LinkedIn Twitter Facebook RSS Feed

Turning sales data into predictive analytics

Podcast episode 28

Article by: Moy Chambers | CEO of Allspice Technologies, and Advisory Board Member for PurePredictive.com Direct Sales
Kenny Rawlins
September 4, 2018

If you run a direct sales company, chances are you capture a lot of data. What are you doing with it? On today’s episode of the podcast, Moy Chambers joins us for a second time to talk about the value hiding in that sea of data points and how machine learning can help you access it.  

Listen in to hear about intelligent retention systems. Lower your churn ratio with predictive analytics that tell you who’s going to leave early—before they even know it. And you’ll hear more about one of our favorite topics lately: how you can use comparative analytics to identify successes in one branch and replicate them across the tree. All that and more on this episode of The MLM.com Podcast. 

Moy is CEO at Allspice Technologies and sits on Pure Predictive‘s advisory board. Check out his previous guest spot on the podcast: Keeping MLM software simple in a sea of complications. 

Full transcript 

Kenny Rawlins: Hello and welcome to the MLM.com Podcast, brought to you by InfoTrax Systems. I’m your host, Kenny Rawlins, and in today’s episode we’re joined by Moy Chambers, who has an extensive background in the IT field and specifically in direct sales. Moy is currently the CEO at Allspice Technologies, which is a consulting company that specializes in direct sales technology and mobile technology especially. And he also sits on the advisory board for Pure Predictive for the direct sales arm of their business, and it’s actually the work that he’s done with Pure Predictive and some analytics that we’re bringing Moy on to talk to us about today. Moy, how are you? 

Moy Chambers: I’m doing good Kenny, how are you doing? 

Kenny Rawlins: Great! Thank you for joining us! So, let’s just jump right into it. You and I have been talking a little bit about some machine learning and predictive analytics and the role that that could play moving forward for successful direct selling companies, and so, I’ll let you just start us off and give us a little bit of background on what you’ve been doing and where you see this technology going. 

Moy Chambers: Yeah, so what’s interesting about this technology—machine learning and artificial intelligence—this industry has so much data from sales figures and everything else and they want to get into big data. And I was sitting with the CEO one day and he says, “Let’s get into big data.” And I started looking around at how we could use utilize big data. And big data can tell you a lot of things, but one of the best things that it can help you with in our industry is the churn ratio in the MLM industry. It seems to be that every company, no matter what size they are, the number of new enrollments that come in—they’ll have that same number of people drop off and so that’s the churn and so with artificial intelligence and machine learning, you can look at thousands of data points and some of these data points are inside your business and some of them are outside your business with big data. So, what we’re trying to do is use machine learning, use algorithms to come up and analyze these thousands of data points to come up with a score that you can use in your direct selling business for your distributors. 

Kenny Rawlins: Yeah, and I mean, you pretty much laid this out, but when we talk about churn, we’re talking about new enrollments minus the people that have left the business, right, and that’s your net growth and if that ever gets negative that’s when you start to see a decline and so that’s what we’re talking about when we talk about churn. And one of the things that I found interesting that you brought up as we were talking in preparations for this is you said almost all companies have what they call a retention team. But the interesting thing about the term retention team is generally what that means is the same thing that it means like with your Comcast’s or your DIRECTV’s or any of these other service providers, which is once you’re already unhappy and you’ve called in and said, “Hey I’m leaving,” then they kick you over to the retention team to see what they can do to make it better and keep you there. And really, what you’re talking about is using predictive analytics and machine learning to prevent that call from ever being made. To be more proactive rather than reactive in responding to people’s complaints because you can see it in the data. Am I summarizing that correctly? 

Moy Chambers: You’re right, you are, and one thing you can see and one thing that this industry doesn’t focus on—you’re right about the retention—but preventative medicine. Just looking at big data and looking at thousands of data points and through these algorithms and getting the score—low, medium, or high—you can use this as if it was preventive. You’re going to prevent something from happening. So preventive medicine is another way of looking at this. Right now, our industry is waiting until someone’s sick and then reacting to it. 

Kenny Rawlins: Exactly. You know, it’s interesting you bring this up cause I was just—I mean this isn’t related to the MLM space, but it’s exactly what you’re talking about—I was listening just a couple days ago to a podcast where they were talking about how in the NBA, the teams that are really leading edge have moved past state-of-the-art injury and rehab technology. And that’s basically baseline, everybody has that now. But the teams that are really cutting edge are getting into preventative medicine, just like you say. They’re doing everything they can to have the bodies of the athletes tell them before an injury happens. Because it’s something like, if you sprain an ankle, you’re 19 times more likely to have a future sprain of that ankle, versus if you never had a sprained ankle. And so, I think that ties in with exactly what you’re saying, right, is we generate just millions and millions and millions of points of data across a customer base, and I won’t even use the term distributor base because this is wider than distributors. You can tell, a lot of times, by people’s ordering patterns and things like that when they are starting to get dissatisfied. And there are companies out there and technologies out there that are now going to help people to capitalize on that. To say, “Hey, we want to make sure we’re providing the best customer experience possible. And we’re going to be proactive in making sure that that happens by using the data of our customers to know when people are unhappy.” And I’m curious, what would your message be to the technology leaders in this industry? 

