Decisions · Strategy and governance
How should you design loyalty for a low-frequency category
For a low-frequency category, do not build a points programme. Build a lifecycle membership with product-linked service benefits and a fixed annual touchpoint. The main condition is purchase interval longer than one year: below that, points may work, but above it the design must sustain memory and preference across the gap.
The decision
The decision is whether to copy a frequency-based points engine or to build a lifecycle membership around the long gap between purchases. In a low-frequency category, the member may buy once a year, once every five years, or once a decade. Points do not change that interval; they merely sit in an account and decay in relevance. The mechanism that actually works is to attach benefits to the product's service lifecycle, then use an annual maintenance or check-in benefit to create a legitimate reason to contact the member without asking for a sale.
That is a different engineering target. A points ledger needs transaction ingestion, earning rules, expiry logic and redemption catalogues. A lifecycle membership needs a product register, a service calendar, and a small set of benefits that expire at the next purchase rather than on a calendar date. The first is designed to reward frequency; the second is designed to reduce the mental cost of staying with you until the next need arises.
What it costs you
The recommended lifecycle membership costs you committed service capacity and a visible annual liability, not just points that may never be redeemed. An illustrative calculation: suppose the product sells for 1,000 dollars every five years and the alternative points design pays 5 per cent back, or 50 dollars per cycle, which is five sixths of a dollar per month over the 60-month gap. If you instead promise an annual service credit of 30 dollars, the five-year liability is 150 dollars per retained member. This looks larger, but the credit replaces the competitive offers you would otherwise make to win back attention. If two competitor offers of 40 dollars each would have been used, the incremental cost is 150 dollars minus 80 dollars, or 70 dollars per retained member. That 70 dollars buys a relationship across the full gap, while the 50 dollars in points buys very little between purchases.
How to decide
Test your own programme with two numbers. First, write down the median time between purchases for one product line in months. Second, write down the cash value of the benefit a member would actually use during that gap. If the median interval is above 12 months and the benefit value before the next purchase is below the cost of one service call, skip points. Instead make a list of the three service tasks that occur most often between purchases, attach a fixed benefit to the two that most influence the next buying decision, and set every benefit to reset at purchase, not at the end of a calendar year. Then ask five of your best members to describe the benefit from memory. If fewer than three can do so without prompting, simplify until they can.
What would change our mind
We would change this recommendation if the category develops a genuine adjacent purchase stream that occurs at least four times a year, such as consumables, accessories, or paid software tied to the durable product. In that case a points ledger becomes worth its overhead, because the member can earn and redeem inside a 90-day horizon and the programme no longer has to carry attention across years. We would also flip if the product itself moves to a subscription or service model, where the low-frequency purchase is replaced by a recurring billing relationship and the design problem shifts from memory to monthly value.