Machine learning and AI algorithms. They’re everywhere nowadays. Companies are using them for all sorts of purposes—but mainly to help them predict the future.
The direct selling industry is no exception. In fact, DirecTech Labs’ artificial intelligence (AI), DISTRO, generates a wide range of predictions and scores to help direct selling companies predict the future paths of their customers and distributors. It’s especially handy for preventing direct selling turnover or “churn”.
Direct selling companies are continually impressed by DISTRO work. But invariably, they begin to ask questions about how DISTRO comes up with its numbers.
The one number that gets questioned most is ‘Risk Score’—the likelihood (from 1 to 100) that a particular customer or sales rep is in his or her last 30 days with your company. It’s churn prediction distilled down to a single two-digit number!
How on earth does DISTRO generate a statistic like that? After all, DISTRO crunches data from thousands of customers and sales reps all around the world every day. Why—and how—can DISTRO call out one particular person as likely to be leaving your company at any given moment?
Meet Jane. She’s a model distributor for your company—one of those bread-and-butter sales reps you hope to have thousands more. Jane has been on autoship with your products for the past 8 months. She recruited a new distributor last month and tallied $250 in commissions for the month. She also added a new customer just yesterday.
Based on these metrics, you wouldn’t think that Jane is a poster child for direct selling turnover. While she may not be heading for your Platinum rank in the comp plan, she’s doing about what you’d hope for at this stage. That’s not what DISTRO sees. Using its powers of churn prediction, DISTRO has given Jane a Risk Score of 87.
Why does DISTRO think Jane is in her last 30 days of activity? Because DISTRO can see patterns of behavior far more complex than any human is capable of. And he sees Jane’s behavioral patterns in ways that we can’t.
DISTRO also compared Jane’s profile to millions of profiles it has identified and created over time. That way it finds people who were very similar to Jane in certain ways. This could include:
What’s happening in DISTRO’s “brain” is really complex, but the output (a risk score) is quite simple.
For each criterion DISTRO considers, there may be thousands of people who match Jane reasonably closely. But DISTRO won’t go off of one number—it will weigh certain criteria more heavily based on the closeness of the match.
The Risk Score that DISTRO comes up with, then, is not just an educated guess at churn prediction. It’s a highly accurate score based on a highly complex series of interrelated factors. In fact, it’s so complex, DISTRO couldn’t even explain it to you if you asked. Silly humans. We just wouldn’t get it
But, just because we won’t be able to fully understand how DISTRO (or any AI like the ones working on you every day via Netflix, Google, etc) comes up with an answer, doesn’t mean you won’t be able to independently verify its accuracy. “We actually guarantee the accuracy of DISTRO to our customers.” says Michel Bayan, CEO of DirecTech Labs.
How accurate? Fair question.
No AI engine is perfect, and no software vendor should be promising 100% accuracy. However… What DISTRO can promise is the ability for you to verify its accuracy. When DISTRO says someone is in their last 30 days with your company, it’s right 80 to 90% of the time. We’ll let you verify through a transparent and simple process. You’ll see how highly accurate DISTRO has been at predicting direct selling turnover.
Remember, predicting churn is nice to do—but at the end of the day, if you can’t turn those predictions into people staying on board longer, then you’re not really getting anywhere.
Learn more about how DISTRO actually retains customers and distributors at risk.