Moy Chambers: The thing is, these companies can’t go out and hire PhDs that understand big data. And so, you need to find a partner that understands big data, understands these algorithms, and can then take your data and… (I’m going to be working here shortly with some small startup companies—one of them has been in business for a year, one’s been just starting up.) …and start gathering this data because of the models that we have have been based on usually two years’ worth of data. You send us two years’ worth of data, we’ll analyze it and we’ll show you your scores. But it’ll be interesting with these new companies, working with them and partnering with them, that we could show them the tendencies that their customers and distributors have, and which ones will be at high risk, because you want to focus on those high-risk ones. And, you know, this model does come from the customer base, and now we’re taking it to the MLM space. Because a distributor might be at a medium rank, if he’s looking at having a high probability of leaving your company, but on average he would stay with your company for another two years. Whatever rank he was, whatever you call him, a diamond or whatever, if he leaves early because the scores are showing that his tendencies are pointing in his data points that he might leave early. But usually he takes his downline with him. So, you just don’t lose one sale, you could lose, you know, a hundred thousand in volume per month. And if he leaves two months or two years early, that’s 24 months of a hundred thousand in volume. So, in this industry, the churn, or the scoring—especially if somebody’s in the upper ranks—becomes really big. It could save, it could be at least—if you’re a two hundred-million-dollar company—it could be a five to ten-million-dollar swing for you on the bottom line. 

Kenny Rawlins: And obviously then, these measures that people will take end up paying for themselves and then some. I’m curious, Moy, I want to get your thoughts, and this isn’t something you and I have talked about but it’s something that in today’s climate I’m curious what your thoughts are. You know, you hear so much in the news and everyday about data privacy and things like that and I can imagine that we may have some listeners who are hearing this that are thinking, “Man, they’re sifting through my data and trying to guess my behaviors beforehand.” I’m curious what you think about this at a privacy level, and I have some thoughts of my own, but I’ll start with getting yours. 

Moy Chambers: Well, you make a good point. The privacy issue here becomes with all these thousands of data points, you’re measuring somebody. And I don’t think this is what these data points are good for. They’re good for the corporation to find out what your score is—high, medium, and low—and I think for security and stuff you might want to keep them in those three ranges. You never want to give these scores back to distributors, but you might want to tell a distributor and his downline, “Here are some people that are showing warning signs.” Maybe rate them, like green you’re okay, yellow’s a warning, and red’s a higher one. But one of the things I’m not too focused on is that we don’t focus on the individuals when we go out there and look at big data, but we look at the individual and we sit and we look at demographics. And you know this guy lives in Ohio, you know this is how old he is, this is you know his sex, and then compare that to big data. All the other guys that live in Ohio, in this zip code range, what are their spending habits? So, these are some of the things that big data does. It looks at your guy and looks at what’s trending out there and compares the two together and then that makes the scores more accurate. So, you could say this guy living in Ohio, compared to every else in Ohio, he’s going to be scoring high. His predictability of churning is high right now. 

Kenny Rawlins: Yeah! And I guess my response to my own question, which is always a funny thing to do, but I think there definitely is an onus on companies to act responsibly with the data that they have. The type of the analytics we’re talking about, I don’t feel like they get into a dramatic breach of confidence, and the end goal is to, like I said, enhance customer service. To make sure people are happy and that they are getting what they want out of their experience with the company. And so, to that end, like I said, I think there is a heavy burden on companies anytime they’re using data provided by their customers, but I don’t think that this stuff inherently crosses that line. And these are practices that retail and other companies are using, you know, just wildly, right? I mean, when you go buy something there are 50 different companies figuring out ways to tie all of your purchases together and understand your habits and your life events even. And so, I think, like you said, there is a sense of being responsible, but then there also the customers and distributors should see benefit from these activities. What I mean by that is, like you said, the MLM space, the direct sales space, is interesting in the sense that you’ve got some data points that you don’t have otherwise, right? For instance, who your upline is, and what organizations you’re a part of. And one of the things even in some more rudimentary analysis that I’ve been a part of is seeing a dramatic difference in behavior based on the upline, and that tells you that there are certain trainings and certain business practices going on within different downlines. Some of those are good and some of those are bad, and out of these measures, you’ll be able to—as a company executive—recognize and support the people who are generating the correct behaviors.  

One of the things that we do here at InfoTrax is commissions consulting and data analytics in that way and it’s amazing the number of times that we’ve gone in and worked with companies and identified downlines who are actually engaging in behaviors that are not good for the customers and they’re not good for the company. And we say, “Hey, you know these people are engaging in behaviors that are counterproductive.” And the CEOs or the marketing directors or whoever we’re working with will say, “What do you mean, these are some of the people we bring across stage every year promoting how great they are.” And that’s because the rudimentary statistics can oftentimes be misleading. And one of the examples of that is there’s a lot of companies I know that track new enrollments or new customer acquisitions but don’t track turnover. So, they see that this organization is bringing a lot of bodies into the door, but they don’t see that for every one person they’re bringing in, 1.3 people are leaving and then it causes them to recognize and acknowledge the wrong people. And so, it is going to be interesting to see the benefits that come to the good actors in the space as their hard work and their correct behaviors become more evident through these advanced analytics. 

