How We Report on Facebook Ads Using Google Analytics

Facebook Attribution vs Google Attribution

The discrepancy in Facebook Analytics and Google Analytics is still as alive today as it was back in 2015. Understanding which platform you should trust, how to account for the reporting differences and why they even exist in the first place are crucial in making smart, data-led decisions when it comes to your performance marketing.

First things first, you’re not alone. Every business advertising across the Facebook network (Facebook, Instagram, Audience Network) is confronted with this discrepancy in reporting between Facebook and Google. It is nothing to be alarmed about.

Before we jump in, worth doing two quick things first to make sure you’re actually set up correctly – kind like checking you’re plugged in before trying to switch on the light:

1. Check your Facebook Pixel is installed correctly. Use the Google Chrome Pixel Helper add-on (turn off Ad-Blocker if you’re running that) as a first check, then walk through your purchase funnel whilst having Facebook Events Manager open to check the Pixel is firing correctly at each stage in real time.

2. Confirm your Google Analytics code is set up correctly

With both installed correctly, you’re half way there but the trouble is you’ll still see differences across the two platforms…

So what do you do? Well, let’s start by outlining why this discrepancy occurs and what the best solution is for your business.

How do Facebook and Google Analytics collect data?

This is the first step to understanding what’s going on.

It simply comes down to understanding the fundamental differences in the way the two platforms collect data.

Google Analytics uses cookies to measure how users interact with your website. Cookies are small text files that are logged in your internet browser (Safari, Firefox, Chrome) when you visit a site. They are only logged on the device you use to visit the site and don’t track you as a person.

Facebook tracks actual people based on their profiles, not just your browser. These two differences are crucial to understanding how to report on the two platforms and is at the core of the major differences between their figures.

More on this in a moment. But first…

What about attribution windows?

So the tracking is different. But to make it just a little more confusing, so too are the attribution windows: the time period during which eligible conversion events can be claimed.

N.b. Conversion events could be an app install, someone adding to cart, purchasing or any other event you set up as a “conversion” on Facebook or Google.

By default, Facebook attributes a conversion on a “28 day click, 1 day view” basis. i.e, if someone clicks on your Facebook advert and purchases within 28 days, or only views your advert and purchases within 1 day, it will be attributed to your Facebook advertising.

An easier way to understand it is Facebook shows conversions based on the time of the ad impression and not at the actual time of conversion. Say someone is served an ad on the 6th but doesn’t purchase until the 10th, Facebook will attribute this purchase in its reporting to the 6th.

Google Analytics operates differently, on a “Last click” basis. i.e, the final referrer before the action was taken. For the same example above, Google Analytics would report a purchase on the 10th, ie, the actual day of conversion.

Obviously, these two default attribution windows are designed to benefit the two different ad platforms. N.b if you’re not already, make sure you’re setting up Google Analytic ‘utm tags’ so you can effectively track facebook data in GA, and attribute where conversions have come from.

This infographic is a useful visual for understanding why Google Analytics might not credit Facebook as the source if the ad was served on the 6th but conversion landed on the 10th, whereas Facebook might take credit because it’s within the 28 day click, 1 day view attribution window.

To understand why Facebook and Google differ in their attribution models, it’s important to understand they are both approaching advertising from a fundamentally different place.

Facebook is interruption marketing, where you are targeting people who are browsing their Facebook/ Instagram feeds, not browsing an online store. As a result, people are far less likely to convert right away from Facebook ads, so it needs a longer window to track conversion.

Google Ads is intent based marketing, designed to target people who have shown intent to buy and are at the point of purchase. They are quite literally searching for what they want to buy right now, so are far more to “convert” immediately.

This is the fundamental reason for the two different attribution windows on Facebook and Google. As a result, it is necessary to read between the lines when reporting on both platforms and understand what is right for your business.

Are you selling a high ticket item, like a holiday, luxury goods or a service? Or are you offering a lower value impulse purchase, such as clothes or a gadget/accessory of some kind?

