Google Ads Reporting: Which Numbers, How Often, for Whom

A store’s Google Ads report should answer three questions at three speeds. Every week: did something break in tracking, the feed or the budget, and what changed? Every month: why did revenue move, through spend or through ROAS, and did conversion rate or average order value change? Every 90 days: what state is the account in? The owner needs the monthly and 90-day answers, each ending in a decision. The in-house marketer runs the weekly checks. An agency owes all three, plus a log of what it changed.

What is a Google Ads report for in an online store?

A report earns its place when it ends in a decision. That is how store owners describe the job in our client conversations: when the numbers arrive, they want to know if all is well or what needs to change. They want the answer in their own language, with a recommendation. Three reports at three speeds give that answer better than one table of 30 columns, because each speed catches a different kind of problem.

Our guide to ecommerce KPIs covers the few numbers that explain a store’s revenue. This guide shows how to arrange them so that the right person reads each one at the right speed.

Which numbers should the report show?

Seven, and simple arithmetic ties them together. Revenue from ads equals spend multiplied by ROAS. ROAS, in turn, equals conversion rate multiplied by average order value and divided by cost per click. A report that shows all seven lines can say not only that revenue moved but also why.

Google defines the conversion columns in Understand your conversion tracking data. “Conversions” counts your primary conversion actions. “Conv. value / cost” divides the total conversion value by cost, and “Value / conv.” divides it by the number of conversions. For Cost and Avg. CPC, see About columns in your statistics table.

Two details in the conversion columns matter for a store. The “Conversions” column may include modelled conversions in cases where not all conversions can be observed. And “All conversions” adds secondary actions and view-through conversions, so it counts more than just your orders.

How to read a Google Ads report explains which columns come first and which ones mislead.

ROAS moves with average order value and conversion rate, not with cost per click

We checked which of the three parts of ROAS (conversion rate, average order value and cost per click) changes together with it. In GetProfit data, 1,360 store-months from June 2025 to June 2026, we compared each store’s monthly change with the median store in the same month. That removes whatever happened to every store at once, such as the season.

Within a store, ROAS went together with average order value (rank correlation +0.540) and with conversion rate (+0.527), but only weakly with cost per click (−0.100). Take the 361 store-months when ROAS rose more than 20% relative to other stores. The median change there, also measured against other stores, was conversion rate up 23.3% and average order value up 19.5%. Cost per click fell by a median of only 2.4%, and spend did not change.

For the report, this means conversion rate and average order value belong on the same page as ROAS. If ROAS drops while conversion rate holds and average order value falls, ask which products sold, not how to change bids. Two limits apply: these are observations, not an experiment, and our ROAS is revenue divided by spend. We don’t have purchase prices, so these numbers leave margin out.

How much of a store’s report is Performance Max?

In the median store, almost all of it. Performance Max runs in 141 of 146 stores in our data and takes a median 95.7% of their ad budget (GetProfit data, June 2025 – June 2026). For a store like that, the account total and the Performance Max total are nearly the same line.

That changes where you look for detail. Performance Max splits products, channels and search terms into separate reports, and Performance Max reporting covers the question each one answers. Google’s own FAQ, Answering Your Top Questions About Performance Max, advises you to start from campaign-level results against your goals. The campaign optimises for marginal return across channels, so the average ROAS of one channel can mislead: use channel and product views to diagnose, not to grade.

Watch for one break in the series. According to About Product Reporting, product reporting expanded in June 2026 to all Performance Max networks, with a possible one-time increase in impressions, clicks and other metrics. The developer guide Retail Campaign Performance dates the change to 15 June 2026. So a product report that compares May 2026 with July 2026 compares two different scopes, and part of any growth may be the new coverage.

The products report in Google Ads shows how to build the product view and read it without cutting your long tail.

What should you check every week?

