How should we measure loyalty program profit?
Our answer
Measure loyalty profit as incremental gross profit against a random holdout not offered the program. Subtract rewards at cost, outstanding points and platform and staff costs. Treat any gap between members and non-members as an upper bound.
- Confidence
- Medium
- Strongest proof
- 1 dataset
- 5 practitioners, 1 opinion or claim
- Sources
- 7 sources
- 7 statements
LRGLJOLFB
New here? Take the 1-minute tour
Why this is the answer
3 reasons and our judgmentMembers were already better customers, so only a random holdout shows what the program caused (Gladly, Loyalty Juggernaut, Loyalty & Reward Co).
Points are a cost the day you issue them. In Eightx's example, a 5% earn rate on $4M of repeat revenue issues $200K a year, $150K after 25% breakage, and needs about a 13% lift to break even.
The raw gap between members and non-members overstates what the program causes. Smile.io, a loyalty app maker, finds members out-spend non-members by 72% at the median, and says itself that engaged customers self-select.
Our judgment
Comparing everyone offered the program with a random holdout, and treating any gap between members and non-members as an upper bound, is our synthesis of the sources. Set the break-even lift before launch by dividing expected program cost by your margin, as in Eightx's example, where about $524K of incremental repeat revenue is needed. A small store that cannot power a holdout should use join-date or first-redemption comparisons and weigh the platform fee against the repeat revenue at stake, since Smile.io says 2 points of repeat share may not cover the subscription at starter volume.
Our judgment, built from Gladly, Loyalty & Reward Co, Loyalty Juggernaut, Open Loyalty, Eightx and Smile.io: our reading of these sources, not a finding from any one of them.
What the sources say
7 statements from 7 sourcesOpen Loyalty, which sells a loyalty platform, says measure loyalty ROI on revenue lift against a control group, with points liability counted as cost. In one client audit a 34% gap shrank to 12%.
“Once we matched cohorts on prior spend and tenure, the true incremental lift was closer to 12 percent.”
Gladly says to measure loyalty program return against a random holdout of eligible customers, since rewards also pay for spending that would happen anyway. In its made-up example, the holdout cut apparent lift 40%.
“Hold out customers at the point they’d be invited or eligible, not after they’ve signed up.”
A loyalty consultancy says member versus non-member spend gaps overstate a program's effect; in one Dutch grocery study, membership explained 4.1 points of a 29-point gap. It recommends a random holdout of about 5%.
“Any ROI figure resting on a member versus non-member comparison is better treated as an upper bound than an estimate.”
Loyalty Juggernaut, which sells a loyalty platform, defines loyalty ROI as incremental profit over net cost, measured against what members would have done anyway. It advises a holdout for each bonus event. No data.
“Loyalty program ROI is incremental profit over net cost, where incremental is measured against what members would have done anyway”
Eightx models a 5% loyalty earn rate on $4M of repeat revenue: about $150,000 net cost, so the program needs about a 13% lift to break even. It says vendor lift figures are self-selected.
“If you cannot measure that lift, you are not running a loyalty program, you are running a discount you forgot to price.”
A study of 21,805 Czech ecommerce customers found paid loyalty members ordered more often and bought more items after joining, but did not spend more overall. It compared members only with themselves.
“these behavioral changes are not reflected in their overall expenditure”
Smile.io, which sells loyalty software, says loyalty members outspend non-members by 72% across 10,766 programs, but notes best customers self-select, so measure incremental repeat revenue.
“engaged customers self-select. Joining a program and redeeming rewards are marks of your best customers, not only causes of them.”
In the full answer
For readers who sign in- What to do this week3 steps
- The numbers behind it3 figures from cited sources
- Where each source standsA chart placing each source on the question
- The strongest disagreementThe best case against the answer, from a cited source
- How sure we areWhat its confidence rests on, and what would change the answer