Understanding Data Discrepancy between Google Ads and Google Analytics

2022-09-27 | Article | Insights

If you are working with Google Ads and Universal Analytics and in parallel already with Google Analytics 4 long enough, it is only a matter of time before hearing or asking the question, “why does my data not match?”. In most cases, the answer typically is, “it depends”. Then after recreating the same reports hundreds of times in the hope of a data matching, the confusion continues, and you start to ask yourself:
  • Am I even comparing the right set of data?
  • How is this data even being collected?
  • Is this normal?

Well, you are not alone. These questions, among many others, are asked every day. Below we explain how data is collected, why it's important to understand the counting method, attribution, and a recommendation of other areas that should be reviewed in order to compare data confidently between Google Ads and Universal Analytics.

Why do we link both tools in the first place?

Let us begin with linking both tools. Both tools are powerful alone but together they can be even more powerful. This is why linking them is worth taking into consideration and most users do. In general, there are two main reasons to link:
  • You can observe the behavior of the traffic coming from Google Ads. When you send traffic to your website and log into Google Ads, you will see metrics such as impressions, click-through rate, etc., but it doesn’t provide information once they get on to your site. That’s where Universal Analytics comes to play because it observes and helps identify the quality of that traffic such as time on site, bounce rate, pages viewed, etc. This enables you to answer the question, “are you buying the right traffic”.
  • The second reason is to create some Universal Analytics remarketing audiences. These are audiences you can create in Universal Analytics based on various metrics or dimensions that resemble the desired behavior (e.g. newsletter subscribers and viewed product detail pages), these audiences can then be pushed to Google Ads and used in campaigns. This gives you an edge and targets users you already know will engage.

Now that they are linked, what should I consider?

When linking both tools you will notice that you have two methods to choose from when tracking conversions in Google Ads. The first method, is using Google Ads conversion tracking and the second, is using Universal Analytics, where you can import goals or eCommerce conversions into Google Ads.

Google Ads provides various conversion tracking options for the website (e.g. sales, or actions), for the app (e.g. install and in-app actions), and phone calls (e.g. track calls from ads or websites or you can import them from another system like Universal Analytics). If you have Universal Analytics connected to Google Ads, you can choose things like your eCommerce conversions (e.g. transactions) or goals (e.g sales) and import them to Google Ads.

Let’s say you use a Google Ads conversion tracker to track sales and you also import a sales goal from Universal Analytics. What is the difference, as they both technically should be using the same data set? Firstly you need to ask, how is the tracking setup? Suppose we use a Google Ads conversion tracker. In that case, it could be installed through a google tag manager (GTM) or directly on the site and this will send data to Google Ads or we set up Universal Analytics and put the tracking code on all the websites, declare a website sale as a conversion, and import it all into Google Ads.

So in essence, the difference in setup is not too dissimilar, however, there are differences, and let's start with the process of counting.

Counting differences

Google Ads has a flexible counting method with two options to choose from, one being counting one conversion per ad click or counting every conversion per ad click. Universal Analytics counts differently on the other hand, as goals are counted only once per user session.

If for example, your goal is to track sales conversions multiple times after a click from Google ads in Google Ads (e.g. user buys various items separately), and you set your settings to count every conversion, it will count every sale after the click individually. If you are tracking the same sales goal in Universal Analytics, you would only see it counted as one conversion within the given session time. So to get a closer number between Google Ads and Universal Analytics, you would also need to adjust the counting to one conversion per ad click.

So, depending on the goal and counting setting, this could drive differences between Google Ads and Universal Analytics. This is why it is important to understand the counting method settings and remember them when comparing later down the road.

Attribution differences

In addition to counting, you must consider how the tools attribute the counted conversions. Google Ads attributes a conversion when a click has led to a conversion with a prior ad click from a Google ads advertisement. That means that Google Ads needs to have two data points in place in order to ask itself, “was there a conversion?” and “was there an ad click from google ads prior to the conversion?”, if yes, then we would see Google Ads get the credit for the conversion. This is called a last-click attribution model.

Universal Analytics, on the other hand, uses two different attribution models (depending on the report) and asks two questions, “was there a conversion?” and “was the last known source Google Ads?”, if Google Ads was not the last known source, then Universal Analytics will not send the data to Google Ads as an imported goal.

This is why understanding the attribution rule is important and why the term “last non-direct source” is a critical term to also be mindful of. By clicking on your Acquisition> All Traffic > Source /Medium in the Universal Analytics interface, you can see sources like: google/organic and (direct) / (none).

Google / organic counts those conversions that have a known non-direct source, such as e.g. referral, social, another advertiser, etc. Unlike direct / none, which counts those that do not have a known non-direct source.

By going into the Conversions >Multi-Channel Funnels > Top Conversion Path of your Universal Analytics interface, you can see examples of the known non-direct source being counted under the google/organic source.

The fact that both tools are counting the “same conversion” but attributing them differently shows how important it is to understand how each tool collects, counts, and attributes conversions, since depending on how you interpret the information you will end up always with a discrepancy.

What else to consider?

As you can see from the above, a mix of configuration differences and tracking differences lead to many of the discrepancies between Google Ads and Universal Analytics. Below is a list of other factors to consider within the configuration and tracking areas:

  • Account linking - is the account correcting linked to the right google ads account and analytics property?
  • Tags Properly firing - Is the Google Ads and Universal Analytics tag firing?
  • Auto-tagging - is the auto-tagging enabled or are manual UTM parameters being applied correctly?
  • Google Click ID - is the google click Id being removed?
  • Filtering - Are there filters in place that are removing google ads data?
  • Lookback window - Are the look-back windows matching and do they reflect the user journey?

  • Attribution Model - Did you consider the attribution model differences between Google Ads & GA?
  • Day of attribution - are you comparing day of click in google ads vs day of conversion in Universal Analytics?
  • GA Session vs. GAds Clicks - Are you comparing sessions from GA vs Google Ads clicks? Did you consider that they can have a systematic errors because of the GA session definition?
  • Attribution time - is the attribution time window between the Google Ads & Universal Analytics too long?
  • Time zone - is the time zone between the engine account and the analytics view the same?
  • Lookback window - is the lookback window between 7 and 90 days compared to the default 6 months window in Universal Analytics?


Understanding all the potential factors that could lead to data discrepancy sets you up to analyze data that is meaningfully comparable, improves the quality of your analysis, and ultimately provides better insights.

Are you interested?

Contact us