A New Dawn for Retail Measurement – Facebook’s New Offline Conversions API and Store Visit Tracking

Executive Summary

On June 14, Facebook announced two new measurement products aimed at helping retailers understand the true value of their Facebook ads: the Offline Conversions API and Store Visit Tracking. These new ad products are not only exciting in and of themselves; they represent the future of retail advertising measurement.

The Offline Conversions API allows retailers to stream their in-store transaction data to Facebook in real time and map it to their Facebook ads. The challenge with this sort of online-to-offline measurement has always been marrying the identity of someone who sees an ad with the identity of an in-store customer; Facebook has addressed this issue by allowing retailers to either upload their own customer personally identifiable information (PII) or work with a partner to determine customer identity. Once the identity match has occurred, Facebook can determine the offline return on ad spend.

Store Visit Tracking is similarly oriented towards mapping offline activity back to an online ad exposure, but is focused on measuring foot traffic rather than transactions. In the absence of an in-store transaction, Facebook uses the location services feature on their users’ smartphones coupled with their knowledge of the physical boundaries of the advertiser’s business to determine if the user is visiting. Since not all users enable location services, Facebook uses polling to verify the visitation data it receives then extrapolates it to the general population to report store visit levels.

Facebook is making both of these products freely available to advertisers. Retailers using Facebook ads should request to be opted into the beta program for these products, and begin the process of integrating the Offline Conversions API into their point of sale (POS) systems and ensuring their business place data on Facebook is correct.

Offline Conversions API: History and More Detail

Facebook has been working on online-to-offline conversion measurement products for years (See Figure 1 below). The first Datalogix online-to-offline ROI measurement studies were conducted in 2013. Unlike many other forms of measurement, these studies used randomized controlled trial (RCT) methodology to determine the causal impact of media – this is the same method used to evaluate the effectiveness of drugs in a clinical setting, and is the best method for determining causal effectiveness currently available. Resolution Media has run more than 100 of these online-to-offline studies for our CPG and retail clients over the past few years, and they have consistently delivered strong results that prove the ROI value of Facebook.

While these online-to-offline ROI studies have seen their dollar minimums come down and utilization increase, the offering has remained out of reach for all but the biggest advertisers. But scale is baked deep into Facebook’s culture. The company doesn’t build ad products that it cannot ultimately deploy to its millions of  small and medium-sized business (SMB) advertisers. Online-to-offline measurement began as something that was expensive, manual, and slow: A minimum of $350,000 USD per campaign, labor-intensive setup by Facebook personnel to create holdouts, and a 90-day post-campaign waiting period as conversions were collected, analyzed, and compiled into a report. Like any new technology, online-to-offline measurement is progressing down the adoption curve; the Offline Conversions API is taking this form of measurement from expensive, manual, and slow to free, automated, and real-time.

Figure 1: A brief history of Facebook’s online-to-offline sales lift measurement offerings
Figure 1: A brief history of Facebook’s online-to-offline sales lift measurement offerings

Initial setup does require engineering work, as a client-side software developer needs to integrate the new Offline Conversions API into retailers’ POS systems. Facebook estimates it takes about half a day of labor for a skilled coder. Once that is done, conversion activity will stream into Facebook’s systems just as an online conversion event would, being mapped to paid media based on 1:1 user identity matching. Retailers can upload PII (email, phone, name, etc.) directly, or lacking that, can work with a data partner to determine who their purchasers are.

It’s important to remember that while this development is exciting and clearly the future of brick and mortar retail, it is an evolution in capability that really only benefits retailers; CPG advertisers will continue to rely on firms like Datalogix to report back on sales because retailers own all of the SKU-level data (meaning they know what products are purchased in each transaction). Resolution has long leveraged online-to-offline measurement to prove the value of our clients’ media investments, and this new capability represents the next step.

Store Visit Tracking: More Detail

Facebook enabled tracking of store visits in tandem with campaigns run against the Local Awareness objective. This allows businesses to see the foot traffic that results from their Facebook media. Here’s the methodology:

  • Business location boundaries: Facebook determines the footprint of business locations using satellite imagery.
  • Location services: When people have location services enabled on their devices, Facebook can count aggregate visits to business locations.
  • Polling: Facebook asks a sample of people it believes are at the brand’s business locations whether they are or not. Polling establishes the accuracy of measurement for a business and is used to adjust Facebook’s extrapolation model.
  • Extrapolation: Visit counts are extrapolated to represent the broader population based on Facebook’s models for measurability and polling data for retailer locations.
The methodology underpinning store visit tracking Source: Facebook Store Visits Materials
Figure 2: The methodology underpinning store visit tracking
Source: Facebook Store Visits Materials

Bear in mind that in major cities the business location boundaries, or “footprint” mapping accuracy will be poor, as multiple businesses will occupy a typical building in cities like Chicago, New York, or LA. The satellite imagery that Facebook relies on for mapping will best pick up big-box stores or fast food locations out in the suburbs: essentially businesses in their own freestanding structures. As a result, Facebook has restricted eligibility for the beta to advertisers with more than 100 retail locations outside of malls or other multi-level structures. Reporting will be available in real time in the Ads Manager UI.


Both the Offline Conversions API and Store Visit Tracking represent a tremendous step forward for retailers looking to track the offline impact of their Facebook media. And while online-to-offline sales lift measurement has been available for some time, this new capability represents a quantum improvement in timeliness, accuracy and cost. Eligible retailers should work with their agency partners and Facebook to gain access to the beta programs and begin onboarding these solutions.