Grocery case study · Transaction Matching

Turning 43.8 million anonymous transactions into real people.

A fast-growing national grocery chain has no traditional loyalty program, no app, and no customer opt-in. Roughly 95% of its in-store transactions carried no identity at all. Transaction Matching resolved about 7 in 10 of them to known individuals.

Company
National grocery chain
Primary solution
Transaction Matching
Client goal
People-based marketing without a loyalty program
32M
Transactions resolved in a single month, in one production run.
~70%
Resolution rate: anonymous transactions matched to a real, known individual.
3×
Sequential match passes: loyalty signal, card-level, store catchment.
43.8M+
In-store transactions processed every month.
The challenge

95% anonymous, 0% actionable

This chain is one of the fastest-growing grocery retailers in the United States, and one of the most data-challenged. As a low-friction, no-frills grocery model, it does not have a traditional loyalty program. That means the overwhelming majority of its in-store transactions are anonymous: a basket of groceries, a payment, and nothing else.

The chain processes over 43 million in-store transactions every month. Before Deep Sync, approximately 95% of those transactions had no customer identity attached. They were commercially invisible. The retailer could not personalize, suppress, retarget, or attribute media spend to in-store revenue.

The standard industry answer, launching a loyalty program, is a multi-year, multi-million-dollar undertaking.

The solution

Resolve identity from the transaction signal itself

Deep Sync's Transaction Matching is a fundamentally different answer: resolve the identity of the buyer using the transaction signal itself, with no customer opt-in required. Three sequential passes run against the full monthly POS file, each capturing what the previous pass missed.

The three-step resolution process
  1. 01

    Loyalty signal match

    For the small share of transactions where loyalty data exists (card swipe, app scan), Deep Sync uses that PII anchor to resolve identity first. High-confidence, exact matches.

    ~9.6Mtransactions resolved
  2. 02

    Card & payment signal match

    Deep Sync matches payment card signals against our identity graph, resolving transactions that have no loyalty data but do have card-level signals, the vast majority of grocery POS events.

    ~3.3Mincremental transactions
  3. 03

    Store catchment match

    For remaining unmatched transactions, Deep Sync uses store-level geographic catchment modeling combined with household identity data to probabilistically resolve the buyer. This is the step that catches what pure exact-match methods miss.

    ~18.1Mincremental transactions
The results

The grocery identity waterfall

Deep Sync processes the full monthly POS file, tens of millions of raw transactions, in a single production run, with no loyalty program required.

Monthly resolution · what the three passes add up to
~32M
Transactions resolved
~64–73% of monthly volume, matched to known individuals.
~11.8M
Unmatched
The remaining 27%, still anonymous after all three passes.
The payoff

What the chain can now do that it couldn't before

Personalize

Reach specific shoppers with relevant offers based on resolved purchase history and basket composition, even without a loyalty program.

Suppress

Exclude existing customers from acquisition campaigns, reducing wasted media spend and improving the cost efficiency of new customer acquisition.

Retarget

Build addressable audiences from resolved in-store buyers and serve them ads across Google, Meta, and CTV, connecting digital spend to physical store behavior.

Attribute

Close the loop between digital ad exposure and in-store purchase, without a loyalty program, making true closed-loop attribution possible at grocery scale.

Why this is Deep Sync's key differentiator

Transaction matching is not a feature most identity vendors offer. It requires a proprietary matching methodology, a deep consumer identity graph, and the operational infrastructure to run monthly production pipelines against tens of millions of records.

“Transaction matching doesn't just improve data quality. It fundamentally changes what the retailer can do with marketing: personalization, suppression, attribution, retargeting, capabilities that were impossible before because 95% of their customers were completely invisible. By resolving roughly 70% of those buyers into known, addressable individuals, Deep Sync transformed invisible purchase activity into measurable customer intelligence.”

CH
Director, Partner Development, Deep Sync
What it proves

You don't need a loyalty program to market like you have one

The chain resolved most of its monthly volume without asking a single customer to sign up. If your stores run on anonymous transactions, the same waterfall applies to your file.

Most of your revenue is happening in the dark

Send us your transaction file. We run a free test against our identity graph and show you how much of your anonymous in-store volume resolves to real, addressable people.