Before you start you must create a GA account and complete the steps in the Getting Started with Google Analytics guide to start collecting basic data from your website.
Many web applications have GA integrations or plug-ins. If you license a web application or use an open-source solution (such as Magento or WordPress), you can refer to the aforementioned guide to check if there's a preferred GA solution.
UTM parameters
Google Analytics defines the following tracking variables within your UTMs that can be appended to your ad destination URLs for campaign tracking.
To add UTM parameters to your AdRoll ads read this article.
| Variable | Required? | Description |
|---|---|---|
| utm_source | Yes | Use campaign source (utm_source) to identify the source of your campaign. |
| utm_medium | Yes | Use utm_medium to identify a medium such as retargeting or brand awareness. |
| utm_campaign | Yes | Use utm_campaign to identify a specific Campaign or AdGroup. |
| utm_content | No | Use utm_content to differentiate the type of ad, for example differentiating your static vs dynamic ads. |
You can use Google's URL builder to create the campaign tracking parameters that will help differentiate traffic from your various AdRoll campaigns.
www.adroll.com/landing_page? is the destination URL
utm_source=adroll indicates the source of this traffic as coming from AdRoll
utm_medium=banner indicates we served a web banner ad
utm_content=static_160x600_v002 indicates the ad size and version
utm_campaign=spring_promo_adgroup_2 indicates the ad is part of our Spring Promo campaign added to AdGroup 2
AdRoll reports available in Google Analytics
Warning
These reports only include visitors that click through. View-through data is NOT included.
Within Google Analytics, you can view basic metrics like visits and page views by navigating to Traffic Sources > Sources > All.
The default view shows sources, and mediums and provides a view of traffic sources for 30 days prior to conversion. Google Analytics processes users who engage with the AdRoll data collection Opt-Ins, such as the Safari banner, as having a UTM Source of app.adroll.com and the UTM Medium of referral.
If you set up goals or e-commerce reporting in Google Analytics, you can view the success metrics attributed to your advertising by browsing the appropriate reports.Set Up Reports in Google Analytics
To set up reports, you must:
- Track conversions using Goals or e-Commerce tracking
- Tag your marketing links with campaign tracking
Discrepancies between AdRoll and Google Analytics
Your campaign reports in AdRoll and Google Analytics reports may not always match up. While both systems aim to track the same trends for your campaigns, each system uses a different approach to reporting that will often lead to discrepancies.
Before deep diving into reporting discrepancies between AdRoll and GA, complete the following checklist to ensure the systems are correctly setup:
- Ensure the Google Analytics code fires correctly and collects data
- Ensure the Google Analytics javascript is present on each landing page
- Ensure your destination URLs have UTM tracking codes that properly assign the traffic source to AdRoll
- Ensure you are tracking conversions in Google Analytics and defining them in AdRoll correctly using Conversion Audiences
- Ensure the ad destination URLs are not prompting redirects. Redirects may strip the necessary UTM parameters
- Ensure you have not setup filters that are removing or editing campaign tracking
- Ensure your Google Analytics date/time is set to UTC to match AdRoll
After you have verified that tracking is set up properly, the source of the discrepancies may be caused by differences between metric definition between AdRoll and Google Analytics:
| Metric | AdRoll Definition | Google Analytics Definition |
|---|---|---|
| Tracked metric | Clicks | Visits |
| Tracked metric definition | When someone clicks your ad | Site visit with at least one page view, expiring after 30 minutes of inactivity |
| Attribution model | Tracks both click-through and view-through conversions, defaulting to a last-touch attribution model | Tracks only click-through conversions, defaulting to a last-click attribution model |
| How it works | Attributes a conversion to the most recent AdRoll advertisement click or view | Attributes a conversion to the most recently clicked ad across the entire marketing mix. If no ad is clicked leading to the session, the conversion is attributed to ‘Direct’ |
| Attribution window | Defaults to 30-day click conversion window, 7-day view conversion window | Defaults to a 30-day click conversion window |
| Cross-device attribution approach | Uses a combination of deterministic and probabilistic cross-device matching | Google Analytics does not have cross-device tracking capabilities. This means that if a user clicks a mobile ad, then visits the site on a desktop browser, Google Analytics tracks that as 1) two separate users & 2) a direct conversion |
Understanding Click reporting discrepancies
-
AdRoll reports more clicks or conversions than Google Analytics for Facebook
- Unless UTMs are appended to the ad destination URL, Facebook clicks are attributed to “Direct” traffic in Google Analytics.
- Cookies placed in a user’s primary browser are not accessible to in-app browsers causing the user to be viewed as two distinct users in Google Analytics
- When someone clicks from a HTTPS environment and enters a HTTP environment, the referrer URL can’t be recorded by Google Analytics.
- Google Analytics is not cross-device aware. So if a user clicks an ad on their mobile device, but later converts on their desktop computer, AdRoll will attribute the click-through conversion to AdRoll, but Google will consider the mobile phone and the desktop sessions to be from different users, and attribute the conversion to direct.
-
AdRoll reports more clicks than Google Analytics reports visitors
- A single visitor clicked your ads multiple times
- A visitor clicked your ad, but then stopped or left the page before it loaded
- A visitor navigates away before Google Analytics tracking code executes
- A visitor opted out of Google Analytics tracking, but not AdRoll tracking
-
AdRoll reports fewer clicks than Google Analytics reports visitors:
- A visitor clicked an ad, then during a different session returned directly to the site via a bookmark or typing directly into the web browser. In this case, the ad is still attributed for the first visit, so the initial click results in more than one visit.
- AdRoll automatically filters out known bots, creative validators, and creative testing activity from clicks data reported in ad campaign results. Google Analytics doesn’t automatically filter out such invalid traffic, which can result in a higher number showing in Google Analytics.
Understanding Page View reporting discrepancies
In GA page views are calculated based on sessions. This means that a user visits the site over a specific period of time (session), ideally, the GA default session time is 30 mins, for example, if an individual user visited their site in that 30-minute session, then after they came back to the site, that time they will be tracked as a new visitors.
- Sessions: It is a group of interactions that take place on your website within a given time frame.
- Users: Users who have initiated at least one session during the date range.
In AdRoll there is no such concept as sessions used to calculate page views. Page views are tracked based on the following sequence.
In addition, AdRoll filters out invalid clicks while GA users are calculated based on sessions since tracking methodologies for both metrics are different you can expect discrepancies while comparing these metrics.
AdRoll makes use of an efficient filtering mechanism to filter out the Invalid/Fraud data and show up only the valid data in our dashboard. There are some possible reasons for the discrepancy:
-
Users using Safari
Safari has been removed completely from AdRoll, so we cannot track the visitor's data and conversion data coming from the Safari browser. -
Pixel Behavior
If the user uses an Ad Blocker or any other plugins that interrupt the Pixel from dropping a cookie, then the users will not be tracked.
We won't track the users who have cleared cookies (or landed in an incognito window).
There are also some smaller factors for such discrepancies, like timezone (all our data is in UTC).
Recommendations
Here are some tips for understanding the reporting discrepancies in your own account.
- Focus on AdRoll click conversions as Google Analytics only attributes credit to click conversions.
- To find AdRoll-attributed click conversions in Google Analytics, combine the last click and assisted conversions. Navigate to Conversions > Multi-Channel Funnels > Assisted Conversions. Sum the last click and assisted conversions attributed to AdRoll and compare to AdRoll reported click-through conversions.
- AdRoll will most likely still report more click-through conversions than Google Analytics. Add a dimension (e.g. source/medium) to drill into which inventory source (e.g. Web/Facebook), then compare this to AdRoll Web- and Facebook-attributed click-through conversions. Typically, AdRoll Web conversions match closely with AdRoll-attributed Web conversions.
- Assuming we’ve isolated the discrepancy to Facebook, click into the AdRoll Facebook line item in the assisted conversions report, then add a secondary dimension of Device Category.
- Compare to AdRoll’s placement report. The discrepancy is usually isolated to the Facebook mobile placements due to the reasons previously outlined.