With broken attribution models, how can you confidently report campaign performance and optimize ad spend?
While perfect one-to-one attribution is impossible, all hope is not lost.
By shifting focus to longer-term trends and incrementality, capturing campaign data through UTMs, agreeing on a single source of truth, and having a plan for when tracking breaks, you can still glean actionable insights from PPC reporting.
With the right approach, paid campaigns can still demonstrate value and guide effective optimization no matter how murky the tracking waters get.
This article explores practical strategies to make paid media reporting work despite the messy post-cookie world.
Follow these tips to prepare your PPC reports for a post-cookie world.
1. Include comparisons
Whenever preparing reports, context is key. One of the best ways to position current data is to show how it compares vs. previous data.
Depending on your industry and how seasonality impacts results, you may want to emphasize comparing to the previous month or the same period last year (or both).
Talk about where you see growth and where you see declines, and think about factors such as budget changes and messaging updates that may have impacted performance.
2. Show longer-term trends
Additionally, just reporting on shorter-term periods lacks significance and may either oversell success or raise unnecessary red flags about performance lags that are ironed out over a broader timeframe.
For instance, you may find that looking at a one-week period in the middle of the month shows a drop in conversions, but your business tends to receive the most conversions at the beginning and end of the month. So once you factor in the entire month, data points to positive performance.
3. Use multiple platforms, but agree on a source of truth
Much is made in our industry about discrepancies between platforms, which often leads to a complete distrust of certain platforms.
For instance, Google Analytics events rarely match to conversions in Google Ads, for various reasons.
Look at the ratio of conversions between platforms, and keep an eye on if that changes over time. You might find that Google Analytics records 80% of the conversions you track with a Google Ads pixel.
You can use this as a rough benchmark for what to expect. However, if you suddenly see the ratio drop to 50%, you may want to investigate potential tagging issues.
Within your organization or through conversation with your client, you should agree on a common source of truth for reporting. This may be an analytics platform, a CRM, or a CDP.
Agreement on a common source of truth doesn’t necessarily mean that you’re seeing these numbers as the end-all guaranteed accurate lead totals.
Instead, you’re simply making sure you’re using the same data set when you present total leads/sales in reporting and can get ahead of potential questions if you only use ad platform reporting and your stakeholder only looks at Salesforce.
4. Capture UTM parameters
Maintain a system for adding consistent UTM parameters onto URLs for key dimensions such as source, medium, campaign, and keyword.
In turn, set up your CRM or whatever system you use to track leads to capture the UTM parameters attached to the initial landing page on which a user arrives.
This will help to provide your own first-party data on the source for each lead or sale. Additionally, iOS may strip some personally identifiable parameters (such as Google’s click ID or a Salesforce ID) in Safari when a user comes via private browsing. However, UTM parameters should not be stripped.
Get the daily newsletter search marketers rely on.
5. Incorporate coupon codes
Using a unique coupon code to tie to specific campaigns or users can be an additional route to tie performance straight to your paid media campaigns.
You have two options here:
- Assign a general code with a landing page that’s only used for a particular campaign.
- Use URL parameters to dynamically populate a code that’s pulled through into a form field when a user submits a lead form or completes a sale.
6. Consider incrementality testing
Another way to measure the overall effectiveness of a digital advertising channel is via incrementality.
Essentially, this involves setting up a test where a subset of users is exposed to ads and others are not and then looking to see if there was a lift in conversions for those who received ads.
There are a few ways to implement an incrementality test:
- Geography: Pick regions with similar populations and demographics to compare against each other.
- Time: Run ads for a set period that’s long enough to get the significance, and then turn ads off for the same amount of time, comparing results at the end. While this route can provide a clean way to test your entire geographic footprint, be wary about accommodating for seasonality.
- Audience split: Show ads to a test group and not a control group and compare performance between the two afterward.
7. Prepare for when tracking breaks
Inevitably, your tracking will break and you’ll overcount or undercount conversions. It’s important to have a plan in place for when this happens.
Tracking data via multiple platforms allows you to pull from a backup source when necessary.
While we discussed agreeing on a main “source of truth” for reporting consistency, you can explain that you’re pulling data from (for instance) the ad platform itself for a particular month instead of relying on CRM data when there was an issue syncing leads to the CRM.
Next, annotate when the tracking issue occurred and when it was resolved. That way, when looking back on results one year later, you’ll remember why there was a gap in data.
Attribution has never been 100% perfect. While there are good efforts in place by ad and analytics platforms to integrate the data, offline interactions, cross-device activity and multitouch behavior have always made attributing conversions to a particular source messy.
For instance, maybe someone who searched for your product and bought it already had a positive perception of it via word-of-mouth from a friend who wouldn’t be tracked in any analytics platform.
Ultimately, the goal of ad platform conversion tracking isn’t necessarily to be able to tie every single conversion to an actual purchase or lead on a one-to-one basis.
Consider conversion tracking more as directional data showing how metrics such as total conversions, conversion rate, CPA, and ROAS have changed over time.
Our ability to track marketing data will get messier. So think through how you approach reporting to your business stakeholders.
Make every effort to set up ad platform conversions, analytics events, CRM fields, and whatever else you’re using to track success.
However, be transparent about the fact that the data you’re able to track doesn’t give perfect attribution and that whatever you report is a guide for optimization and measuring business success.
Treat your data as directional and look at change over time, while looking to alternative methods to track success outside of what you can track directly in platforms.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.