For centuries, marketers have struggled to understand exactly which of their marketing programs and touchpoints deliver value. Many companies are now in the process of modeling and measuring cross-channel attribution across their businesses. Unfortunately, partner marketing data -- including affiliate sales activity -- hasn’t always been a part of those models.
All marketers care about getting attribution right. But for partner marketers, the stakes are particularly high. Including partner data in attribution models is critical to demonstrating that sales through this channel are profitable and incremental. Many senior execs at brand marketing companies still have their doubts on this score, so you need hard numbers to prove the value.
Fortunately, more and more companies are focused on including all marketing touchpoints -- including those powered by partners -- in their analytics and attribution modeling. If you’re a brand looking to get your partner and affiliate marketing efforts a seat at the attribution table, and thereby prove their value, here are three key things you must do.
Recognize that attribution is a cross-channel analytical challenge.
In partner marketing, some people think of attribution as a single channel analytics challenge -- that it is all about assigning credit to specific partners according to preset, rules-based methodologies. But that’s not really attribution. Consumers don’t just interact with one form of media on one type of device. Rather, they seamlessly migrate from touchpoint to touchpoint and expect to find their brands ready to meet their needs everywhere.
Understanding the interplay across media and devices is what attribution is all about. To gain this understanding, you need to include a comprehensive set of data on consumer events for the partner marketing channel. Through advanced analysis, that data -- when combined with performance insights across the marketing mix -- helps to identify the relative importance of different touchpoints toward driving a conversion. Across channels. Across devices.
Deliver the right partner marketing data inputs.
When you commit to including partner marketing data in the modeling, you need to ensure that you collect all of the data necessary to provide a complete picture of consumer behaviors across every partner-driven touchpoint.
For most customer journeys, there are a number of measurable events between the first partner click and the final purchase or conversion. Each step in the process -- for example, product searches, browsing, check sizing, add to cart, use of a discount code, etc. -- must be separately measured and the data compiled at a granular user level. Only with such a complete set of data can you map the entire customer journey and understand how the partner channel contributes to conversion.
Ensure timely data ingestion and processing.
Finally, it’s important to streamline the process of getting partner event data into your attribution solution. Whenever available, high-quality APIs are a great way to simplify data ingestion and processing. That’s as true for partnerships as for any other marketing channel.
The timeliness of the data that you ingest is important -- real-time data has far more value for optimizing results than data that is days or weeks old. Further, real-time information at the user level is essential for powering the most sophisticated attribution models.
While APIs aren’t entirely necessary to transport data from one place to another, they make it far easier -- and with less potential for errors. They speed initial integration and ensure that data is consistently delivered in the most timely way possible.
Doing attribution right -- across channels, including partnerships and with complete data -- will help enrich your understanding of what works. The journey toward more complete attribution isn’t an easy one. But by incorporating partner and affiliate marketing into your attribution models, you are able to empirically show that these efforts deliver real, incremental value and that they impact far more buyer decision making processes than one might expect. Not a bad combination, right?
