Before using an industry figure as a target, check whether the research measures the same behavior, among comparable people, over a comparable period.
When a team asks whether its loyalty program performs well, an external benchmark seems like a useful starting point. The difficulty is deciding what deserves comparison.
A membership figure, a participation rate and a measure of customer preference describe different things. Similar labels and percentages often make the differences less visible.
I would begin with the decision the benchmark needs to support. A team assessing market adoption needs different evidence from a team reviewing first-purchase activation or reward redemption.
Read the population before the percentage
Mando and YouGov’s Understanding Loyalty in Europe 4.0 overview reports 63% of adults across 24 European markets belong to at least one loyalty program. This describes membership across the surveyed markets. [3]
The figure does not provide a participation target for an individual retailer. The retailer’s customers form a different population, and joining one particular program is a different outcome from belonging to any program.
Deloitte’s research introduces another distinction. Its survey covered 5,564 US adult loyalty members. Several questions, including behavioral impact and redemption, concerned respondents’ preferred program in an assigned industry. Deloitte explicitly cautions against generalising those findings across all memberships. [2]
A result about someone’s preferred program needs this context when used to assess your entire membership base, including occasional and inactive members.
Neither source becomes less useful because of these limits. The limits tell you which decisions the research supports.
The denominator changes the story
Consider two programs reporting redemption.
Program A has 10,000 registered members. During the period, 2,000 hold an available reward and 1,000 redeem. Program B has 10,000 registered members, 5,000 with an available reward and 1,500 redeemers.
These figures are illustrative.
Measured against all registered members, Program B has the higher redemption rate: 15%, compared with 10%.
Measured against members with an available reward, Program A has the higher rate: 50%, compared with 30%.
Both calculations are valid. They answer different questions.
The first describes how much of the registered base redeemed. The second describes redemption among members with an available reward. Program B reaches more members with reward eligibility, but a smaller share of those eligible members redeem.
Even here, the comparison needs consistent treatment of eligibility. Someone receiving a reward on the last day of the period has had less opportunity to use the benefit than someone receiving a reward three months earlier.
A benchmark labelled “redemption rate” leaves too much unresolved unless the denominator and observation period accompany the figure.
Decide what deserves comparison
For an external comparison, I would look for a similar market, sector, purchase cycle and program maturity. A new program has a different member mix from one operating for ten years. A weekly shopping habit produces different opportunities from an annual purchase.
Research method matters too. A customer recalling their behavior in a survey provides different evidence from recorded transactions. Both have uses, but combining them into one performance ranking hides the difference.
The same applies to collection dates. Publication year alone does not tell you when respondents answered or when transactions occurred.
Where a source omits a definition essential to the comparison, I would keep the figure as background context. Giving an uncertain comparison a precise target creates more confidence than the evidence supports.
Use your own baseline alongside the market
An external benchmark helps identify questions. Your own consistently measured results help evaluate progress.
For example, a team reviewing activation should follow new members over equal periods after joining. Comparing this month’s joiners with last quarter’s joiners before allowing the same observation time creates an avoidable distortion.
Once the baseline is clear, the team has a stronger basis for setting a target linked to a specific change. A revised joining journey should have an expected effect, a measure and a review period. The industry number helps challenge the ambition without replacing the business case.
For each benchmark in a management presentation, I would include a short definition beside the figure: who was measured, what counted, when the measurement happened and how the data was collected.
If the team cannot explain why the comparison fits, I would leave the number out of the target-setting discussion. A smaller set of comparable measures gives the team more useful direction than a long list of impressive percentages.
References
[2] Deloitte. Reshaping loyalty programs in an era of value seeking. 12 January 2026. Survey methodology and endnote 4. View source
[3] Mando and YouGov. Understanding Loyalty in Europe 4.0. Publisher overview, 2026. View source