Whether you are using Facebook Ads or Google Adwords to get paid traffic to your site, there is a big probability that to some degree, the data reported in those platforms does not match with what is being reported in Google Analytics.
And if your website gets many sales or you are getting lots of traffic from these platforms, the situation only gets worse. The bigger your sales and paid traffic from these sources are – the higher the discrepancies in data reported.
To bring some light into this subject, and some possible fixes, we decided to create this post where we will analyze each individual case in which the data discrepancies might occur, and if possible, how to fix them.
- Facebook Ads and Google Analytics data mismatch
- Google Adwords and Google Analytics data mismatch
Facebook Ads and Google Analytics data mismatch
When it comes to Facebook ads, the majority of data mismatch fall under 2 big categories: 1) discrepancy in Facebook Clicks and Google Analytics sessions, and 2) discrepancy in Facebook conversions and Google Analytics conversions.
The major reasons for discrepancy in Facebook clicks and Google Analytics sessions are:
- Using “all clicks” instead of “link clicks” – Facebook offers a variety of metrics to count user clicks and ad interaction, the most popular ones being “Clicks (All)” and “Link Clicks”. The “Clicks (All)” metric includes all clicks on the ad such as likes, shares, link clicks, etc., while “Link Clicks” includes only the clicks on the actual links from the ad. Often times users confuse these two metrics and instead of using the “Link Clicks” metric in their Facebook reports, they use the “Clicks (All)” metric.
Another important note is that the “Link Clicks” metric counts all link clicks, even the duplicates, so when a user clicks for 2 times on a link from an ad during the standard Google Analytics session duration of 30 minutes, Facebook will report 2 “Link Clicks” while Google Analytics will show only one session. To exclude this mismatch, it is better to use the “Unique Link Clicks” metric instead of “Link Clicks” in the Facebook reports.
On the other hand, Facebook doesn’t require cookies to be enabled in order to track clicks on the ad, so in this way, Facebook is recording the click while Google Analytics is not recording the session, thus the discrepancy between the two.
Users exiting before the GA code loads on the site – In some cases, the user that clicked on the Facebook ad might exit the website so quickly that the GA code doesn’t manage to load, thus leaving that session unrecorded in Google Analytics. Facebook would have still recorded the click so this is another way in which data discrepancy between the two appears.
For this reason, it is very important to fire the Google Analytics tracking code as soon as possible when the page is just starting to load.
Viewing the GA report with sampled data – By default, Google Analytics reports are not sampled, but in some cases, when the property records a lot of sessions or when segments are used, the report may contain sampled data. In this case, some data from Facebook or other sources may show inaccurately. To learn more about Google Analytics data sampling, please read this article.
Referrer data removal by mobile apps – Sometimes mobile apps that display Facebook advertisement remove the referral data when the user clicks on the ad, thus Google Analytics is unable to know the source of the visit, and in majority of cases, assigns that visit to direct traffic.
Currently this is the most common cause of data mismatch for paid mobile traffic from Facebook. Adding UTM parameters to the links you share on Facebook may potentially fix this, but not always.
Different reporting time zones – If your Facebook ad account uses a different time zone than your Google Analytics account, then when you compare data for the same time period on both platforms, you will get different result. To fix this set the same time zone on both platforms.
Google Analytics is not installed correctly – If Google Analytics is not installed (or the installation is broken) on pages where you send the paid Facebook traffic, then those sessions will not be counted in GA thus causing data discrepancy between the platforms.
In terms of discrepancy between Facebook conversions and Google Analytics conversions, the most common reasons are:
Improperly installed Facebook pixel – Although not as frequent as it used to be in the past, this is still a very common reason why the conversions (purchase, add to cart, custom events) reported in Facebook and Google Analytics do not match.
Different attribution models used by Facebook and GA – By default, Google Analytics uses the last non-direct attribution model which means that if a user clicked on a Facebook ad and came to the website, then exited and came back using a Google organic search and converted, by default, Google Analytics will attribute that conversion to Google organic.
On the other hand, the default attribution model used by Facebook attributes any conversion to itself if the user either clicked or viewed the ad.
So in the above example, the default attribution model from GA will attribute the conversion to Google organic while Facebook will attribute it to itself.
In GA, you may view the conversions using different attribution models by going to “Model Comparison Tool” from Conversions report while in Facebook, you need to go to Setting -> Attribution -> Edit .
Different date of click vs date of purchase – Facebook attributes the conversion based on the date users see the ad and not on the date they convert, like it is the case in Google Analytics. This means that if a user saw the ad yesterday, and converted today, Facebook will attribute the conversion to yesterday’s date, while GA will attribute it to today’s date.
Cross device tracking – Because users need to log in to use its services, Facebook has the ability to track cross device conversions, a feature that Google Analytics does not have by default.
This means that if a user clicked or saw an ad on his mobile phone, then went to the website directly using his desktop computer and converted, Facebook will be able to track the conversion, while Google Analytics will fail at this and will attribute the conversion to direct traffic. The ability to accurately track users across all devices is one of Facebook’s biggest advantage over Google Analytics.
As you can see from the above, except for the issues related to having Facebook Pixel and Google Analytics properly installed and configured, there is very little what you can do to fix the data discrepancy between the two platforms.
And in most cases you do not even need to fix the discrepancy. It is very important to know that Google Analytics and Facebook ads are different tools, which use different techniques and metrics to track and report user interactions and conversions. So it is a good idea to view them as complementary platforms, working together towards providing you a better view and understanding of your business’s advertising and marketing activities, each one with its benefits and its drawbacks rather than expecting one to perform like the other.
