I used to roll out the classic “£10 off for you and a friend” referral mechanic for all my clients because it’s simple, easy to communicate, and customers understand it instantly. But after a few programmes where it felt like we were systematically rewarding the wrong behaviour — customers recommending heavy spenders who then never returned, or worse, customers churning because they’d been “poached” by friends who only signed up for the discount — I changed my approach.
Replacing flat referral discounts with value-based credits isn’t just a tweak. It stops rewarding free riders, protects margin, and aligns incentives with the long-term value you want to drive. Below I share how I decide when to make that switch, practical mechanics that work for SMEs, and measurement guardrails so you don’t break acquisition while improving retention and LTV.
Why flat referral discounts can be a problem
Flat referral offers (e.g. £10 off per referral) are popular because they’re easy to understand and cheap to implement. But they create three common problems I’ve seen again and again:
Poaching highest-value customers: If your average order value (AOV) is £100 and you reward £10 per referral, you might be encouraging customers to bring in one-off high spenders who never return. You then lose margin without sustained revenue.Gaming and fraud: Flat amounts are easier to game with shared coupons, disposable emails, or friends who sign up just to get the voucher then disappear.Misaligned incentives: You want more repeat buyers and higher LTV, not just more first orders. Flat discounts reward the initial transaction only, not the behaviours you actually value.Those issues become painful when referral volume grows — suddenly your loyalty investment is shrinking margin without delivering better retention.
Signals that it’s time to replace flat discounts with value-based credits
Not every brand should flip overnight. Here are the signals I look for in client accounts before recommending the transition:
Low repeat rate among referred customers: If referred customers have a significantly lower repeat purchase rate than organic ones, that’s a red flag.High cost per new active customer: Calculate CAC (from referral) divided by first-year gross margin. If cost approaches or exceeds your target CAC for other channels, rethink the reward.Outsize coupon abuse: Increased instances of refund/return patterns, multiple sign-ups from the same IP, or unusually low average orders from referred cohorts.Margin pressure: If rising referral redemptions are eroding profit, especially in businesses with thin margins.Clear cohort LTV tracking: You can measure cohort repeat rates and lifetime value. If you can’t measure, that’s the first fix — don’t change rewards without the data.What I mean by value-based credits
Value-based credits are referral rewards that scale with the quality or value of the customer’s behaviour, rather than being a flat number. Examples:
Percentage of first order: Give the referrer 10% of the friend’s first order back as credit (capped to protect margin).Tiered credits: £5 credit if referred friend spends <£30, £15 if they spend £30–£80, £30 if they spend >£80.Credit conditional on second purchase: Award the referrer only after the referred customer makes two purchases (signalling they’re sticky).Points proportional to basket value: The referrer earns points equivalent to 5–10% of the friend’s spend, which can be redeemed later.These mechanics reward behaviour that correlates with higher lifetime value and discourage low-effort sign-ups.
Implementation blueprint for SMEs
When I help a small brand move from flat discounts to value-based credits, I follow a simple phased plan:
Phase 0 — Prepare the data: Build a referral cohort report (referred vs non-referred) with first order value, repeat rate at 30/90/365 days, and return/return-rate. If you can’t get this from your platform, do a manual sample.Phase 1 — Pilot selective value-based offers: Start with a small segment — say, customers in Tier A (repeat customers with high AOV) — or select one geographic region. Test percentage-based credits (e.g. 10% of friend’s first order) vs tiered credits.Phase 2 — Test time- and condition-based crediting: Run an A/B test where one cohort receives immediate credit on first purchase and the other receives credit only after the friend completes a second order within 90 days.Phase 3 — Roll out with safeguards: Implement caps (max credit per referral), expiry windows (e.g. credits expire in 12 months but are restricted to full-price items), and fraud detection rules (unique email verification, device/IP monitoring).Phase 4 — Continuous monitoring: Weekly referral volume, CAC per active referred customer, referred cohort repeat rate and average order value. Recalibrate percentages/caps quarterly.How to communicate the change without losing referral momentum
Switching mechanics risks confusing or disappointing customers. Here’s how I’ve handled the messaging:
Be transparent and positive: “We’re updating our referral rewards to give you credits that grow with the friends you bring — the more they shop, the more you earn.”Use examples: Show 2–3 scenarios: “If your friend spends £20 you earn £2 credit; if they spend £80 you earn £8 credit + a bonus £5 after their second purchase.”Frame as value enhancement: Position credits as more flexible and valuable in the long term compared to one-off discounts.Provide a grace period: Honor old codes for a short period or grandfather existing ongoing referrals, then migrate new referrals to the new system.Metrics to track — the ones that matter
When I switch mechanics I watch a handful of KPIs daily/weekly so I can act fast if something goes wrong:
Referral conversion rate: Percentage of referred visits that become customers.Referral-to-active ratio: Percentage of referred customers who make a second purchase within 90 days.Average earned credit per referrer: Helps gauge whether value-based credits push more long-term accrual or simply reduce immediate payout.Net promoter and churn: Referral programmes affect satisfaction. Track NPS and churn among both referrers and referred customers.Impact on margin: Referral cost as a percentage of gross margin per cohort.Real examples I’ve used
For a mid-sized online kitchenware brand I worked with, flat £10 referral vouchers were bringing in lots of first orders but a 25% repeat rate vs 45% for organic. We moved to a 7% credit on the friend’s first order, with an additional £5 bonus for the referrer if the friend ordered again within 60 days. Within three months referred cohort repeat rates rose by 12 percentage points and referral cost per active customer fell by 18%.
For a boutique hotel group, we tied credits to nights booked rather than flat discounts. That change cut abuse (multi-booking for single-night stays) and increased the average nights per booking for referred guests. The programme became more profitable and easier to justify to finance because credits were proportional to revenue and occupancy.
Common pitfalls and how to avoid them
Too-complex rules: If customers can’t explain the reward in one sentence, adoption will drop. Keep the headline simple and provide details for those who want them.No measurement plan: Don’t change without tracking. Set up cohort tracking before you flip the switch.Overly generous caps: Capping too high defeats the purpose; caps that are too low disincentivize big spenders. Model the economics first.Poor UX for credits: Make credits visible in the account, explain how they were earned, and show expiry. Confusion creates support tickets and churn.If you want, I can draft a simple experiment plan tailored to your data (what to test first, how to segment, and the exact KPI thresholds to guide a full rollout). I find a short, data-driven pilot removes most of the guesswork and keeps referral programmes delivering real value rather than just volume.
