Discrepancies in the analysis of the number of clicks and user sessions

The first differences between Google Analytics statistics and Facebook statistics can already be seen in the method of counting clicks and user sessions. So let’s start by explaining these concepts.

– A click is a parameter that sums up all users’ reactions to a given ad. So if you click the ad once and then click again, Analytics counts that as two interactions with the ad.
– The session, in turn, shows how many users clicked on the ad. Two or more clicks of the same user will be considered by the system as one session, provided that they occur within a 30-minute time window. Only when the time interval between clicks is greater, will the actions be considered as two separate sessions.

Up to this point, everything is quite simple and it would seem that there is no room for discrepancies in the counts here. And yet. Because when can you start counting users who actually reacted to the ad? Google Analytics claims that only when the user is on the page to which the ad is referring him. After all, he might have clicked the ad by accident and stopped loading or closed the tab in the browser before the landing page was fully loaded, and this is hardly an ad success. Therefore, such “incomplete” transitions are not counted in the Google Analytics system.

Facebook, on the other hand, will record each click, i.e. the user’s interaction with the ad, even if the recipient does not fully see the landing page or does not go beyond the Facebook platform at all. It is true that in the Facebook Ads Manager you can measure not only all clicks, but also clicks on links that lead beyond the platform – to the advertiser’s website. However, the Facebook system will also count those clicks that did not fully load the landing page. What is the conclusion of this? The number of clicks reported in the Facebook Ads Manager may be greater than the one we see in Google Analytics.

Tracking users between devices

Analytics is therefore much more accurate when it comes to counting sessions and clicks on a single device level. What about the situation when the user is using different devices and clicks on the ad, for example, first on the phone and then on the computer? Here, Facebook is much better at counting. This is because in order to use Facebook and thus click any advertisement there, the user must be logged in to his account. Most users use the same account on all devices, so Facebook’s algorithm has no problem distinguishing when it is dealing with the same user and when not.

In Google Analytics, the default data collection is based on cookies, and these are assigned to a specific device. Hence, each change of the device by the user qualifies him as a new person.

What does this mean in practice?

If the recipient first clicks on the ad on their phone and then reacts to it again on their computer, Google Analytics will register the ad as two separate users from different traffic sources. On the other hand, Facebook, if the recipient is logged into the same account on both of these devices, will recognize him as the same person.

Differences in attribution models

Conversion may be preceded by a variety of user actions. This means that before he or she fulfills our goal, e.g. buys an item in an online store, he may deal with the website in various ways. Let’s see it with an example. The user sees an ad for shoes from your Facebook store. He clicks it, looks at the shoes, but decides not to shop. After a few days, he decides that he needs shoes and enters a query to the Google search engine, for which he will receive an advertisement in the search engine with a link to your website. He clicks on the link, looks at the same shoes again, but decides to ask his wife for opinion before buying. In the evening, he will enter the search query again, but will not click the sponsored link, but choose an organic result, and this time he will buy shoes.

So to which interaction with the website can this purchase be attributed? Did the first contact with the offer, i.e. Facebook advertising, decide that the user has performed the assumed conversion? Or maybe the user did not remember it anymore when he searched for shoes in the search engine? So then the sponsored link would make it hit your store and make a purchase there. But after all, he did not buy shoes by entering the site through an advertising link, but through an organic result, so maybe he should be “responsible” for the conversion? On the other hand, there is a chance that the user initially had no plans to buy shoes and it was only advertising on Facebook that awakened this need in him. In this situation, the first ad would do its part to persuade him to buy.

The questions are constantly multiplying and it is not easy to find an unambiguous answer to them. Therefore, individual analytical tools may adopt different methods of calculating ad effectiveness, i.e. they have different attribution models. We will now show you what it looks like in the case of the Facebook Ad Manager and Google Analytics.