Google Analytics and Facebook Ads use different techniques and metrics to track clicks and conversions, thus the discrepancy in their reporting, so it’s a good idea to view them as complementary platforms rather than expecting one to perform like the other#Data #Mismatch: Don't expect Google Analytics to report like Facebook Ads - They use different techniques and metrics to track clicks and conversions Click To Tweet
Google Adwords and Google Analytics data mismatch
Now that we explored the reasons why data in Google Analytics and in Facebook might not match, we can look at the causes why there might be discrepancies between Google Analytics and Google Adwords.
Being native platforms, Google Analytics and Google Adwords usually report exactly the same data, but in some instances, there still appear some mismatches, so it’s important to explore the causes and the ways of fixing them.
Before continuing further, it is important to note that clicks and sessions are different metrics thus they are recorded and reported differently. A user can click on an ad multiple time during the standard GA session duration of 30 minutes and Google Adwords will record all of those clicks while Google Analytics will show only one session.
The most common reasons for the data discrepancy between Google Analytics and Google Adwords are:
Not having the two accounts linked – This may seem as self-explanatory, but sometimes users simply forget to link the Google Analytics property and the Google Adwords account. Or, in some cases, they disable it for some time, and re-enable it back, only to discover later that this caused some data mismatch in their reports.
So the first thing to check if you are experiencing some data mismatch is to see if the Google Analytics property and Google Adwords are linked and the linking is enabled for the GA views you want to see the data.
Missing Google Analytics tracking from the landing page – Before proceeding to other checks, you should always ensure that your Google Analytics tracking is firing correctly on your landing page.
Disabled auto tagging in Adwords – If auto-tagging is turned off in Google Adwords, and you did not tag the destination URL’s from your campaigns manually, then Google Analytics will attribute the traffic from Adwords to Google organic. To fix this, you just need to enable the auto-tagging in adwords.
Having liked multiple Adwords accounts with your GA account – It is common for advertisers that run multiple online businesses to run all their campaigns from the same adwords account, causing data from other website’s campaigns to show in every GA property that is linked with the particular adwords account.
To fix this, you simply need to create a destination filter in every GA view from all properties, and to include only the campaigns were destination url matches (via regex) with the default url of that particular Google Analytics property.
Filters (both in GA and in Adwords) – If you have a filtered view in GA which is excluding some valid traffic from Adwords, you will get a mismatch in the data reported in that view and Adwords. For this reason, it is important to always have a raw, unfiltered view in Google Analytics.
Another cause of data mismatch may be the built-in fraud filter that Adwords has. If a user is causing fraudulent clicks, like clicking on an ad many times during a small interval of time, Adwords will exclude those clicks from their reports while Google Analytics will still record the session.
Different attribution models – By default GA uses last non-direct attribution model while Adwords reports all conversions where a user clicked on an ad and attributes it to the campaign and creative that generated the last AdWords click.
This means that if a user clicked on an adwords ad, exited the website, and came back again using a Google organic search and converted, Google Analytics will attribute the conversion to Google organic while Google Adwords will attribute it to the last campaign that generated the user click.
Counting model: one conversion vs every conversion – Goals in Google Analytics are tracked only once per session while in Google Adwords, depending how you have configured the conversion tracking, you can count only one conversion per ad click or you may count every conversion per ad click.
This means that if a user came from a paid Google search and clicked 3 times on a button from your website that you track as a goal, in GA you will see one goal completion while in Adwords, if you have configured the conversion to track every conversion, you will get 3 conversions.
Date of conversion – In Analytics, conversions are reported against the date of the conversion while in Adwords, conversions are reported against the date when users clicked the ad. For example, if a user clicked on the ad on June 10th and converted 2 days later or June 12th, in Google Analytics the conversion will be attributed to June 12th while in Google Adwords it will be attributed to June 10th, the day of the ad click.
Phone call conversions – Since this is only tracked in Adwords, Google Analytics has no way to report them, so if you are comparing the total conversions from GA with total conversions from Adwords, don’t forget to exclude the phone call conversions from Adwords total conversions.
Redirect to different pages – If your landing page is redirecting to another page using a 301 or 302 redirect, then Google Analytics will not fire on the landing page and will not record its referrer information, so the real referrer information of that session will not be tracked in Google Analytics.
Ad blockers – If users use ad blocking or privacy extensions in their browsers that might prevent Google Analytics from firing, then sessions from those users will not be recorded in Google Analytics, while in Google Adwords the clicks would still show, thus generating the mismatch.
Users return to your site using a bookmark – Google Analytics uses the gclid parameter from the landing page to identify traffic from Adwords so if a user has bookmarked the landing page that contains this parameter and comes back using that bookmark, Analytics will mark that traffic as it is coming from google/cpc while Adwords will not record that as a click.
As we have seen, there are many instances where data mismatches might occur between Google Analytics, Facebook Ads and Adwords so when dealing with a discrepancy, it is best to check all of these points to see if it can be fixed or not. With Google Adwords, most of data discrepancy can be fixed however in the case of Facebook, besides having the GA code and Facebook Pixel properly installed, there is little what can be done to fix the mismatch, so as stated earlier, it is better to view the data from each platform as complimentary, each helping the other in creating a better understanding of your marketing and advertising campaigns rather than expecting one to perform like the other.
When submitting the audit form, if you have some data mismatch, please write some details about them in the comments field, so that our experts would take a closer look at the issue.