Paid search analytics: the reports that answer real questions, and the settings that quietly change the numbers
How to measure a Google Ads search account properly: conversion setup first, then the search terms, auction insights and attribution reports, read with the settings that shape every figure in them.
Published by AI Knows Us · Updated 21 October 2026
Paid search analytics is the work of measuring what your search ads actually produce and why. It rests on one foundation and three reports. The foundation is conversion measurement: until purchases, leads or calls are recorded as conversions, every other number is just traffic. The three reports are the search terms report (what people really typed), auction insights (who you compete with and how often you show), and attribution (which clicks get credit). Read them alongside the settings that shape them, especially the conversion window and attribution model, because changing either one changes your results without a single extra sale.
Start with conversions, or nothing else adds up
Google's help page on conversion measurement says you choose what counts as valuable, such as a purchase, sign-up or phone call, and Google Ads then reports and optimises against it. It lists what that makes possible: knowing which keywords, ads, ad groups and campaigns drive valuable activity, understanding return on investment, and using Smart Bidding strategies such as Maximize Conversions, target CPA and target ROAS.
Decide your conversion actions on paper before you touch the account:
- Primary actions are the outcomes you pay for: an order, a qualified lead form, a booked call.
- Secondary signals are useful but not the goal: newsletter sign-ups, brochure downloads, time on site. Track them, but do not let bidding treat them as sales.
Importing from Google Analytics
Many accounts import conversions from Google Analytics instead of using a separate Google Ads tag. The help page sets three requirements: Admin access in Google Ads and Edit access in the linked Analytics property, the event marked as a key event in Analytics, and the right property actively linked. It warns that after linking or marking a key event, it can take 24 to 48 hours before the import option appears.
Give Smart Bidding time
The same page says Smart Bidding strategies need a standard 7 to 14 day learning phase, and advises against frequent manual changes to budgets, targets or conversion goals in that period. In analytics terms, do not judge a bidding change on its first week.
The settings that change every number
The conversion window
Google defines the conversion window as the period after an ad interaction during which a conversion is recorded, with a default of 30 days. You can change it per conversion action, and Google says changes apply only to conversions going forward, with no retroactive counting. A shorter window will show fewer conversions for exactly the same sales. To choose, Google points to the time lag report in attribution reports, which shows how long people take between ad interaction and conversion.
The attribution model
Google's attribution help page says the first click, linear, time decay and position-based models are no longer supported, and that conversion actions which used them were upgraded to data-driven attribution. Two models remain:
| Model | How credit is given | What it means for your reports |
|---|---|---|
| Last click | All credit to the last clicked ad and keyword | Brand and late-stage keywords look strongest |
| Data-driven | Credit spread across interactions based on your account's past data | Earlier, generic searches receive a share of the credit |
Google's restaurant example makes the point: with four searches before a booking, last click gives 100% of the credit to the final branded search, while data-driven gives each keyword part of it. Use the "Model comparison" report to see how a switch would move your numbers before you make it, and never compare a period on one model with a period on another.
Report 1: search terms, where the waste and the ideas are
Google explains the difference simply: a keyword is what you add to your account, and a search term is what a person actually typed. The search terms report shows the real searches that triggered your ads and how each performed. Use it to:
- Find searches that spend money and never convert, and add them as negative keywords.
- Find converting searches you do not yet target as keywords, and add them.
- Turn on the "Keyword" and "Match type" columns to see which of your keywords matched each search, which shows whether broad match is reaching sensible queries.
- Pull ideas for ad copy and landing pages in the customer's own words.
One limit to plan around: Google says some search terms without enough query activity are left out of the report for privacy. The separate search terms insights view groups searches into themes and includes those low-volume queries, so the two views will not match exactly.
Report 2: auction insights, who you are up against
Auction insights compares your performance with other advertisers in the same auctions. For Search campaigns, Google lists six statistics: impression share, overlap rate, outranking share, position above rate, top of page rate and absolute top of page rate. Shopping campaigns get three of them.
The most useful one is impression share: impressions received divided by the estimated impressions you were eligible for. Low impression share with spare budget usually points to ad rank; low impression share with a capped budget points to money. Two notes from Google's page: the report shows no insights when impression share is below 10%, and a competitor's impression share in your report may differ from what they see in their own account, because your eligible auctions only partly overlap.
Report 3: attribution and time lag
Attribution reports show the paths to conversion. Alongside the model comparison, the time lag view tells you how long buyers take. Google notes it reflects ad interactions on google.com only, not the Display Network or search partners. For a business with a long buying cycle, this is often where a "failing" generic campaign turns out to be starting journeys that brand searches finish.
The metrics worth a place on the dashboard
| Metric | Question it answers |
|---|---|
| Conversions and conversion value | What did we get? |
| Cost per conversion (CPA) | What did each result cost? |
| Return on ad spend (ROAS) | How much revenue per unit of spend? |
| Conversion rate by keyword and landing page | Where does traffic turn into results? |
| Search impression share | How much of the available demand do we reach? |
| Spend on non-converting search terms | How much is wasted? |
Clicks and click-through rate belong on a diagnostic view, not on the summary, because they can rise while results fall.
Paid search is also only half the search picture. If organic results and ads are competing for the same queries, the numbers are best read together; our guide to running organic and paid search as one programme covers how.
Questions about paid search analytics
What is the most important paid search metric?
Cost per conversion or return on ad spend, measured against conversion actions that reflect real business outcomes. Without reliable conversion tracking, no other metric tells you whether the ads pay.
Why do Google Ads and Google Analytics show different conversion numbers?
They can use different attribution, different conversion windows and different counting rules, and Google notes that conversions can arrive with a lag. Agree which source is the record for each decision and compare like with like.
How often should I review the search terms report?
Weekly for active accounts with meaningful spend, and after any change to match types or keywords. Smaller accounts can review monthly.
Which attribution model should I use?
Google Ads now offers last click and data-driven. Use the model comparison report to see how each would credit your campaigns, choose one, and keep it consistent when comparing periods.
What does a low impression share mean?
Your ads showed for a small part of the searches you were eligible for. The cause is usually budget, ad rank, or both. Check whether the campaign was limited by budget before raising bids.
The search that never reaches the ads report
Some buyers now start their research in ChatGPT, Gemini or Perplexity before they type anything into Google. A buyer who asks an assistant "which supplier should I use" and gets a shortlist may only search for the brands on that list, which shows up in your account as brand searches with no visible cause. Paid search analytics cannot see that step. AI Knows Us asks assistants the questions your buyers ask and shows whether your company is on the shortlist they return.