Most misleading ad reports come from a handful of predictable errors. Learn why clicks and sessions differ, how Google Ads dates conversions and how to stop double counting.
The most common advertising reporting mistakes are not calculation errors but mismatches: comparing metrics that measure different things, dates that follow different rules, or conversions counted under different settings. Fixing them rarely requires new tools; it requires knowing how each number is defined before putting it in a report.
This guide covers ten common Google Ads reporting errors, explains why each happens and how to fix it, and ends with a checklist to run before any report is shared.
Advertising data passes through several systems, each with its own definitions. An ad platform counts clicks and the conversions it can attribute to them, an analytics tool counts sessions and users, and a CRM or finance system counts leads, deals and revenue. Each is correct by its own rules, and most misreadings happen where two of them meet.
Three habits prevent most of the trouble: define every metric in the report, align dates and settings before comparing anything, and annotate every change that could affect the numbers.
Why it happens. Both numbers seem to describe “visits from ads,” so a gap looks like an error or a sign of fraud.
What is actually going on. Google’s own troubleshooting guide for clicks and sessions starts from the fact that Google Ads tracks clicks while Analytics tracks sessions and users. It lists many reasons the two differ, including:
gclid parameter create sessions without new clicks.How to fix it. Stop expecting the two to match. Track the click-to-session relationship over time and investigate when it changes suddenly or differs sharply for one landing page or device. Our guide to how ad click tracking works explains the path from click to session in detail.
Why it happens. Spreadsheets make it easy to average a column of CPCs, conversion rates or daily return-on-ad-spend figures.
What is actually going on. An average of averages gives a small segment the same weight as a large one. Consider a hypothetical account:
| Campaign | Clicks | Cost | Average CPC |
|---|---|---|---|
| A | 1,000 | $1,000 | $1.00 |
| B | 100 | $300 | $3.00 |
| Simple average of CPCs | $2.00 | ||
| True blended CPC ($1,300 ÷ 1,100 clicks) | about $1.18 |
The simple average overstates the real cost per click by about 70%, because campaign B has a tenth of the traffic.
How to fix it. Recalculate ratios from totals: total cost divided by total clicks, total conversions divided by total clicks. The same rule applies to conversion rates averaged across days. See our guide to measuring cost per click correctly for more on mix effects.
Why it happens. Platform cost, invoiced amounts, agency fees and figures converted for a group-level report all get called “spend.”
What is actually going on. Depending on where and how you are billed, invoices may include taxes or other charges that platform cost columns don’t show, and agency fees sit outside the platform entirely. Accounts in different currencies need a conversion rate, and different reports may use different rates or dates.
How to fix it. State in every report which figure “spend” means: platform cost, invoiced amount, or cost including fees. For multi-currency reports, name the exchange rate source and date, and use the same one throughout.
Why it happens. Each tool has its own time zone setting, and date pickers make it easy to compare a partial week with a full one.
What is actually going on. Google says your reports and statistics follow the account time zone, while items on billing pages are shown only in Pacific Standard Time. It also says time zone changes are no longer supported in standard serving accounts; using a different time zone requires a new account. An analytics property or CRM set to another time zone will therefore assign late-evening clicks and conversions to different days.
How to fix it. Record the time zone of every data source in the report. Compare whole weeks or months rather than single days when sources differ, and check that every comparison uses the same inclusive start and end dates.
Why it happens. Reports pulled on Monday for “last week” look worse than the same report pulled a month later, and people read the first version as final.
What is actually going on. Google explains that Google Ads has reported conversions by the date your ad was clicked in its standard columns: if someone clicks this week and buys next week, the conversion appears on this week’s date. Separate “by conv. time” columns, such as “Conversions (by conv. time),” report conversions at the time they happened instead. Google also notes that conversions can be reported up to 90 days after the click, depending on the conversion window, so recent periods look weaker until late conversions arrive. For complete data, Google recommends a date range that ended at least 30 days ago.
How to fix it. Label which column a report uses. Use click-date columns to judge the traffic you bought, and conversion-time columns when reconciling with sales or finance records, which are dated by transaction. Treat the most recent weeks as provisional and say so in the report.
Why it happens. Month-over-month comparisons are quick to produce, and nobody remembers what changed three weeks ago.
What is actually going on. A new bid strategy, budget, location option or conversion action can move the numbers more than any market trend. Without a note, the comparison silently mixes two different setups.
