How MentSpot made it's product more engaging by implementing a recommender model

Dig into engagement

Mentspot is a platform that connects people who need advice with those willing to give it. It's an innovative way to get help and learn from others without ever leaving the comfort of your couch.

Mentspot's Product Manager wants to provide users with a platform that is both engaging and intuitive. They hope to keep people coming back for more and using the platform on daily basis.

Engagement is a key factor in retaining customers. We use the Aha-moment template to examines what leads to week-2 retention.

It gives insight into what behavior will lead to users coming back. This gives product managers the ability to optimize their products for that customer behavior.

Exploring the patterns

In just a minute, we get to know that in order for users to keep coming back, they need to view at least 4 mentor profiles in the first four hours.

If you're able to help them with this, then you're on track for higher retention and engagement!

The idea is to optimize the listing of mentors and get users to view more mentor profiles so that they are more relevant to the user currently visiting the website.

We do this by implementing a recommender model that will recommend the most relevant mentor profiles for conversion based on time of day, week, device, location, and other factors.

Creating your recommender model

It's easy as pie! All you have to do is run the "online recommender" Template. Simply answer a few questions about what attribute of the "view mentor" event we want to show recommendations on, and voila! The model will be built for you just like that!

Integration

The last step is Implementation. Luckily, it's super easy! Just click the "Deploy" button on the recommender results to put your recommender modal into production.

Once the recommender model is deployed, you can simply click a button to get code snippets that are ready to be copy-pasted onto your website and you're ready to go!

A/B testing

We want to test if the recommender model is really improving retention, or if it's hurting it by accident. This we do by using the A/B testing template and choosing the retention experimentation goal.

The results show the impact of using a recommender model to find mentors. When 50% of customers saw results recommended by the personalized model, their conversion rate was 18% better than when they were shown randomly selected mentors! 🎉

Have you ever wished for a better way to retain your customers and increase user engagement? Well, with this powerful model, you'll be able to do just that! With a few clicks of the mouse, you can apply this new idea and watch it work its magic.

In no time at all, you'll see an improvement in retention rate and user engagement too! Simply connect your data and get blown away by the results! 🤯