Running ads
Why do my two ad platforms disagree?
The short answer
Usually because the two platforms are not measuring the same thing, and sometimes because they report the same thing in different units. Google divides conversions by ad interactions; Microsoft divides by clicks. Google states return on ad spend as a percentage and Microsoft as a ratio, so 500 percent and 5 mean the same result.
Two reports, two numbers, one month. Somebody is going to suggest that one platform is lying or that the tracking is broken. Usually neither is true.
The columns share names because the industry shares vocabulary. They do not share definitions. In some cases the two platforms calculate genuinely different things and call them the same word. In at least one case they calculate the same thing and print it in units a hundred times apart. Both are documented, in public, by the platforms themselves.
Once you know which kind of disagreement you are looking at, the comparison becomes possible. Until then, every conversation about it goes in circles.
Conversion rate is not the same calculation on both platforms
Start with the one that catches almost everybody, because the column header is identical.
Google defines conversion rate as the average number of conversions per ad interaction, shown as a percentage, with the worked example of 50 conversions from 1,000 interactions giving 5 percent.
Microsoft's documentation gives the formula explicitly as conversions divided by clicks, times 100, with its own example of 300 clicks and four conversions producing 1.33 percent.
Interactions and clicks are not the same denominator. An interaction is the chargeable engagement for the ad format, which on a search ad is a click but on other formats is not. So two accounts running comparably can show different conversion rates purely from what sits underneath the division.
This is not a rounding difference and it does not average out. It is two different questions wearing one label.
Return on ad spend is the same result in units a hundred times apart
This one produces the most dramatic misreadings, and it is entirely presentational.
Microsoft states its formula plainly. In its own documentation: "The formula for calculating the ROAS is (Revenue / Spend)." That produces a ratio. Spend 1,000 and make 5,000 and the number is 5.
Google works to a percentage. Its guidance for setting a target says to add the Conv. value/cost column, then multiply conversion value per cost by 100 to get your target ROAS percent. Its own worked example is explicit: 5 dollars in sales divided by 1 dollar in ad spend, times 100 percent, equals a 500 percent target ROAS.
So 5 and 500 percent are the same performance. Put the two reports side by side without converting, and one platform appears to be delivering a hundred times what the other does. Nobody has done anything wrong. One prints a ratio and one prints a percentage.
There is a naming trap sitting next to it. Somebody looking for a column called ROAS in Google Ads will not find one. The column is Conv. value/cost. ROAS appears as the name of a bid strategy, Target ROAS, which is a setting rather than a result. People conclude their account is configured wrongly when the metric is simply filed under a different name.
Even converted to the same units, the two figures are not quite measuring the same thing, and that matters more than the units do.
Microsoft's formula uses revenue. Google's uses conversion value, which is a number somebody assigned when the conversion action was set up. If the person configuring the account decided a form submission was worth 200, then every form submission contributes 200 to the numerator whether or not a single job resulted.
That is a legitimate way to work. Value-based bidding depends on it. But it means Google's figure is only as honest as the values entered, and those values are frequently a guess made once and never revisited.
The other thing neither platform does is subtract what the work costs you. A return of 5 sounds excellent and is a loss if your margin is 15 percent, because the platform counts the sale, not the profit on it. Both numbers are revenue-side. Your margin is the part only you know, and it is the part that decides whether the campaign is worth running.
One campaign, both vocabularies
It is easier to see all of this at once on a single set of figures.
Take a campaign that spent 2,000 dollars, took 500 clicks, recorded 560 ad interactions, produced 40 conversions, and had conversion values totalling 10,000 dollars.
Google would report a conversion rate of 40 divided by 560, which is 7.1 percent, and a Conv. value/cost of 10,000 divided by 2,000, which is 5, described against a target expressed as 500 percent.
Microsoft, given the identical activity, would report a conversion rate of 40 divided by 500, which is 8.0 percent, and a return on ad spend of 5.
Four numbers, two of them different, and only one of those differences is real. The conversion rates genuinely differ because the denominators differ. The 500 and the 5 are the same result in different clothes. Treat the figures as an illustration rather than a benchmark; the shape is what matters.
Click-through rate is the one they agree on
Worth knowing where the ground is solid, because it tells you which comparisons are fair.
Google gives click-through rate as clicks divided by impressions, with the example of 5 clicks from 100 impressions being 5 percent. Microsoft documents the identical formula, written as clicks divided by impressions times 100, with its own example of 50 clicks from 2,348 impressions producing 2.13 percent.
Same numerator, same denominator, same units. A click-through rate from one platform can be read next to a click-through rate from the other with no conversion and no caveat.
So the rule is not that cross-platform comparison is impossible. It is that it is metric by metric. Some columns are directly comparable, some need converting, and some are answering different questions entirely.
How to compare them honestly
Six steps, and none of them require platform access.
- Write down the formula each platform uses for the metric in dispute before arguing about the number. Both publish them.
- Convert units before comparing anything. If one figure is a percentage and the other a ratio, one of them is a hundred times the other by definition.
- For conversion rate, check the denominator. Interactions and clicks are different divisors and the column header will not tell you.
- For return on ad spend, ask what the conversion values are set to and when they were last reviewed. A number assigned once and forgotten is still in the numerator.
- Apply your own margin to both figures before deciding anything. Neither platform knows what a sale costs you to deliver.
- Compare each platform against its own previous period as well as against the other. A platform improving against itself is a real signal even when the cross-platform comparison is muddy.
The uncomfortable version of this answer is that there is no single number that settles which platform is doing better. There is your spend on each, the work that actually arrived from each, and what that work was worth. That calculation lives in your books, not in either report, and it is the only one that is denominated in the same units on both sides.
Reading platform reports next to each other is one of the things we do, so if two numbers have been disagreeing for a while, that is work we take on.
Terms used on this page
Sources
- Conversion rate: Definition
- About Target ROAS bidding
- Report Attributes and Performance Statistics (Microsoft Advertising)
- Clickthrough rate (CTR): Definition
Last reviewed 2026-09-11.