Introducing New Predictive Metrics to Analyze Store Visit Behavior

Introducing New Predictive Metrics to Analyze Store Visit Behavior

By Rachel Wolan | News and Events | 05 December, 2017

Euclid leverages proprietary machine learning and large data set to predict which customers will visit a physical location

Since Euclid’s founding, our mission has been to map the physical world. As the foundational layer of the offline world, Euclid’s network has seen 100s of millions of unique devices and trillions of data points collected solely at commercial locations in 2017 alone.

Earlier in 2017, we launched new ways to identify visitors and classify visitors, as inside and outside your location and as your customers or staff. As a next step, we leveraging our large data footprint to predict specifically which visitors will be visiting your commercial location soon.

Today, I’m excited to announce Likelihood to Visit in the Next 30 Days (L30), a powerful predictive metric to help clients target visitors who are likely to visit their physical location in the next 30 days. Another way to interpret L30 is as a score for someone’s propensity to visit, even if they don’t.

Online marketers have been using predictive metrics for years to predict where and why users click. Financial analysts have used forecasting to predict purchase patterns. But until today, marketers and analysts alike have been limited to using clicks and anonymous footfall to predict who will visit.

The L30 machine learning model makes predictions based on key data points from identified visitors including number of visits, days elapsed since visit, visit duration, number of locations visited and more. The L30 machine learning model learns which behaviors predict future visits by being trained on historical visit and behavioral data. Euclid expects the predictions to continue to improve with more time and data.

Practically speaking, what does that mean?

Marketers are unlocking the power of L30 in many innovative ways including:

  1. A large national retailer is targeting visitors with a low likelihood to visit in the next 30 days and incentivizing them with a higher-than-average coupon that expires in 14 days.
  2. A fast food chain is targeting visitors with a medium likelihood to visit at lunch in the next 30 days with their loyalty program and giving them a free coffee for the month of December when they visit for breakfast instead of lunch.

Likelihood of Next Visit

L30 is available immediately to clients in the Euclid dashboard, and will be available soon in all major integrations including Salesforce, MailChimp, Oracle, AgilOne and more.

Check out the announcement to learn more about L30 and the future of predictive metrics at Euclid.

Euclid Careers

Tags
Rachel Wolan

Rachel Wolan

Rachel is the VP of Product at Euclid. She is a seasoned product executive and brings over 15 years of experience in B2B SaaS product, engineering and analytics.

More posts by Rachel Wolan