Facebook Ads Manager – attribution model

If the user makes a conversion on the page within 7 days from clicking on the ad, Facebook Ads will assign it to his ad (of course, if we have a well-configured FB pixel). The same will happen if the recipient meets the target within one day of being shown the ad. It is true that the user could later have contact with ads in other channels, but Facebook will attribute this success to itself anyway. This is currently the default attribution model for this channel and may be objectionable but legitimate. Facebook is not a strictly sales channel, so the task of advertising on this platform is to arouse the user’s desire to buy. So even if an advertisement did not lead to a purchase, but sowed a seed in the user’s mind, which then germinated and gave fruit, it is also a merit of advertising on Facebook.

For example, the recipient might want to compare prices with competitors before buying, or look for opinions about the presented product in order to finally go directly to the company’s website. However, he found it through an ad on Facebook, so the system rightly recognizes its participation in such a conversion. In addition, the Facebook system is not able to control whether the user was later exposed to ads created by other systems, so the assumption that the Facebook Ads campaign influenced the recipient’s decision is the best way to count conversions.

However, is it reliable? Unfortunately not. For this system, clicking an ad is not synonymous with loading the landing page, as we have already written about above. It’s even more difficult when counting views. Facebook’s algorithms take into account whether the advertisement was displayed to the user, but they do not measure the time that the advertisement was visible on the user’s screen. So it may turn out that yes, it appeared on the page, but it was scrolled so quickly that the user did not have a chance to even notice it, let alone read its content. Nevertheless, if the user subsequently performs a conversion within a certain period of time, Facebook counts the conversion as obtained from its ad.

Google Analytics attribution models

By default, Google Analytics uses a different attribution model – the last indirect input is counted as the one that decided to convert. What is an indirect input? Simply put, it can be any way to enter a website, except for directly typing its address in the browser bar. So if, for example, a user clicks on a Google Ads ad, but does not buy anything, and then enters the page by typing its address into the browser and converts on the page, Google Ads will receive the credit for it. Comparing this counting system with the Facebook system, it is easy to notice that the results of Facebook Ads campaigns in Analytics may be weaker than in Facebook statistics.

However, Google Analytics also offers other attribution models, and changing the default settings can significantly change the results presented by this tool. We can choose models such as:

– Last interaction – then the system assigns all credit for the conversion to the channel that the user used immediately before the conversion. This setting makes sense when your ad is targeting people who are determined to buy, not those at an earlier point in the funnel.
– Last Google Ads Click – All 100% “responsibility” for the conversion is attributed to the last ad displayed by the Google Ads system that the recipient clicked before converting. It is worth using this model only when you want to compare the effectiveness of Google Ads. Does not apply to statistics for Facebook ads.
– First Interaction – In this model, Analytics considers all credit for conversion to go to the channel where the user first encountered our site. It can be used in this case when we want to build brand awareness with advertising.
– Linear – considers 1 conversion = 100% and grants equal percentage of share to all channels through which the user came into contact with the page. So if he first saw an ad on Facebook, then clicked a sponsored Google Ads link and made a purchase, he will attribute 50% of the value of this conversion to Facebook Ads, and the other 50% to Google Ads. It is best to use this model when each user contact with the website is important to us.
– Timing – assigns different conversion shares to channels, depending on how many days before the conversion was done by the user. The closer the contact with a given channel is in time, the more important that channel is.
– Including items – this model also divides the share between channels that are on the purchasing path, but does not take time into account. Most often, in this model, 40% of the conversion value of the first interaction is assigned, 40% of the last interaction, and the remaining 20% ​​is divided equally among the other channels on the purchasing path.

Google Analytics or Facebook Ads – which statistics should be taken into account?

Advertisers often try to unify the results obtained from both tools as much as possible. You can, for example, change the attribution in Facebook Ads so that conversions are counted only within 1 day of clicking on the ad. However, before taking such steps, it is worth analyzing your business and industry or tracking the average length of your audience’s conversion path. Sometimes it’s not necessarily about fixing the discrepancy, but about accepting that both Google Analytics and Facebook are complementary tools that can be used in parallel to better understand your business with web analytics and make better marketing decisions.