How to fix it. Keep a dated change log alongside reports and add annotations to charts. Before presenting a trend, check the account’s own record of changes for the same dates. If results changed abruptly, follow a structured Google Ads performance investigation rather than explaining the change from the report alone.
Why it happens. Different conversion actions, tools and reports can each use a different way of giving credit, and the setting is easy to overlook.
What is actually going on. Google’s page on attribution models lists last click and data-driven as the available models; first click, linear, time decay and position-based models are no longer supported. Data-driven is the default for most conversion actions. Google notes that the model affects how conversions are counted in the Conversions and All conversions columns, and that changing it applies going forward. Analytics tools and CRMs may assign credit with their own rules, so the same sale can be credited to different channels in different systems.
How to fix it. Record the attribution model for each conversion action, and annotate the date of any model change. Don’t compare conversion totals across tools without stating each tool’s rule. Google’s Model comparison report shows two models side by side, which helps explain differences. Our guide to ad attribution explains what each approach can and cannot tell you.
Why it happens. Accounts accumulate conversion actions over the years: an old tag, an imported action from another tool, a newer tag added by a different team.
What is actually going on. Google says primary conversion actions are reported in the Conversions column and used for bidding, while secondary actions appear in the All conversions column and are for observation only. If two primary actions track the same form submission, one lead counts twice. Counting settings matter too. Google explains that with the Every and One counting options, “Every” counts every conversion after an ad interaction, while “One” counts only one conversion per ad click, which suits leads. Changes to the count setting only apply to future conversion reporting.
How to fix it. List every primary action and confirm that each represents a distinct business outcome. Demote duplicates and supporting actions to secondary. Use “One” for lead actions unless repeat submissions from the same click genuinely add value. Report from the Conversions column, and treat All conversions as a diagnostic view rather than a total.
Why it happens. Dashboards show percentages with two decimal places whether they rest on ten clicks or ten thousand.
What is actually going on. As an illustration, imagine two ads with 40 clicks each. Ad A has 4 conversions (a 10% conversion rate) and ad B has 2 (5%). Ad A appears twice as effective, yet the difference is two conversions, which chance alone can easily produce.
How to fix it. Show the counts next to every rate. Wait for more data, combine longer periods or broader segments, and avoid declaring winners on small differences. When a decision can’t wait, describe the result as directional.
Why it happens. A click that cost money and produced nothing feels like a loss, and fraud is a vivid explanation.
What is actually going on. Many non-converting clicks come from real people who were researching, comparing or not yet ready to buy. Separately, Google defines invalid traffic as traffic that doesn’t represent genuine interest in your business. It says the Invalid clicks column shows traffic its systems already filtered, that you aren’t charged for invalid traffic detected before billing, and that later detections are credited. Reported clicks have therefore already been through a filtering step, even though no filter is perfect.
How to fix it. Separate three questions in reports: is the traffic genuine, is it relevant, and does it convert at an acceptable cost? Irrelevant but genuine traffic is a targeting problem, covered in our guide to Google Ads budget waste. For how filtering, the reporting columns and credits work, see our explainer on Google Ads invalid clicks, and for how clicks are counted in the first place, our guide to the Google Ads clicks metric.
| Check | Question to answer |
|---|---|
| Definitions | Is every metric defined, including what “spend” and “conversion” mean? |
| Ratios | Are rates and averages recalculated from totals? |
| Dates | Same time zone, same inclusive dates, whole periods? |
| Conversion dating | Click-date or conversion-time columns, and is that stated? |
| Freshness | Are recent periods marked as provisional? |
| Attribution | Is the model known and unchanged across the compared periods? |
| Conversion actions | One primary action per real outcome, with suitable counting? |
| Sample size | Are counts shown next to rates? |
| Changes | Are setting changes in the period annotated? |
| Sources | Does each figure name the system it came from? |
Keep reading
Measurement
Attribution divides credit for a conversion among ad interactions, but it does not prove the ads caused the sale. This guide covers what current models do and how to read them.
Measurement
Follow one ad click from the auction to your analytics report, see where identifiers are added or lost along the way, and use a checklist to keep tracking healthy.
Click fraud
When results change suddenly, the order of your checks matters. This diagnostic sequence rules out tracking, account changes and market shifts before blaming fraud.