How We Raised Click through Rate by 50% with a Simple Change

By on March 14th, 2015 in Machine Learning
TrueAccord Blog

The technical and analytical vision behind TrueAccord is to add data-driven decisions to the communication model in debt colelction. Digital communication enables better data collection, and better understanding of customer behavior patterns. We can collect and observe open, click and browsing patterns that sometimes do more to explain how to engage with a customer than explicit communication. By connecting customer behavior to their mental state, and responding to that state with corrent language, we were able to substantially increase engagement on our collection communication.

Continue reading “How We Raised Click through Rate by 50% with a Simple Change”

Using a complex model for measuring success in interactions with customers

By on March 4th, 2015 in Machine Learning

The TrueAccord product can be compared to an automated marketing and sales campaign, focused on identifying payment intent and acting on it. The system classifies customers (we use “customer” to refer to those in debt) by their most probable reason for non-payment, and the “voice” we think is going to drive that to action. Then, it has to decide, out of the hundreds of content items we have, which goes out to which customer, through what channel, and when. Of course, we’d like to learn from history and send the combination most likely to succeed. This raises the important question: what’s “success” in our context?

Continue reading “Using a complex model for measuring success in interactions with customers”

Beware Of The Disposable-Business Customer

By on June 20th, 2014 in Machine Learning
TrueAccord Blog

disposable-business-customers I’d like to share a disturbing phenomenon we’ve observed in our data, one that probably impacts many of you who provide services to online businesses.

When a new customer is entered into our system, we crawl for additional information about them on the web as well as look for them elsewhere in our system. As we grow and see more companies and the people who owe them money, we start to see chains of debts that are linked to one another. Deeper linking checks revealed that the individuals connected with these debts owned multiple websites, some of them dysfunctional, and often with web hits that indicated some kind of foul play.

As we dug deeper, the picture cleared. There is a large number of individuals, operating online, starting and closing businesses in a highly irresponsible manner bordering on fraud. These individuals, by and large (as far as we can see) unrelated to one another, tend to serially open businesses with hardly defensible business models, contract the majority of the work for them to third parties, charge unsuspecting consumers large amount of money, and move on to the next business. We are unable to determine, at this time, whether this is pure fraudulent behavior or just extreme irresponsibility, but the pattern exists. We see it across fashion, insurance, financial planning and several other services.

As a business owner, you should be aware of this pattern, especially if you are providing services to online businesses. Design, eCommerce, Hosting and other related services are especially vulnerable since they are key in starting a new online property. TrueAccord customers can ask for access to our Data Furnishing program, that can provide data to help identify these individuals before they are able to do business with you. We recommend that, before you take on new customers for large projects, you run a manual search of their web history, previous businesses, and other owned domains. You may (or may not) be surprised to discover a few of those in your own customer roster.