Skip to main content

How the Remarkety recommendation engine works

By February 13, 2014June 15th, 2015Email Marketing

Remarkety is more than just an eCommerce email marketing automation platform. We have tools built into the platform that make it like having a consultant at your side.  For example, we constantly analyze your data and suggest what to do next. We call it Remarkety’s email marketing recommendation engine.

email_marketing_engine

New customers are usually curious how the recommendation engine actually works so here’s an in-depth perspective of our recommendation engine. Basically the recommendation engine has two purposes – to suggest new campaign types and to optimize existing campaigns.

 New campaign suggestions

By analyzing your data, the recommendation engine tries to find opportunities that are relevant to your online store. For example, if you are a relatively new store the engine will recognize that your data is not suitable for an “inactive customers” campaign, and will not suggest such campaign. But if you already have many inactive customers in your data, the engine may suggest launching a new “inactive customers” campaign, based a prioritization algorithm that considers the likelihood of success based on statistics gathered from thousands of other stores.

The example of the “inactive customers” campaign also illustrates how the recommendation engine identifies the opportunities themselves. Every type of business may have a different characterization of inactive customers. For example, if a consumer of ink cartridges has not purchased another cartridge in the past two months, should he/she be considered an inactive customer? Of course not. The recommendation engine analyzes all the historic data and establishes this characterization automatically. Along with the new inactive customers campaign recommendation, the recommendation engine also suggests the appropriate settings for the campaign (e.g. Sending the email to customers who have not purchased in the last 90 days).

The recommendations are sent by email as well as appear in the Remarkety dashboard. See below for an example.

email_marketing_recommendation

Fine tuning and optimization of existing campaigns

Remarkety’s recommendation engine also suggests methods for optimizing the current email remarketing campaigns that you have in place. The engine identifies optimization opportunities by combining several factors:

  • Best practices
  • Comparison between your performance metrics and the ones from other stores
  • Actual monitoring of the resultsof the various actions (what actually works and what doesn’t)

The recommendations can be related to any of the rules and settings of the campaigns as well as to the content of the email itself. For example, the recommendation engine monitors the length of the subject lines of the emails sent and recommends what works best for the specific type of campaign.

When applicable, you’ll see a small icon at the top right corner of an email campaign. This icon indicates there is an optimization recommendation in this campaign.email_analytics_kpis

After clicking the icon, you’ll see a more detailed explanation of the recommendation.email_subject_line_recommendation

I hope this helps clarify a bit of what is behind all the recommendations that we constantly push to our clients. Our statistics show that following these recommendations have significantly improved email performance. And by performance we mean actual orders and purchases, not just open and click rates.

If you haven’t started sending eCommerce emails to your customers with Remarkety yet – start now.

How to add product recommendations into your emails


Photo credit: Relish Tray Media