What metrics should you track for a referral program?

Which referral metric matters most?

Referral conversion rate — the share of referred people who become customers. It isolates the quality of the mechanic from the size of your audience. Total referrals can rise purely because you emailed more people; conversion rate only rises when the offer, the landing page or the targeting genuinely improved.

• Participation rate — share of eligible customers who share at least once
• Referral rate — referrals sent per participating advocate
• Referral conversion rate — share of referred people who convert
• Referred-customer CAC — total reward cost divided by referred customers acquired
• Payback period — months for a referred customer's margin to cover their acquisition cost
• Outstanding reward liability — rewards earned but not yet redeemed
• Fraud rate — share of referrals rejected by screening

How do you calculate referred-customer CAC?

Divide the total cost of rewards paid — both sides, plus platform fees — by the number of referred customers acquired in the same period. Include the referee incentive, which programs routinely omit because it looks like a discount rather than an acquisition cost.

The common error is counting only the advocate's reward. In a double-sided program the referee incentive is just as much an acquisition cost, and excluding it can understate true CAC by roughly half.

Match the period properly too. Rewards are often paid weeks after the referral converts, so comparing this month's rewards against this month's signups mixes two different cohorts.

What is a good referral conversion rate?

There is no single benchmark that transfers across categories — conversion depends on price, purchase frequency and how much trust the product requires. Compare against your own other channels rather than a published industry figure, because those figures rarely state the definition or the sample behind them.

Any benchmark you find should be treated as directional at best. Programs define a 'referral' differently — some count a share, some a click, some a qualified signup — so two published rates are frequently measuring different things.

The comparison that does hold is internal: referred customers versus customers from your other channels, on conversion, retention and margin.

Why does reward liability need tracking?

Rewards earned but not yet redeemed are a real obligation sitting off most dashboards. If a program pays in credit or points, liability accrues continuously while redemption is unpredictable — so a successful quarter can create a cost that lands in a later one.

What is referral payback period?

The number of months it takes for a referred customer's gross margin to cover what you spent acquiring them. It is the metric that decides whether a program is affordable, because a referral can have excellent ROI on paper and still create a cash problem if payback runs long.

Calculate it as total reward cost per referred customer divided by their monthly gross margin. A $60 combined reward against $20 of monthly margin is a three-month payback.

Compare it against your other channels. A referral program that pays back in three months while paid search takes nine is a strong result even if the raw ROI numbers look similar.

How should referrals be attributed?

Attach a durable identifier at the click and carry it through signup and first purchase. Decide in advance how conflicts resolve — if someone clicks two different referral links, most programs credit the last touch before signup, but the rule must be written down before it is contested.

• Attach the identifier at click, not at signup
• Persist it across sessions and devices where you legitimately can
• Write down the conflict rule — usually last touch before signup
• Record the referral even when it does not convert, or you cannot measure conversion rate

How often should referral metrics be reviewed?

Monthly for the operating metrics and quarterly for the economics. Participation and conversion move fast enough to act on monthly; CAC, payback and referred-customer retention need a quarter of data before the numbers mean anything, because cohorts are small early on.

Which referral metrics belong on a leadership dashboard?

Three: referred-customer CAC against your blended CAC, payback period, and referred-customer retention against other channels. Those answer the only question leadership actually asks — whether this channel is cheaper and better than the alternatives — without drowning the answer in program mechanics.

What is a good referral participation rate?

Participation is usually the lowest number in the funnel and the one with the most headroom. Rather than chasing an external target, split it: what share of eligible customers saw the program at all, and of those, what share acted. The first is a distribution problem, the second a motivation problem.