October 26 - 28, 2020
Hilton Austin, TX
Reducing Buyers Remorse
In this exclusive presentation, Tim Sheppard from Samsung & John Sedej from OnProcess discuss reducing buyer’s remorse returns.
John Sedej: This is a picture of one of the customers we’re talking to. We asked them how his buyer’s remorse return rate was going, so we took a quick snapshot of them. You know, I’m John Sedej from OnProcess Technology. I’ll be joined by a colleague Dan Gettens and we’re very pleased to have Tim Sheppard here from Samsung, speaking. We’re–our–we’re kind of carrying on the theme from the panel discussion about how the importance of collaboration between the partners in the ecosystem to help reduce buyer’s remorse returns. Part of our this innovation spotlight is we want to share with you OnProcess’ methodology of combining our customer proactive outreach program with a set of predictive analytics modeling to reduce buyer’s remorse returns and with that I’ll turn it over to Tim to kick it off.
Tim Sheppard: Thank you. So hope you guys can hear me. Good morning. So we just got a quick slide here just to kind of set this up. So following on from the panel discussion the devils in the details of buyer’s remorse, just put it in perspective from a Samsung point of view. We have a pretty much a machine for launching products and we launch lot of different styles of technology. You guys have probably seen the watches coming out which should be very interesting from a buyer’s remorse point of view. But the fundamental thing here is about we launch about a new product a week in the US market. So we have a machine. We’ve got–you got to figure out the marketing, the other things, build the fundamental infrastructure to think about what’s actually happening from the consumer expectation point of view. So we worry a lot. And my point here is about the customer experience. When you’re producing that many products across multiple channels for multiple types of consumer, you’ve got to already think about what’s happening. We focused historically in two fold. One is the marketing setup. What is it we want to be as a brand? And I think we’ve changed a lot over the last couple of years. I think the next best thing campaign has resonated largely across the consumer base, most can identify Samsung has a point of view. We focused on–definitely on traditional product quality. Our return rates are very low from a basic fundamental quality off the buyer’s remorse experience. The opportunity really for us is to figure out what do we do next? What do we do in that 0-14 day period in the telecommunications industry where a consumer can return the device? The challenge really is there’s a lot of things we can do but we can’t do it by ourselves as a manufacturer. We have to collaborate with the channels, whether it’s national retail, whether it’s the carrier, whether it’s the channel in the supply chain. There are many, many touch points for the consumer. So partnership is very important. On the right-hand side of this slide we’re talking about areas that we can go focus very specifically from a tactical point of view. Branding, I will refer to UI/UX brands. We design and redesign extensively the UI. Most of our phones run on Google. We also have Windows phones. We also have traditional proprietary operating systems. Design enhancements and easy design, we think a lot about those issues and try to figure out how can we make the phone or the handset or the watch easier to use based on the experience we think we’re going for. So with that, please go on the detail.
John Sedej: All right. Thank you, Tim. What we talked about is, you hear a little bit about Nicky talking about the concierge service. What we have is the importance of a proactive customer outreach program to really collect that voice of the customer data to understand those causation factors that are really driving those buyer’s remorse returns and by gathering that voice of the customer codifying it quickly and understanding provides the ability to actively manage that user experience to understand what are the usability issues, what’s the frustrations and address those concerns before that customer makes that step like in a wireless within the 14-day time period before they come back and return the particular device. With that type of proactive outreach and the processes around it, yeah, you will be able to drive short-term, reduce the buyer’s remorse returns in a short-term by looking at certain segments of the customer base in addressing those buyer’s remorse returns, preventing them on a customer–customer basis. The challenge there economically is which customers you reach out to? How do you contact them? How do you concentrate so that your hit ratio will game it in such a way that your–when you reach out, you have a higher propensity to your touching the highest propensity to return customers and with that, I’ll turn it over to Dan. He’s going to talk about how OnProcess has developed a predictive analytics modelling which we call propensity return to target those customers selectively look out for those customer base.
Dan: As John said, if you want to reach out, you want to reach out to exactly that audience with the right message, concentrate the message and then aggregate the learning across your whole base. So that’s really where the predictive analytics comes into play. So the first step is to take your clients dataset. Most of these datasets as we heard earlier–rich datasets, the challenge is just to bring the data together to make it highly actionable and that’s where this PTR model propensity to return model comes in. So the first step is the model which all the data attributes, which is the client’s historical dataset and scores from a low–from a high propensity to return which is a 1, to a low propensity to return which is a 10. What we’re looking for is separation and this case, this case example, we got it. So the average returns rate was 10% in this case. But through the model, the first decile was highly concentrated. We got a great separation between the first decile which was returning at a 45% returns rate and the last which was only returning at 2% returns rate. So the obvious first place to go is to go after the biggest issue. That decile, it has a 45% returns rate, take the model the other aspect of the model, it helps us understand who they are, how they behave, how–why are the–why is that audience being undeserved. And then to take that data, apply the appropriate concentration of effort to that data, so as Tim said, with collaboration, with a little bit of analytics, with the concentration of effort, we can achieve what the panel talked about last time with the dramatic and very significant transformative improvement reduction in the remorse returns rate.
John Sedej: Thank you, Dan. And so conceptually this is how we put together the proactive outreach processes with the predictive analytics model. What Dan was showing is in that entire customer base, the average return rate was 10%. We’re able to concentrate and identify those customers have the highest propensity to return in 10% and put 45%. So every 1 out of 2 customers that we reach out to, that 10% sample base, were hitting those that have the highest propensity to return. So what we were able to do is after we build the predictive analytics model by product, by model or configuration, they were able to feed into that model the sales data and with the model able to score that particular customer and rank them where they fall within that scale, the highest propensity to return to the lowest propensity to return and coupled with the products information, understanding the underlining usability issues. Understanding the demographics of that particular customer, we’re able then to do a proactive outreach, selectively identify those customers, already in an armed fashion to know what those major issues are with the intent to address the issues before that customer gets frustrated and brings the device in. It’s not, you know, this is not a full proof, you’re not going to get–and the results were not going to say that every customer you touch are going to reduce return. But we’re looking at over 20 to, you know, 30 to 40% reduction in that return rate which makes a huge difference depending on the volume and the types of devices. There’s other uses of this model, too, is in cases in asset recovery were identifying on a B to B situation. Those customers have the highest propensity not to return. So you’re applying the predictive analytics for marketing purposes so you can segment your customer base, selectively do cross-sell, upsell, enhance service offering for selling them new warranty or extended program. So this ability to couple the predictive analytics model segregating the population, concentrating them but more importantly, affect the change that you want. In this case, we’re applying it with in the consumer electronic space to reduce those buyer’s remorse return rate.
So we promise to be done within 10 minutes, kind of following the theme here, working together, collaborating between other parties we think we can crack the code on buyer’s remorse return. All I could say it’s not all rocket science that definitely a part of it is rocket science and that’s the innovation part of the solution and presentation. Any questions?