Moy Chambers: You made a great point, if you’re able to now take these scores from big data and everything and then go into people’s downlines and compare down lines against each other, you might say this guy’s downline has high scores. Maybe the guy is really good at recruiting people and getting them to sign up, but he’s terrible at retention. And then you can look at people that might not be as good at signing people up, but they’re great at retention, and you can have that… You’ll have a score to compare them with—an average amount of churn—and you can start analyzing them by trees, through the branches. So sometimes everybody looks at the whole company, but sometimes it’s just the behavior of a leader and his downline. 

Kenny Rawlins: Exactly. And, you know, this is something that—we hear about advanced analytics in all areas of our lives whether it be sports, or you know, what Facebook and social media are doing, or even, you get into, we’ve got the midterm elections coming up and they talk about the demographic data there and getting down to the micro level of precincts and things like that and it’s only natural that we in the direct sale space need to utilize that as well and it can be a powerful tool for good, especially when it comes to, like I say, improving customer experience, also when it comes to terms of regulating ourselves. There are things in the data that will allow companies to better regulate what is going on and so that we have less people that feel manipulated or feel misled in terms of the business opportunity and I think that’s a powerful thing on the horizon. 

Moy Chambers: So, one thing is, what do you do once you get the score? You know, that becomes a big thing. You can break it down by downlines, look at leaders, and so forth, but if you were to present to your customer service when somebody calls in, you know you have a red flag up on the on the screen or yellow flag. The customer service should treat every customer as if they’re the number one customer. But if you see a red flag or a yellow flag, maybe you go the extra mile, okay, and then you make note of what action was taken by that customer service rep. Some of these yellow flags… You could do an email campaign or you could do a text campaign or you could have your internal departments, the elite teams, reach out to these people. But whatever they do, what’s really important is to record their action so you could feed that back into the system, so the system sees these people and sees what works. Did the call work, did the text work, did the emails work? You got to keep feeding this thing so it keeps learning. But that’s the beauty of this. Sometimes, for example, a downline might take off or something might happen in a downline. In this valley, in Utah Valley, there was a company [going through that] last year, and at the DSA, their president got up and spoke about this toothpaste story. And they had no idea why their toothpaste sales just skyrocketed up and so they’re asking sales, “You running anything?” They said no. “Marketing, you do anything?” “No.” And then they turned to the Commission Department and said, “Hey, show me somebody’s got a spike in their commission, their earnings, and it ended up being a group of women that just loved the toothpaste, started promoting it out on social media, and the bottom line was a woman was trying to earn five hundred dollars to repair her husband’s truck. At the end of the day the CEO says, “You know, your next check, you could buy your husband any truck he wants.” And so, they did—if they would have had big data or some other things they could have. These are events that occur and you want to measure them and replicate them so everybody can be successful. 

Kenny Rawlins: Absolutely. And I think you make an excellent point in the sense that this is about replicating those good behaviors. Well, Moy, you’ve been very generous with your time and I’m grateful for you coming on and giving us a little insight into what’s coming and what we need to be looking out for because, you know, it’s important that as the direct selling space that we’re taking advantages of these technologies, and, like I say, especially using them for the betterment of our distributors and our customers. 

Moy Chambers: In closing, I just want everybody to understand, this is science. It’s not black magic. It’s down to a science and these scores come back with thousands of data points. If you have your sales or your marketing or your BI team try to analyze this, on the surface, they will not understand it because of the algorithms and how deep they go and how wide they go with all this big data. So, it’s technology you just gotta trust. It’s like getting in a self-driving car—are you going to trust it or not trust it? But this technology, I think, is a lot safer than the self-driving car. 

Kenny Rawlins: And it’s something that’ll continue to be improved and I think you’re really going to see companies be separated by those who get on board early versus wait until it’s too late. But thank you, Moy, and we look forward to having you on the podcast again soon and we appreciate the work that you’re out there doing! 

Moy Chambers: All right, thank you. 

Kenny Rawlins: And that does it for today’s episode of the MLM.com Podcast brought to you by InfoTrax systems. As always, I’m your host, Kenny Rawlins, and I want to give a special thanks to Moy Chambers for his time and his insights into machine learning and predictive analytics. Production support from Jana Bangerter and we hope that you’ll join us again next time. 

Read more About

The MLM.com Podcast, Software for Companies, Network Marketing and MLM Software

Moy Chambers | CEO of Allspice Technologies, and Advisory Board Member for PurePredictive.com Direct Sales

Moy M Chambers is CEO of Allspice Technologies, and Advisory Board Member for PurePredictive.com Direct Sales.

Mr. Chambers is a technology...

Read more Articles by Moy Chambers

Kenny Rawlins

Kenny Rawlins has been fortunate enough to have been around the network marketing industry his entire life and has experienced its power...

Read more Articles by Kenny Rawlins

Share Article

Be the first to Comment