Deep-dive guide
Loyalty and Cashback Cost Simulator: Repeat Lift vs Reward Cost and Breakage
Executive industry analysis
Loyalty programmes are attractive because they promise repeat customers, which are cheaper than new ones. A cashback or points scheme gives members a reason to return, and a return is worth the margin on another order. The risk is that the reward is paid on every member order, not only on the extra ones. A member who would have bought anyway collects the same reward as one who came back because of it. At four percent, a reward can look small, yet it may absorb a large share of the margin the programme creates.
The structure of the problem is the gap between two groups of orders. The programme earns margin only on the extra orders it creates. It pays rewards on all member orders, extra and ordinary. Whether it works therefore depends on the repeat lift, the margin on each order and the size of the reward. Breakage helps, because unredeemed rewards cost nothing, but it is a double-edged sword if members feel cheated. Platform fees and abuse add a smaller layer of cost.
The domino cascade begins with the extra orders created by the programme. These earn extra margin at your gross margin. The programme cost is then subtracted, made up of the rewards redeemed, the platform fee and the abuse. The ratio of cost to extra margin shows how much of the benefit is spent on getting it. When the ratio passes one hundred percent, the last box turns negative, because the programme costs more than it earns.
The iceberg separates the reward most people calculate, paid on the extra orders, from the larger hidden reward paid on the orders that would have happened anyway. Below the waterline are also the loyalty software and messaging, which is charged per member whether or not they buy, and the abuse and goodwill that leak out of every programme. In many programmes the hidden cost is several times the visible one. Use the simulator to find the lift you need to justify the reward rate.
Formula and methodology blueprint
The model works per month and measures the programme only. Members place a baseline number of orders each month. The repeat lift adds extra orders on top. Margin is earned on the extra orders. Rewards are paid on all member orders at the reward rate, but only the redeemed share costs money after breakage. A per-member platform fee and a small abuse share are added. Net result is extra margin minus programme cost.
- Extra orders
- E = members × baseline orders × lift
- Orders that exist only because of the programme.
- Extra margin
- M = E × AOV × margin
- Gross profit on the orders the programme creates.
- Rewards redeemed
- R = (baseline + E) × AOV × reward rate × (1 − breakage)
- Cost of rewards actually used, paid on every member order.
- Programme cost
- C = R + platform fee × members + abuse × R
- Rewards, software and leakage together.
- Net result
- Net = M − C
- What the programme earns or loses in the month.
- Cost to margin
- C / M
- How much of the extra margin the programme itself consumes.
Cascading elasticity
The repeat lift is the engine: it is the only thing that earns margin. The reward rate is the brake, because it is paid on the full order base, not on the extra orders. Raising the reward rate by one point costs more than raising the lift by one point earns when the base is large. Breakage cuts the cost of every reward by its share, but pushing it too high turns the programme into a source of complaints.
Three real-world case studies
Bootstrapped starter
A small shop with a simple points scheme
A small shop with a few hundred members offers a modest reward and sees a good repeat lift. Platform cost is small and the rewards are well covered by the margin. Load it to see how a simple scheme with a low rate pays for itself.
Scaling mid-market
A large programme with tiers and expiry
A large retailer runs a tiered programme with expiry, which lifts breakage and keeps the base rate low. The repeat lift is strong at scale. Load it to see how breakage and tiers protect the margin as the member base grows.
Black-swan crisis
A generous scheme with no real lift
A store offers a large cashback to a big list of members who mostly buy cheap items anyway. Few extra orders appear and almost every reward is redeemed. Load it to see the programme cost far exceed the extra margin.
Industry benchmark matrix
| Metric | Danger zone | Industry average | Top 10% elite |
|---|---|---|---|
| Reward rate on member orders | Above 8% | 2–5% | 1–3% with tiers |
| Repeat-order lift from members | Below 5% | 10–20% | Above 25% |
| Breakage of earned rewards | Below 5% or above 50% | 15–30% | 20–35% with fair rules |
| Programme cost as share of extra margin | Above 100% | 50–80% | Below 40% |
| Platform cost per member per month | Above 0.5 per member | 0.1–0.3 | Below 0.1 |
Variables and levers
Active loyalty members
Members who can earn and use rewards in a typical month.
Lever: Grow the number of members who really buy again, not sign-ups, because inactive members add platform cost and no orders.
Average order value
What a member order is worth before rewards are redeemed.
Lever: Set a minimum spend to earn rewards, which raises the basket and reduces the cost of rewards on small orders.
Gross margin
Share of the order value left after the cost of goods, before rewards.
Lever: Fund rewards from higher-margin categories first, so a reward never exceeds the profit on the order.
Repeat-order lift
Extra orders from members, as a share of the orders they would place without the programme.
Lever: Reward the second and third purchase more than the first, because a repeat habit is the only source of income.
Reward rate
Cashback or points value earned on each order, as a percentage of its value.
Lever: Lower the base rate and use tiers, because the whole reward base is paid on every member order.
Breakage
Share of earned rewards that expire or are never redeemed.
Lever: Add fair expiry windows and clear reminders, so rewards are used within a period without feeling like a trap.
How to use this simulator
- 1Enter active members, average order value and gross margin.
- 2Estimate the repeat lift from a holdout comparison and set the reward rate and breakage.
- 3Read the five boxes to see whether the cost to margin ratio turns red first.
- 4Press Find my breaking point to learn the reward rate or lift that cancels the profit.
- 5Compare the current programme with tiered rewards and expiry over 24 months.
Mistakes to avoid
- Counting every member order as a result of the programme. Only the extra orders earn margin, but rewards are paid on all of them.
- Counting sign-ups as members. Inactive members cost platform fees and produce no orders.
- Raising breakage with strict expiry only. Members who feel cheated leave, and the lift disappears.
- Funding the same reward on every category. A reward on a thin-margin item can exceed the profit on the order.
Frequently asked questions
Do loyalty programmes always pay?
No. They pay when the repeat lift is high enough that the margin on the extra orders exceeds the rewards paid on all orders. A generous scheme for customers who buy anyway usually loses money.
What is breakage?
It is the share of earned rewards that are never redeemed, for example because they expire or members forget them. It lowers the cost of the programme, but fair rules matter for trust.
Should rewards be cash or points?
Both work. Points are often cheaper in practice because they are redeemed on future purchases and breakage is higher, while cash is simpler. Check local rules on how rewards must be treated.
How do I measure the lift?
Compare members with a similar group of non-members or with a holdout group over several months. Comparing before and after is distorted by seasons and campaigns.
Is this financial advice?
No. It is an educational model with simplified assumptions, such as a fixed baseline order rate. Accounting and consumer rules for rewards vary, so check current rules and replace the inputs with your own data.
This is an educational planning model with stated assumptions. It is not financial, tax or legal advice. Replace the inputs with your own figures and verify important decisions with a qualified professional.