What broke and what changed. Leave performance verdicts for the monthly report. A week is too short to judge ROAS, but long enough for a broken tag or a lost feed to cost real money. Five checks:

1. Conversions are still arriving. A day with clicks and no orders is a tracking question first, and only then a demand question.
2. Average order value looks real. Google’s “Value / conv.” should sit close to the average order value your store’s own system shows. If every order carries the same value, a default has replaced the real amount.
3. Spend matches the plan. Look for campaigns that run out of budget early, or spend far below it.
4. Products are still in ads. Google’s product reporting help suggests checking the Status column in the Products tab: if top products move to “Not eligible”, the issues column shows why.
5. You know what changed. According to Review your account history, change history covers the last 2 years, and you can filter it by user and by tool. It includes changes from automated rules, the Google Ads API and Google Ads Editor.

Read the last few days with care. Google’s primary conversion columns put each conversion on the day of the click. So an order placed this week after a click last week lands in last week, and last week’s ROAS can still grow.

About data freshness adds two more delays. Conversions attributed with a model other than last click arrive later: in Google’s table, Sunday’s data is ready on Monday or Tuesday, depending on the time zone. And reported metrics can change days later, after Google removes invalid traffic or late conversions arrive.

Let a weak week pass without a change. Google’s conversion data help warns that frequent manual changes to budgets, targets or conversion goals reset the standard 7–14 day learning period.

A weekly Google Ads review gives the full routine and what to leave alone. Google Ads anomaly detection shows how to set alerts that catch real breaks and stay quiet through normal swings.

How do you tell noise from a real change?

Compare ROAS with your own history, not only with last month. We looked at 110 stores in GetProfit data with at least eight months of data between June 2025 and June 2026. In the median store, a typical month’s ROAS sat 16.0% away from the store’s own median ROAS.

A quarter of stores swung less than 11.0%, and a quarter more than 24.1%. The best month’s ROAS was a median 3.1× the worst. This is raw ROAS, with the season included.

So a month somewhat below your median is ordinary for most stores. A move well beyond your typical swing is an event worth explaining, and so is one that lines up with a change, a stock-out or a tracking edit. The portal’s trend alarm uses the same logic. It treats ROAS falling by 15% or revenue falling by 10% in two consecutive months as a negative trend, rather than reacting to a single month.

How much ROAS swing is normal covers when a drop justifies touching bids or budgets. Comparing periods in Google Ads shows how to keep partial months, spending gaps, broken data and the season from producing a fake rise or fall.

What should the monthly report explain?

Why revenue moved, in two steps. First, did it move through spend, through ROAS or through both? Second, if ROAS moved, was it conversion rate, average order value or cost per click?

In GetProfit data, June 2025 – June 2026, we looked at 484 store-months where revenue grew more than 20% relative to the median store that month. Measured the same way, against other stores, spend had risen by more than 20% in 61% of them, and ROAS by more than 20% in 55%. Both had risen in 23%, and neither in only 7%. So each route matters, and the report needs spend next to ROAS: the spend route was there in most of these months.

Example store, not client data.

A tableware store with 3,000 products compares this month with last month:

Revenue fell 4.7% although spend rose 10%. Clicks grew with spend, and conversion rate and cost per click held. The whole ROAS drop comes from average order value, which fell from 600 to 520. So the monthly report asks which products sold this month, not how to change bids, and gives the owner one line of explanation and one decision.

A monthly Google Ads report template gives the sections one by one, with the questions an owner should ask.

Which campaigns belong outside the store’s ROAS?

Awareness campaigns: Video, Display and Demand Gen. Their job is to bring people in rather than to close orders on the last click. Few stores run them, and those that do spend little on them (GetProfit data, 146 stores, June 2025 – June 2026):

Google’s own columns treat them differently too. According to View performance across campaign types, the “Interactions” column counts clicks for text ads but video views for TrueView video ads. View-through conversions appear only in the “View-through conversions” and “All conversions” columns, not in “Conversions”.

The portal judges Search, Shopping and Performance Max strictly on purchases. It lists Video, Display and Demand Gen separately and doesn’t penalise them for having none: their spend counts in the account’s totals, but not in its performance part. A report can do the same with two blocks: product campaigns with revenue and ROAS, and awareness campaigns with spend and the job you gave each one.

