the most queries. They will be the teams that connect search evidence to profit, stock and advertising automation before the next euro moves.

Operative Perspektive

So nutzen Sie diese Erkenntnis

Reine Kennzahlen-Sicht

Betrachtet Umsatz, Klicks, ROAS oder Bestellungen als getrennte Signale. Das ist schnell, kann aber Marketplace-Gebühren, Retouren, Bestandsdruck und Margenverluste verdecken.

Marketplace-Intelligence-Sicht

Verbindet Kanalperformance mit Deckungsbeitrag, Pricing, Advertising, Bestand und Operations, damit die nächste Aktion kaufmännisch klar ist.

FAQ

Fragen, die Marketplace-Teams zu diesem Thema stellen

Was ist die wichtigste Kennzahl für E-Commerce-Analytics?

Beginnen Sie mit dem Deckungsbeitrag und interpretieren Sie danach Kanalmetriken wie Umsatz, ROAS, Conversion und Bestandsreichweite in diesem Profit-Kontext.

Wie können Marketplace-Teams E-Commerce-Analytics nutzen, ohne mehr manuelle Arbeit zu erzeugen?

Nutzen Sie verbundene Marketplace-Daten, wiederholbare Dashboards und klare operative Regeln, damit Teams Ausnahmen prüfen statt Tabellen neu aufzubauen.

Wo passt FiveX in diesen Workflow?

FiveX bringt Marketplace Analytics, Advertising, Repricing, Bestand, Integrationen und Exporte in ein Cockpit für Seller, Marken und Agenturen.

Brauchen Sie zuerst einen trader‑geführt Walkthrough, or einen rollout‑tauglichen Finanz‑Plan?

Schicken Sie Ihr Marktplatzportfolio, wir zeigen Connector‑Deckung Repricing‑Einstieg Advertising‑Schicht sowie Exportpipelines für einen schnellen Optimisationszyklus.