\ Understanding Attribution for Paid Social - UTMs and GA Attribution Models
Feature: Page (1) of 1 - 01/24/18

Understanding Attribution for Paid Social - UTMs and GA Attribution Models

By Tim Brown

When you're running a paid social campaign - in particular, Facebook ads, a strange thing happens if you don't know to use manual UTM (stands for Urchin Tracking Module) tagging in your ads - Google Analytics doesn't quite know what to make of the traffic. The lack of out-of-the-box analytics performance here is frustrating if you don't know what to do next. It makes sense that the same level of ease for tracking wouldn't exist when compared to Google AdWords; however, the results are surprisingly useless. 

Facebook ads will typically show up as Source/Medium (source being the site that sent the traffic, and medium the classification for the type of traffic - typically direct, referral, organic, cpc/paid), Facebook/Referral, or m.facebook.com. Well, that tells you absolutely nothing. It doesn't tell you if traffic coming to your site is from a Facebook ad, or someone who clicked on a post you had out in the wild - it tells us absolutely nothing. That's a big problem when you're trying to evaluate the success of Facebook ad campaigns you're running, but thankfully there's an easy solution.

UTM Tagging

The primary fields you want to define for Google Analytics are as follows: utm_source, utm_medium, utm_campaign, and utm_content or utm_term. Below is a quick-and-dirty sample of what a utm tag looks like


The information that you put into source/medium/campaign are pretty simple; you want Google analytics to know that it's ad traffic from Facebook, and you're likely just going to use the campaign name as utm_campaign. We can have more fun and define other features in the secondary fields like content or term. Here you want to either use the Ad Set or Audience name, and may want to note anything you're testing. If you're testing two ads in one ad set, and you're testing the headlines against each other try adding the headline into the utm_content field (ex. utm_content=adset1_headline1) so in Google Analytics you can tell the difference between the two versions! This not only makes analyzing your campaign performance easy, but it's also not hard to implement. 

One quick thing to note, there are two different ways to add UTM's into Facebook Ads

1. Add to the end of your url (making sure to use a ? to separate the url and the tag): www.website.com?utm_source=facebook&utm_medium=cpc&utm_campaign=campaign1&utm_content=adset1_audience1

2. Place the utm tag into the URL parameters section of the ad creation screen, it's on the bottom left when you're making or editing your ad. This requires no ? at the beginning, simply put your tag in.

Bonus: Dynamic Tagging In URL Parameters Section

If you are running more complex ad campaigns and are combining your ad placements (rather than only running ads in Newsfeed, or separating your placements into their own ad sets) together you will not be able to tell in Google Analytics if there is a difference in conversion rates or on-site behavior between users from Instagram or Facebook Mobile Newsfeed. However, Facebook has you covered. It can take some patience, and is a completely different approach - but go here and behold the opportunity before you! Your dynamic tag would look something like this: 

utm_campaign={{campaign.name}}&utm_source={{site_source_name}}&utm_content={{placement}} - {{adset.name}} - {{ad.name}}
This would make it so you never have to write another utm tag, you could simply copy and paste this every time and get the information you want fed into Google Analytics. In my example above utm_content would tell you the placement, the name of the ad set, and the name of the ad - this is incredibly useful information but requires sticking to a good naming structure so you ad set and ad names mean something.

Google Analytics Attribution Models

The default method for giving credit for conversions in Google Analytics is always going to be last-click. What this means is 100% of the credit for a lead or sale is given to the channel that drove the last click to the website when the conversion occurred. In most cases, it makes sense that this would be the default method because we ultimately care the most about the way people last came to our site.

It's nice to think of a world where someone sees an ad we ran on Facebook with the perfect targeting and timing and simply decides to buy right at that moment - it rarely unfolds that way. In comparison to a channel like Google Search where there is demonstrated intent to seek out a product or service like yours, you're typically addressing someone at a time where they were not trying to find you (and possibly were not even aware of your existence) so this means we often have a longer path to conversion. 

Because this is the case, you may be severely underestimating the value of traffic from sources like Facebook. So what attribution models should you look at to get the whole picture? First, let's start by mentioning how this is done. Under the Conversions section of Google Analytics, we merely click Model Comparison Tool and select the models you want to compare against last-click attribution. The two models you should be looking to most often for comparison are first-click attribution and last non-direct attribution.

First-Click Attribution: 100% of the credit for a conversion or sale is given to the channel that drove the users first click to your website. Best used when evaluating awareness/high-funnel channels or campaigns. 

Last Non-Direct Attribution: Similar to last-click but ignores a direct click to your site if this was last, 100% of the credit for a conversion or sale is given to the channel preceding a direct site visit. This is useful as direct traffic to your site (someone typing in your URL) doesn't really deserve credit for a conversion action, looking at the previous steps would be much more revealing.

When we select these two models to compare to last-click attribution, we will want to look at the data by Source/Medium and see if Facebook/CPC is showing varied data in either of these new attribution models. This is a simple, efficient way to determine the overall success of your campaigns - use these models to guide your optimization and budgeting efforts, and you'll be able to make much smarter marketing decisions. Without this data (and the UTM tagging that came before it) marketing is more of a guessing game, don't let that happen to your marketing.

Tim Brown is the owner of Hook Agency, serving clients with their Web Design, SEO, Paid Social and PPC Management needs. 

Take a look at our blog to learn cutting-edge methods to get the most out of your digital marketing strategies. Tweet him at @hookagency

Related Keywords:Urchin Tracking Module, Google Analytics, UTM Tagging

Source:Digital Media Online. All Rights Reserved

Our Privacy Policy --- @ Copyright, 2015 Digital Media Online, All Rights Reserved