Blended ROAS in Google Ads asks whether one number for everything belongs in the report at all.

When is the ROAS in a report not worth reading?

When the conversion value can’t be trusted. If every order carries one fixed number instead of its real amount, revenue in the report is just the number of orders multiplied by that number. ROAS computed on it means nothing. Google’s conversion data help gives the first check: if values are missing or default to $1.00, make sure “Use different values for each conversion” is selected in the conversion action’s settings.

The portal follows a fixed rule here. As long as it treats the order amount as real, it shows revenue and ROAS. Where the amount can’t be trusted, it hides ROAS, shows revenue in grey with a “not trusted” mark and counts by the number of orders and the cost per order. A report can follow the same rule: switch the headline to orders and cost per order until you fix the value.

Before you read any ROAS, and before you change bids or budgets, run a conversion tracking audit.

What does the 90-day review look at?

The state of the account, not the month’s result. Ninety days is long enough to show what a month hides:

• Do the campaigns collect enough conversions to learn on, or is the budget split too thin?
• Which products sell, and which spend without selling?
• Which changes worked, and which ones hurt?
• Can the conversion data be trusted?

Our Google Ads audit guide gives these checks step by step.

The portal’s store score puts the answer on one screen: a score out of 100 and a letter from A to F. It always uses a fixed 90-day window, whatever period you pick on screen. Four areas carry the points: products 40, campaign structure 30, the data Google receives 17 and the history of changes 13.

If the conversions can’t be trusted, the portal lowers the score and says why. The score section of the portal explains how the score is built.

The portal’s change log answers the agency question: “what did we change, and how did it end?” It gives a first conclusion a week after a change and a confident one after two weeks. Once a change is more than a month old, the portal stops judging it: too many other causes pile up.

How to judge a single change yourself, and when you need an experiment instead: did that Google Ads change work. A change log for Google Ads covers what to record next to the numbers that Google’s history doesn’t hold, such as stock-outs, price changes and product data edits. Google Ads benchmarks for online stores shows how to compare your store with others and which comparisons mislead.

Which view do the owner, the marketer and the agency need?

The same data, cut three ways.

The owner needs one page with the decision at the top. The reason sits underneath, in four lines: spend, ROAS, conversion rate and average order value. Everything else can live in the marketer’s view.

The in-house marketer works from saved views and a dashboard. A Google Ads dashboard in Looker Studio covers what belongs on it and how to work around the connector’s blind spots.

The agency usually works across many accounts. According to About Google Ads manager accounts, a manager account lets you compare performance across all linked accounts and run reports for several accounts at once. In the Report Editor, a manager account can put several accounts in one report, but some data appears greyed out on screen until you download the report. Reporting across many client stores covers the point where manager account reports and Looker Studio stop being enough.

An agency report also owes the owner the list of changes. Google’s change history records who made each change and with which tool, so the owner can check that part.

How do you set up the reports in Google Ads?

Google’s help covers the mechanics. Here is the order that works for a store:

1. Build the view in a statistics table. According to Create, save, and schedule reports from your statistics tables, you set the date range, columns, filters and segments, and segments show up as rows in the report.
2. Add what’s missing as custom columns. About custom columns describes metrics built from existing ones and several date ranges side by side in one table. Custom columns for online stores lists the formulas that make loss-making products stand out.
3. Build charts in the Report Editor. Create custom reports in Report Editor gives the path: Campaigns, then Insights & reports, then Report editor. You can build a table or a line, column, bar, scatter or pie chart, and add conditional formatting to tables.
4. Schedule it. From the download icon above the table, choose Schedule, the frequency (daily or weekly, for example) and the format. Reports go to people who have access to the account. Reports for individual accounts start running at 1 A.M. in the account’s time zone.
5. Keep it in use. If you don’t open or download a saved report for more than 18 months, directly or through a dashboard, Google removes it from the account automatically.

Google Ads reporting tools compares spreadsheets, Looker Studio and dedicated tools by what each costs in hours and in blind spots.

