Case study · Card-linked offers · Retail

Most Offers Can’t Prove Anything. This One Returned 428%.

A leading national liquor retailer ran a single card-linked offer on the PokitPal platform. Three months in, the users who could access it were spending 29% more with the retailer, taking roughly one extra shopping trip every seven weeks, and spending 31% more across every other merchant in the network. Return on investment: 428%.

Results at a glance

• +29% Spend with the retailer vs users not eligible for the offer
• 428% Return on investment On a 30% gross margin assumption
• 2.71 vs 2.11 Shopping trips per month Eligible vs non-eligible users
• +31% Spend across the rest of the network vs users who did not engage

Figures cover the first three months of a campaign that is still running. All spend is observed at the transaction level.

The challenge: offers are easy to run and hard to prove

Any retailer can run an offer and count redemptions. Far fewer can answer the question the finance team actually asks: would that spend have happened anyway?

Redemption volume does not answer it. Attributed sales do not answer it either, because attribution tells you which customers touched the offer, not which customers changed their behaviour because of it. A campaign can look excellent on both measures while delivering close to nothing in incremental revenue.

This campaign was built to answer the harder question. Rather than reporting activity, PokitPal set out to isolate the offer’s incremental effect: the behaviour change attributable to the offer itself, measured against what comparable users did without it.

Why the network makes the measurement possible

PokitPal operates a card-linked offer network. Offers from participating merchants are surfaced to users through participating publishers, and because each offer is linked to the user’s payment card, the resulting spend is observed at the transaction level rather than modelled or estimated.

For this campaign, the liquor retailer ran a single offer on the platform. Some users on the network became eligible for it. Others did not. Both groups carried on shopping as normal, with the retailer and with every other merchant in the network.

That structure is what makes the result readable. The offer runs as a genuine commercial campaign, and the network produces a clean comparison group as a by-product of how eligibility is distributed. No holdback had to be engineered, no panel had to be recruited, and no survey had to be fielded.

How the impact was measured

The results

At the retailer

Spend: +29%. Eligible users spent 29% more with the liquor retailer than non-eligible users across the three-month period.

Shopping frequency: 2.71 trips per month, against 2.11. Eligible users averaged 2.71 shopping trips a month. Non-eligible users averaged 2.11. That is roughly 0.6 of an additional trip per user per month, or about one extra visit every seven weeks, sustained across the base.

ROI: 428%. Return on investment for the retailer, calculated on a 30% gross margin assumption.

Across the network

Spend at other participating merchants: +31%. Among users who engaged with the retailer’s offer, spend across the rest of the network rose 31%.

No lift among users who did not engage. This is the control that makes the previous number credible. If simply being on the platform lifted spending, the non-engaged group would have moved too. It did not. The network effect tracks engagement with the offer, not exposure to the platform.

What the data tells us: frequency, not basket size

The composition of the lift matters as much as its size, and this is where the campaign gets genuinely interesting.

The increase came almost entirely from more frequent visits. Basket size stayed virtually flat. That distinction separates two very different commercial outcomes.

• A lift driven by larger baskets can reflect purchases pulled forward, or customers stockpiling. It borrows from next month.
• A lift driven by frequency reflects a change in how often the customer chooses that retailer. It is the behaviour retailers are actually trying to buy.

This campaign delivered the second kind. In a repeat-purchase category, that is the difference between a promotion and a habit.

The frequency lift was also growing month over month across the measured period. On that basis, PokitPal expects incremental upside to keep rising as the campaign continues. To be clear about what that is: an observed trend and a forward expectation, not a result already banked.

Beyond one merchant: the 31% nobody plans for

The network figure is the part of this campaign that is easiest to overlook and hardest to replicate anywhere else.

Users who engaged with the liquor retailer’s offer did not simply shift spend towards that retailer. They increased spend across other participating merchants as well. One strong, relevant offer raised engagement and relevance across the whole network, not only for the merchant funding it.

That changes the unit of value on both sides of the network.

For publishers and media networks, a single well-matched offer improves the performance of the inventory around it. Offer quality stops being purely a merchant concern and becomes a yield lever.

For merchants, joining a network where other participants are running strong offers is itself commercially useful. You benefit from your own offer, and from theirs.

The business takeaway

Across the first three months of a live campaign, one offer on the PokitPal platform produced:

• 29% higher spend with the liquor retailer
• 428% ROI on a 30% gross margin assumption
• A frequency-led lift of 2.71 against 2.11 shopping trips per month
• 31% higher spend across the rest of the network among users who engaged

Every one of those figures rests on a comparison against users who did not receive or act on the offer, measured at the transaction level. That is the difference between reporting activity and demonstrating incremental behaviour change.