Just these factors alone will have a massive impact on how long it takes people to buy, how many devices they use and, as a result, how you should report on your marketing activity.

So, with all that in mind, let’s move onto the main causes of the discrepancy.

Facebook can track conversions across multiple devices, Google Analytics can’t

When you visit a website on your mobile, then visit it later on your laptop, Google Analytics sees this as two different people visiting the website because you have used two devices. Facebook recognises that it’s still you, just on a different device because of the ability to recognise your Facebook identity.

Because Google can’t join the dots here and Facebook can, you can get a broader perspective on how people interact with your website after seeing your ads on Facebook.

Let’s take this a step further and pretend you are selling products online.This is how it might look in the real world when someone buys something from your online store:

They’re browsing Facebook at work > click on an ad (on mobile) > Remember they should be working > Leave site > go home, eat dinner > Search directly for product on Google on their laptop > Click Google Search Ad > Purchase

This is a fairly typical conversion path in the mobile era, particularly amongst a younger, mobile savvy audience.

Facebook counts this as a conversion in Ads Manager, as the user clicked on the ad and then purchased within 28 days. Google attributes it to the final place the traffic came from, in this case the Google search ad on desktop. As Google doesn’t recognise the customer on their laptop, it doesn’t see Facebook in the conversion path.

Trouble is, amongst a younger audience this cross-device conversion path is very common. In fact, we often see 90% of Facebook ads served on mobile as this is where people use the platform the most.

This plays a significant part in creating a discrepancy in reporting between Facebook and Google Analytics, as Google’s default attribution model is based on last interaction, which worked perfectly until cross-device became a big part of modern purchase behaviour.

So, is there a way to test the effectiveness of Facebook Ads based on performance of Google Ads?

Whilst a water-proof test is not possible, it is possible to closely estimate the impact your Facebook advertising has on the performance of your Google Ads.

Whilst it does have extensive tracking abilities, Facebook isn’t able to account for all its purchases for various reasons, including people not being logged into Facebook on their browsing device. As a result, your Facebook Analytics might actually show a conversion figure lower than the real impact your Facebook advertising is having.

Once good solution is a conversion lift test. In effect, splitting your audience in half and only showing ads to half while tracking conversion behaviour from both. What this gives you is the intangible uplift in conversions from your advertising and allows you to calculate a % figure to estimate real ROI of your campaigns.

Any other ways to account for different figures in Google Analytics and Facebook Analytics?

Yes. If you are seeing discrepancy in conversion numbers, it is possible to analyse the data from your Facebook ads to compare the ‘impression device’ that people first saw a Facebook ad on, and the ‘conversion device’, the device on which they actually bought.

This will allow you to identify the percentage of your customers that have seen your product on mobile and switched device to actually purchase. As a result, you can put a number on the difference you expect to see in reporting between the two platforms.

Facebook can attribute conversions by impression as well as clicks

Because Facebook can track what people are doing within the app, you can also report on whether someone buys your product within one day of seeing one of your ads, without necessarily clicking. This is known as a ‘view-through’ conversion.

View-through allows you to still get an idea of the impact of an ad campaign, even if you aren’t including a link in your post.

Why does Facebook report using view-through? Consider this: you’ve seen an ad on your Facebook timeline but haven’t clicked it. Then you go straight to the business by manually typing the URL and then make a purchase. Very common.

Google Analytics sees that as an organic referral, whereas Facebook is smart enough to recognise you’ve been served an ad and made a decision off the back of it, hence it’ll attribute a conversion in Facebook. The view is, Facebook has contributed to the sale in some way through the user’s awareness of the product.

Taking this another step forward: imagine you have Dynamic Product Ads (DPAs) set up, which dynamically create adverts for the user based on what products they’ve seen on your site, or even added to cart. If you have multiple SKUs and don’t have these set up, get on it asap.

