To Measure Hybrid Transactions, Opted-In Location Data Is Getting A Dose Of AI

Retracing the customer journey from impression to transaction is the holy grail of marketing attribution.

But consumer behavior is messy. In today’s hybrid retail environment, customers can window shop in person or comparison shop online before completing their purchase in a store or on the internet. Or they partake in “buy online, pick up in store” (BOPIS).

The need to capture the customer’s journey between online and offline behavior was the seed for Foursquare’s new Closed Loop feature, the location data platform’s latest addition to its attribution product, said James Kung, Foursquare’s senior director of product management.

The product upgrade represents Foursquare’s expansion into attribution for online and hybrid conversions to give marketers perspective on the wider consumer journey.

Holistic campaign overview

The Closed Loop tool is designed to make it easier to analyze hybrid conversion events.

When a customer buys a product online, then visits a physical store for curbside pickup, or visits a physical store after using an online store locator tool or researching a particular promotion, marketers can see that information in the tool’s dashboard, which is refreshed every couple of days.

By integrating information from first-party and third-party tracking pixels tied to a marketer’s ad campaign with Foursquare’s user panel data (from its owned apps as well as SDK data sources), Foursquare can show marketers if someone who was exposed to an ad campaign completed a transaction. These closed-loop customer journeys are then measured against an AI-powered baseline model to determine the actual impact the campaign had on completed transactions, whether they were in-store or online.

The user interface displays campaign metrics like number of impressions served, unique users reached, message frequency, ad spend, number of conversions and cost per action. And it is suited to measuring omnichannel campaigns across online mediums like websites, apps, CTV, social media, video games, streaming audio, podcasts and digital out of home, as well as offline mediums like billboards and linear TV.

Each individual publisher or platform running the campaign, as well as each DSP serving the programmatic display portion of a campaign, gets its own tab breaking down the advertiser’s return on investment.

The goal is for customers to use this dashboard to optimize toward partners delivering higher conversion rates or lower cost per actions, Kung said.

But Foursquare is also tackling upper-funnel metrics, such as brand lift. It extrapolates consumer behavior patterns from its end-user panels.

A machine-learning model uses the panel data to predict what real user behavior would be across a wider, hypothetical audience; this predictive data set serves as the baseline of precampaign user behavior. The marketer measures brand lift by analyzing the difference between the actual outcomes driven by the campaign and the predicted outcomes of the baseline model, Kung said.

Location data requires special privacy considerations. Instead of tracking location across the board, Foursquare relies on an opted-in panel where the users agree to share data. Foursquare has a data supplier review program to assess its partners’ data-gathering practices for compliance with privacy legislation and best practices.

The data comes from more than one place. Foursquare’s user panels are built using data from a variety of sources, including Foursquare’s owned apps, such as its city guide and its location data platform, plus third-party apps that use Foursquare’s SDK. (As a reminder, Foursquare bought location attribution company Placed from Snap back in 2019.)

For users in the opted-in panel who visit a store, for example, “We can see that activity for days prior to the ad impression being served. We also see their online activity, including clicks on a particular store locator page or clicks on the online order action,” Kung said. That activity calibrates Foursquare’s baselines for hybrid conversions that include online and offline activity.

The Closed Loop feature was tested during a beta release in Q1 2022 by an undisclosed number of Foursquare’s advertiser partners from a variety of verticals, including QSR (quick-service restaurants) and retail. The feature is now available for all customers.

This article was corrected to remove a reference to Placed SDK being used as a data source in the Closed Loop Attribution feature.

Enjoying this content?

Sign up to be an AdExchanger Member today and get unlimited access to articles like this, plus proprietary data and research, conference discounts, on-demand access to event content, and more!

Join Today!