Yes, You Can Reconcile Attribution Data Discrepancies between Google and Facebookadmin
One of the most common issues brands see when advertising on Facebook is trying to understand and reconcile attribution data discrepancies between Google Analytics and Facebook Ads Manager. Wouldn’t you love to be able to report to your CEO that you can explain why these numbers don’t always match up? Even better, you could detail how to find a closer estimate of your actual attribution.
Below, I break down a step by step guide on how exactly you can gather your data and then analyze it so you can provide a clearer picture of your digital advertising performance.
Just as the saying goes, “There are three sides to every story: yours, mine, and the truth,” the true story of your digital advertising attribution is somewhere in the middle of these two sources.
Two Main Reasons for These Discrepancies
While data discrepancies can be due to an array of factors, there are two common reasons behind the majority of them.
(1) Conversion Timing
The timing of when the platform registers the conversion event is one of the major reasons behind data discrepancies.
Google Analytics (GA) reports last-click conversions as its default setting while Facebook reports instead on any conversion within a 28-day window that comes after a click.
This means that any user that clicked on a Facebook ad and didn’t convert, but later came back through a different channel (search, display, direct, etc), would be counted by Facebook but not by GA.
(2) Cross-Device Conversions
Cross-device conversions are not counted in GA, but they are counted when you’re using the Facebook pixel.
In practice, this is when a user clicks on an ad on one device and eventually converts on another (perhaps switching from smartphone to desktop to do additional research). Facebook will count this conversion, but GA will not.
It’s important to note this effect is significantly stronger for more expensive products and services such as high-end retail and online education since research plays the biggest role in their decision-making.
Whenever it’s the case that the click on the Facebook ad was not the last click, or the user switched devices, it’s even more crucial that we still account for their role in the conversion funnel.
The reason it’s so crucial to take Facebook’s role into account here is because without that click on your Facebook ad, the user may not have been aware of your brand, been considering your brand, or needed that reminder to get back into your funnel.
If we were to rely solely on last click reporting, the reporting would fail to give Facebook any value in that customer journey even though it likely influenced at least a part of your consumer’s decision.
Reconciling Data to Determine Attribution
Despite what digital advertisers may wish, there is sadly no magic formula that can calculate attribution perfectly. Even the most sophisticated and expensive solutions leave out many qualitative and quantitative factors.
The two methods described below connect the dots to give you a much better sense of how to reconcile your reporting discrepancies.
3 Questions to Answer before Reviewing Your Attribution Data
Before we can move to how to reconcile your data differences, there are three important questions you must answer to ensure your calculations are as accurate as possible.
(1) What percentage of your conversions within the 28-day window (or whatever window you are using) come in on Day 1?
This percentage usually ranges between 50-85% for most brands. It also depends on your product price, the split within your budget between prospecting and retargeting, and what the consideration time is for your product.
(2) What is your ratio of Facebook conversions to GA?
In our own research, we typically see this ratio in the 1.5-2.0 range. By this, we mean for every 1.5 or 2 Facebook conversions, there is a single GA conversion.
Just as with any calculation for digital advertising, this number varies based on a number of factors including consideration time as well as the relative size of your other paid channels.
(3) Is the above ratio consistent over time or are there major shifts?
If your ratio is consistent, then you can skip ahead to the section below. But if there have been changes in your ratio, then there are some additional questions you need to to consider.
First, do these shifts line up with any budget changes you made either on Facebook or for any other paid channels?
Second, were you running any non-digital paid advertising such as TV commercials?
Lastly, have there been any technical changes made on your end, such as a site redesign or pixel placement?
You’ll want to take these questions into consideration because any fundamental external shifts in your ad budgets on other channels could easily explain any changes in your Facebook to GA ratio.
For example, say you started running search display ads, having never run them before, and your ratio ends up going from 1.5:1 to 4:1. You can assume that this increase is due to the new traffic from your new ads and therefore the ratio you see presently has become your new normal.
Two Reports to Pull for Reconciling Discrepancies
Now that we have a high level view of the differences within the data, it’s time to dig into the details and start to quantify how conversions should be weighted.
Both of these methods below tell a piece of the attribution story that we can aggregate into our reconciliation:
(1) Google Analytics Top Conversion Paths Report:
This report can be found within your Google Analytics dashboard and will show all click conversion paths that users took to eventually convert.
Use this report to analyze all of the conversion paths potential consumers take that include Facebook as a step, but not as the last step.
If the steps after Facebook are paid channels such as display retargeting or paid search, we can derive a formula to attribute credit to the other channels within the path.
However, if the later steps are mostly unpaid formats such as direct traffic or organic search, we can attribute most, if not all of the credit, to Facebook.
The reason we can do this is because it’s accepted knowledge that Facebook plays a big role in brand awareness when it comes to increased direct or organic search traffic following a brand running advertising on Facebook.
We get to that belief because users who arrive to your landing page either directly or via organic search or directly are already aware of your brand. A user that goes directly to your site had to type in your brand’s name, meaning that they must have discovered your name somewhere.
And since we have proof that they clicked on your Facebook ad, we know that they either learned of your brand or were reminded of it by that same ad. As the bulk of organic traffic for most brands is brand name searches, if consumers navigate to your page by organic search, it’s most likely that they searched for your brand name.
(2) Halo Effect
Another way to measure the effect of Facebook on your overall conversion volume is to create a scatterplot of monthly direct and organic conversions as compared with your monthly Facebook spend.
As expected and explained above, we see a strong positive correlation between Facebook spend and direct and organic conversions for every client we have ever analyzed.
While the size of this effect varies from client to client, the correlation is always there.
Once you’ve built your chart in Excel, we can then add a trendline and look at its equation. The slope of the line is what you’ll use to estimate how many incremental conversions you earned as a result of your Facebook advertising.
For example, assume the equation of the trendline is y = 0.015x + 500.
Leveraging our 7th grade math abilities, we can estimate that if we ran no Facebook we would have 500 direct and organic conversions per month.
From there, every $10,000 spent drives an additional 150 conversions (10,000 * 0.015). If your Facebook CPA is $100, that means that each $10,000 spent would drive 100 Facebook conversions, and an additional 150 “halo” conversions.
The purpose of this Google Analytics Top Conversions report is to see how many of the conversions ended from an unpaid channel vs paid channel. If a majority of the conversions are from the paid channel (whichever paid advertising platform that may be), then the same paid channel should get the bulk of the credit.
But if a majority of the consumer journeys end in unpaid (i.e. direct + organic), then Facebook should get the credit.
Not to mention, giving credit to direct and organic channels won’t help you to scale your business
Being conservative when reviewing this data, we often discount the halo effect by 50% or more, depending on the overall volume of data.
This data volume is obviously more reliable if a client has been running ads consistently on Facebook for two or more years versus a client that has just begun ramping up their spend over the last six months.
You’ll use this report in addition to the the Google Analytics Top Conversions report to showcase a more in-depth story surrounding your customer’s journey to conversion and reconciling the discrepancies between your GA and Facebook Ads Manager.
No model is perfect on its own, so it’s important to look at all inputs with a qualitative view. Using the two methods described above, you can create your own method for how you weigh Facebook’s impact on your overall business.
By reconciling the attribution data discrepancies between your Google Analytics and Facebook Ads Manager, you’re able to get a more accurate picture of what is driving your traffic.
Once you’re able to determine which channels most influence your customers, you’re then able to work to scale them and further grow your overall business.