ActionNext 2.0: Making a Good Thing Even Better

Our award-winning predictive modeling engine just got an upgrade.

Test. Analyze. Optimize. Test. Analyze. Optimize….

As marketers and data scientists, this is how we roll. So when we launched our award-winning automated predictive modeling engine, ActionNext, we knew it was only the beginning. There would be more to come, and it’s here: ActionNext 2.0.

Here’s a refresher.

In today’s world, consumers have so many choices. So how do you know, out of all the products or services you have available, that you’re marketing the right ones to the right customers? We created the first iteration of ActionNext to answer that exact question for our clients.

We researched. We tested. And finally, we released ActionNext. With it, we developed highly accurate predictive models in just minutes. A process that would take days using more traditional methods. This allowed us to send targeted communications featuring the products, offers, or experiences customers were looking for, creating a connection between them and our clients’ brands while reducing marketing waste.

It was a hit. One retail client, in particular, tested a direct mailer with customized deals chosen with ActionNext vs. a one-size-fits-all piece, and the personalized version yielded a 30%+ lift in incremental revenue. How’s that for results?

ActionNext has quickly become one of our most popular offerings. But in true Havas Helia fashion, we got back to work. How could we make it even better? Even more accurate? Even more efficient?

Through extensive R&D, we found that embedding a new technique that helps model non-linear relationships between predictor variables and consumer choices led to as much as a 15% improvement in model accuracy. And this isn’t just with the samples we used for model development. We’re seeing the same lift with random hold-out samples. And all of this comes without compromising model development efficiency.

We can’t wait to see where this leads next.

The updated ActionNext automated predictive modeling engine is ready to help you personalize your customers’ experiences with your brand. Want to learn more? Email us at hello@havashelia.com.