What to set up this week

1. Write down the three questions and who answers each. What broke this week, why revenue moved this month, what state the account is in after 90 days. A report without an owner for each question turns into a table nobody reads.
2. Check the value line first. Put “Value / conv.” for your purchase action next to the average order value your store’s own system shows. If they don’t match, fix tracking before you read any ROAS.
3. Add conversion rate and average order value next to ROAS. With them, the monthly report can say not only that ROAS moved but also why.
4. Split awareness campaigns out. Report Video, Display and Demand Gen in their own block, with their spend and the job you gave them.
5. Measure your own typical swing. Take your monthly ROAS for the last 8–12 months, find the median and how far a typical month sits from it. Treat months well beyond that as events to explain.
6. Mark 15 June 2026 in every product-level comparison. Product reports for Performance Max before and after that date cover different networks.
7. Keep a change log next to the numbers. Google’s change history lists changes made in the Google Ads account. Stock-outs, price changes and edits to your product data happen elsewhere and need your own notes.
8. Schedule the weekly view; have a person write the monthly one. The weekly checks can arrive by email. The monthly report needs someone to explain why the numbers moved.

What state your ad account is in. One score instead of a dozen tabs of figures — plus a breakdown of exactly where the money leaks. The portal changes nothing without your consent.

Get your account score →

Sources

• Understand your conversion tracking data — definitions of “Conversions”, “Conv. rate”, “Conv. value / cost” and “Value / conv.”; modelled conversions in “Conversions”; secondary and view-through conversions only in “All conversions”; primary conversion columns counted by the time of the click; values missing or defaulting to $1.00 and the “Use different values for each conversion” setting; frequent changes reset the 7–14 day learning period. Checked 2 October 2026.
• About columns in your statistics table — definitions of “Avg. CPC” and “Cost”. Checked 2 October 2026.
• Answering Your Top Questions About Performance Max — start from campaign-level performance; Performance Max optimises for marginal return; average ROAS of a single channel can mislead. Checked 2 October 2026.
• About Product Reporting — product reporting expanded in June 2026 to all Performance Max networks; one-time increase in Performance Max metrics; the Products tab and its Status column. Checked 2 October 2026.
• Retail Campaign Performance — from 15 June 2026 the Shopping Performance View includes data from all Performance Max networks, with a possible one-time increase in reported metrics. Checked 2 October 2026.
• Review your account history — change history for the last 2 years; filters by user and tool; changes from automated rules, the API and Google Ads Editor. Checked 2 October 2026.
• About data freshness — timing of conversions attributed with models other than last click; metrics updated days after the event for invalid traffic and late conversions. Checked 2 October 2026.
• View performance across campaign types — what the “Interactions” column counts for each ad format. Checked 2 October 2026.
• About Google Ads manager accounts — comparing performance and running reports across several accounts. Checked 2 October 2026.
• Create, save, and schedule reports from your statistics tables — building a report from a statistics table; scheduling, frequency and formats; recipients with account access; 1 A.M. start time; removal of saved reports unopened for 18 months. Checked 2 October 2026.
• About custom columns — custom metrics from existing ones; several date ranges side by side. Checked 2 October 2026.
• Create custom reports in Report Editor — the path to the Report Editor; chart types; conditional formatting; several accounts in one report from a manager account. Checked 2 October 2026.
• GetProfit data, 1,360 store-months, June 2025 – June 2026 — how ROAS moves with conversion rate, average order value and cost per click; the 361 months when ROAS rose.
• GetProfit data, 484 store-months, June 2025 – June 2026 — whether revenue growth came through spend, ROAS or both.
• GetProfit data, 110 stores, June 2025 – June 2026 — how far monthly ROAS strays from a store’s own median; best month to worst.
• GetProfit data, 146 stores, June 2025 – June 2026 — Performance Max and awareness campaigns: number of stores and median share of budget.
• GetProfit portal methodology — purchases-only judgement for Search, Shopping and Performance Max; awareness campaigns listed separately; no ROAS when the order amount can’t be trusted; the negative trend rule; the store score’s 90-day window and area weights; the change log’s 7, 14 and 30-day windows.