That same user is then later served an ad on Facebook with the exact content in their cart, remembers they haven’t checked out so just manually types in your URL and completes the purchase. This would make sense because there’s little reason to click the ad and they are already very product aware. The ad alone has reminded them that they haven’t completed the purchase yet and so Facebook should by rights take some credit for prompting them to complete checkout.

GA on the other hand, will again see this as organic because of the manual URL input.

View-through isn’t necessarily ideal for all advertisers and is often discounted in some attribution models, but it does serve a purpose. You also have the flexibility on Facebook to set your attribution window to your desired length and include/exclude ‘view’ conversions to suit how long your typical customer takes to purchase.

It’s an extremely useful feature but if view-through is a major issue for you, you can easily change the settings in Ads Manager to add or remove ‘view-through’ attribution depending on what you think is best, but…

Facebook needs lots of data to work effectively

Facebook’s attribution is often labelled as ‘greedy’. However, it’s important to recognise that this viewpoint doesn’t take into account the different roles that Google and Facebook play in your marketing mix.

Facebook is a ‘discovery’ platform, Google is a ‘search’ platform.

Facebook ads use unrivalled machine learning to match people to products that they’re likely to enjoy based on their interests and the actions they take online.

Google’s marketing products look to capitalise on people who have shown high intent to buy, as they have directly searched for your product.

As you can see when you break it down, both are designed very differently in order to make their ad platforms work as effectively as possible.

Facebook’s ability to report using ‘people based’ tracking is crucial to the success of the platform, as its algorithm relies on huge amounts of data to be able to show the right ads to the right people, at the right time to have the greatest impact.

This is arguably Facebook’s main benefit, as you are able to drive awareness of a brand or product to a new audience of people who are similar to your existing customers in a very targeted, measurable way, which is very difficult to do using other mediums.

In comparison, Google’s advertising solution, Adwords, allows you to use specific keywords relating to your product to pinpoint people who have shown clear intent to buy.

Because the person effectively tells you that they are very likely to buy by using a high intent search term, Google doesn’t need to process anywhere near as much data as Facebook in order to show its ads effectively.

Facebook and Google are competitors

It is worth considering that by reporting on your Facebook ads solely in Google Analytics, you are using software created by a direct competitor to measure results from an ad platform that operates in a very different way.

Once you are aware of the differences between the two ad platforms and why they report differently, you can take results from both and read between the lines to get closer to the truth.

So what does this barrage of information actually mean for you?

It means that the truth lies somewhere in between.

With so many marketing channels now playing a part in growing a business, the problem of attribution is at the forefront of nearly all marketers’ minds.

You need to be using a tool like Google Analytics to track the actual behaviour of users on your website, the performance of the website itself, as well as gaining an understanding of where your traffic is coming from. This is crucial to effective performance marketing.

However, once you have an understanding of how Facebook fits into your marketing mix, you can start to join the dots within your data and see lifts across all channels when your Facebook campaigns are performing well.

Facebook is an incredibly powerful tool for discovering, attracting and converting new customers who’ve never heard of you or your products, as well as reengaging warm traffic in a place they’re spending a lot of time. It allows you to “feed the funnel” with more potential customers and supercharge your search volume, which in turn helps your Google Ads performance.

In terms of solving the attribution model once and for all, there are third-party solutions available to help attribute value to your different marketing channels, some of which are using machine learning to really start to help close the gap, but we are still far from seeing the problem solved.

The real answer is to stop worrying about perfecting attribution and start helping more people discover you and your products across the Facebook network, whilst running Google Ads to help convert customers when they decide it’s time to purchase:

Increase brand awareness, social proof, super-charge search volume and drive purchases by putting your ads in front of potential consumers on the world’s #1 marketplace for consumer attention: Facebook, Instagram and its dominant social media network; and

Ensure you convert organic traffic, and make sure people you’ve advertised to don’t slip through the net to a competitor’s site when they come searching with the undisputed best intent-based marketing tool available: Google Ads.

This is somewhat simplistic but by setting up your ad campaigns in tandem this way you’ll maximise your return on ad spend (ROAS) and improve your overall ROI.