Spur Reply | Thought Leadership

Channel Planning Is More Than Just A Good Algorithm

Written by Richard Flynn | May 16, 2020 7:11:45 AM

We've spent quite a bit of time here talking about how using data is essential to taking your channel business to the next level. A fundamental flaw of using channels is partners always have a different agenda from you. This means there is a always a drift, either slight or big, from your objectives. The goal of channel management is to minimize that drift and create revenue acceleration for your company.

With the shift to the cloud, the ability to effectively plan has never been easier, with software available to track partner performance, customers, and the solutions they consume. We've also seen some improvement in systems that make measuring partner performance for lead management, deal registration and marketing effort possible. Analytics engines can now assimilate data from multiple sources, manage complex models, and create bots to derive insights.

The era of channel management big data is here. What more could we ask for?

Big Data isn't enough

At The Spur Group we've completed many projects that boil down to becoming a data-driven channel organization. During that time, we've seen what works and what doesn't.

In the latest issue of the MIT Sloan Management Review they have a great article on why big data isn't enough. While the article isn't specifically about channel management, the warning is appropriate.

Algorithms are powerful but they aren't perfect. Too often they take a casual relationship and can exaggerate it's importance. They lack the natural ability to filter on the important data points and sometimes look at too much information to draw clear conclusions. They also tend to magnify inherent biases that are build into your data set.

The bottom line is you need to be cautious when jumping into the big data pool.

It's all about having a hypothesis

The key is how you use big data and modelling for channel management. It is critical that you formulate a hypothesis and use big data to test that hypothesis.

For example, we have a hypothesis that a partner's revenue growth is always a function of improving some mix of a partner's sales velocity (contribution), product mix (capability), market presence (coverage), competitive share (commitment), and customer-lifetime-value (consumption). We call this the 5Cs model.

Once you have a hypothesis you can use big data techniques to test your thinking and drive your analysis.

Let's look at our example again. We use big data to find out what lever to use with each partner to have the greatest impact. This informs specific business plans with each managed partner and creates a structure to developing programs for unmanaged partners.

The important takeaway is get more from your data by having a well thought out hypothesis.