Build a Smart Spree Shop Using Machine Learning

David Jones · Jul 25, 2014 · 2 min read · Share:

Machine Learning (ML) can improve your customers’ experience, increase conversion rates, and boost sales across channels. Although we focus on Spree Commerce, ML is relevant to any eCommerce site with at least a few hundred products and a bit of historical data.

Learning from Amazon & Netflix

What if you could understand your customer’s behavior based on past purchases and browsing patterns? What if your site could predict which products he or she is predisposed to buy?

Amazon.com and Netflix have been doing it for years. It’s called “predictive analytics” and is one of the many benefits machine learning can provide your eCommerce business.

We know that if you show your customers the right content at the right time they’re more likely to take the action you want. For a shopping site that means click on the “add to cart” button and complete a purchase.

Offering Personalized Recommendations

What is the right content? The fact is every visitor has different goals, desires and motivations. ML algorithms continuously adapt to your customers' behavior to provide relevant, personalized product recommendations. This individualized approach improves the visitor experience. It results in up-selling and cross-selling opportunities that in turn lead to higher conversion rates and larger average order sizes.

Can Machine Learning Help my Spree Store?

You don’t have to be a Fortune 500 company to benefit from machine learning. ML is especially effective on sites with at least a few hundred products, where subtle differences between them may not be immediately obvious. The more history you have, the better. Ideally each product will have been purchased and/or rated at least ten times.

It comes down to this: can visitors easily discover products they want or do they need help finding new merchandise?

A recommendation engine sorts through all the options to deliver specific product suggestions for each customer based on their purchase history and preferences.

Limitless Possibilities

In a future blog post we'll reveal the preliminary results of a recommendation engine using Prediction.IO. But there are so many other things you can do that could have a huge impact on ROI.

  • Promote the highest margin products to increase revenue
  • Send personalised recommendations via email marketing
  • Use recommendations to tailor search results
  • Up-sell through the check-out process
  • Integrate with your CRM to provide real-time suggestions for your sales team


"Build a Smart Spree Shop Using Machine Learning" was written by David Jones.


Join The Conversation

Share and start a conversation about this post: