“How incremental is the affiliate channel?”

This is, far and away, the most common question we get about the impact affiliate marketing can have on a brand’s cross-channel growth. There are a lot of ways to get answers to that question, but there are also a lot of potential obstacles to making sure that information is useful and at least directionally accurate.

In this post, with a big assist from our Associate Director of Search and Social, Carrie Abel, I’ll run through some initiatives you need to do to get the right answers – and potholes to avoid along the way.

Do #1: Align on your definition of “incrementality”

When we start digging into incrementality measurement for our clients, the first step is to make sure everyone on the brand side is on the same page regarding the exact thing they’re trying to measure.

A general definition of incrementality is impact that would not have happened without an event, whether that’s a promotion, a new channel, a new partner, a new ad, etc. Note: be careful not to conflate incrementality (the measure of lift) with attribution (the act of distributing credit to a channel or tactic).

Do: #2: Get nuanced

As mentioned above, there are lots of things you could measure for incrementality, so don’t just stop at affiliate in general. We recommend measuring specifics, like bottom-funnel partners or a new placement, to get more accurate (and actionable) information on what’s actually driving growth.

Once you’ve pinpointed what it is you’d like to assess, you’ll have lots of options for testing methods, ranging from holdout tests to comparing historical overall data to data from a specific time frame when a campaign or placement was running.

Brands will also want to consider weighing different factors, including seasonality or segmenting returning vs. net-new customers, in their calculation of incrementality.

Long story short: make sure you’re running through a few baseline questions to align stakeholders before jumping into a measurement model.

Do #3: Invest in measurement

Hat tip to Carrie (whose team runs paid search, paid social, and SEO campaigns for our clients) for this one: if you’re big enough to invest in omni-channel advertising, you’re big enough to invest in measurement solutions, or you risk compromising performance across your account.

Even if you’re “just” using a free tool like GA4, you can and should make sure you’re using the right attribution models (anything but last-touch, really, but we recommend the data-driven model as a default). It’s also important to understand where your internal data and your affiliate platform’s data (e.g. from Impact or CJ) diverges and how to reconcile the differences.

If you do have the budget and the internal resources to stand up a third-party (read: objective) tool like Rockerbox or Northbeam, it’s a worthwhile consideration – getting closer to understanding the impact of each channel will make for smarter investments throughout your mix.

Don’t #1: Make assumptions

Lots and lots of brands enter a relationship with DMi assuming bottom-of-funnel affiliate partners, like Rakuten and Honey, aren’t incremental. That can be true, but it’s always something you have to test to know for sure. We’ve seen brands turn off those partners and get a corresponding drop in revenue.

The same can happen on the paid media side, with brand search and retargeting campaigns: there are frequent assumptions about incrementality that don’t always hold up when run through analysis. So: before assuming anything, get some data to support or debunk your claim.

Don’t #2: Get comfortable with silos

Often, we find that brands with work to do in understanding incrementality have a bad side habit of keeping paid media and affiliate teams in silos. Siloing your strategies can cause a lot of targeting overlap and a potential mismatch in your messaging, which serves to damage both your incrementality and your brand presence online. And on the other hand, there can be blind spots and gaps in audience coverage if the teams aren’t aligned.

Other repercussions: inefficient spend from the fragmentation of measurement with different platforms and methodologies. A much healthier set-up is to use one source of truth across all campaigns.

Don’t #3: Test for incrementality in the high season

Well-structured holdout tests (one popular way of measuring incrementality) are effective, but they should be deployed strategically – e.g. not at a high-tide time of year when throttling back spend to any audience with high intent will curtail revenue.

Yes, you’ll get valuable learnings (probably faster) when purchase intent is up across the board, but we always recommend that you take your time and set up incrementality testing when the stakes aren’t so high.

Even with all of these addressed, there are many ways to skin the incrementality-measurement cat, and we work with all of them across our client portfolio. If you’re looking for a clearer understanding of the incrementality of your affiliate efforts, we’d love to start a conversation.