CRM KPIs: how to measure success in online casino, sportsbetting, and e-commerce

The most useful CRM KPIs connect customer activity, retention, and contribution after costs. Start with a defined active-customer cohort, track how many return, then test whether campaigns create incremental profit. Opens and clicks diagnose delivery and response; they do not establish commercial impact.

This guide gives formulas for online casino, sportsbook, and e-commerce, with the denominator and cost basis each needs. There are no universal performance targets here: compare like-for-like cohorts and your own history rather than borrowing a benchmark with a different definition.

Quick reference: which CRM KPIs answer which question?

Use one reporting period, timezone, currency policy, and customer identifier throughout. In gaming, choose what "active" means: for example, at least one settled real-money bet. Keep deposit-active and login-active counts separate.

Rates with a zero denominator are undefined. A ratio using zero or negative revenue can be misleading even when technically calculable; show the underlying amounts and flag it. No single KPI proves that CRM caused a business-wide change.

What are turnover, deposits, and active customers?

Gaming turnover is the sum of eligible stakes, not deposits. Ten €10 bets create €100 turnover even if the same account balance is reused. State how voided, unsettled, and bonus-funded bets are treated.

Deposits are cash moving into player accounts. They are not operator revenue. Net deposits = deposits − withdrawals is a cash-flow measure; it does not reconcile to profit without balances, bets, settlements, and costs.

If you report net deposits / deposits × 100, label it a net-deposit ratio. A falling ratio is not automatically a retention problem, and encouraging fewer withdrawals is not a sound KPI objective.

In e-commerce, report gross sales and net sales separately. An order is the commercial event; cash receipts and refunds can land in different periods. Do not compare a gaming stake total directly with retail sales as if they represented the same economics.

Define active customers before any per-customer calculation. An e-commerce purchase-active cohort and an email-engaged audience are different populations. The same applies to bettors, depositors, and logged-in players.

How should you calculate GGR, NGR, and contribution?

For a simple settled-bet management view:

GGR = eligible stakes − corresponding player payouts.

Reconcile the stake and payout ledger, including returns of stake, voids, bonus play, and settlement timing. This management formula is not a substitute for regulatory reporting. Great Britain's Gambling Commission reporting guidance requires GGY rather than GGR and specifies its own calculation and bonus treatment.

NGR = GGR − the deductions included in your documented NGR definition. Those deductions may cover promotional costs, gaming duties, provider fees, or other agreed items. List them; do not assume every report uses the same definition. NGR is not automatically net profit.

Contribution = revenue − the variable costs included in the stated contribution definition. Keep fixed overhead visible when moving from campaign economics to programme or company profitability.

For e-commerce:

• Net sales: gross sales less the defined returns, refunds, and discounts, with tax treatment stated.
• Contribution: net sales less included product, payment, fulfilment, and other variable costs.
• Profit: requires the remaining relevant costs too; net sales alone is not profit.

Finance should approve the mapping before CRM compares campaigns across products.

Must GB gambling operators report GGY rather than GGR to the Gambling Commission?

Yes. The UK Gambling Commission's regulatory returns guidance, updated 15 September 2026, says it requires operators to report gross gambling yield (GGY) rather than gross gambling revenue (GGR) in regulatory returns.

How does the Gambling Commission calculate gross gambling yield for GB operators?

The Commission's GGY guidance, updated 15 September 2026, gives GGY as A + B − C, where A is stakes, B is other amounts accruing to the operator and C is prizes. Use that for Great Britain regulatory returns, not the management GGR above.

How do you interpret casino and sportsbook margins?

Realised gaming margin (%) = GGR / corresponding turnover × 100. Both figures must use the same scope and settlement basis.

Theoretical casino edge = 1 − theoretical RTP, expressed as proportions. A constructed example with 96% theoretical RTP has a 4% theoretical edge. This does not promise 4% revenue in a particular session or campaign.

The Gambling Commission's RTP calculation guidance explains why actual results depend on play volume and volatility. Compare realised performance with the actual games and configurations played, rather than a generic industry margin.

In sportsbook, keep odds overround, modelled expected margin, and realised hold distinct. For a market of mutually exclusive, exhaustive outcomes, overround is the sum of implied probabilities minus 100%. It describes the quoted prices, not guaranteed realised profit. Bet mix, settlement outcomes, and exposure matter.

For e-commerce, a comparable decision uses gross or contribution margin on net sales, with the cost definition shown. A positive revenue response can still lose money when discounts, product cost, or fulfilment absorb the return.

Does the GB RTP tolerance narrow as play volume rises?

Yes. The UK Gambling Commission's RTP guidance, checked 30 October 2026, says actual RTP is measured against the game's calculated figures and "as the volume of play increases the tolerance will decrease". The volatility of the game must also be considered.

How should you measure bonus and discount costs?

Use recognised economic cost, reconciled with finance, rather than assuming every issued bonus costs its face value or every unredeemed offer costs nothing. Track issued, redeemed, expired, and recognised amounts separately where your system supports them.

Bonus-to-GGR (%) = recognised bonus cost / GGR × 100. You can also report bonus-to-NGR, but label it separately: a denominator already reduced by bonuses will produce a different ratio.

Constructed example: €20,000 GGR and €4,000 recognised bonus cost give a 20% bonus-to-GGR ratio. If the operator's NGR definition subtracts that bonus cost and €3,000 of other deductions, NGR is €13,000. Bonus-to-NGR is about 30.8%. Neither ratio alone says whether the offer created incremental profit.

Do not infer bonus profitability from wagering requirements multiplied by theoretical edge. Completion, cash-out, eligibility, game mix, and the campaign's effect on activity also matter. The bonus-conversion explanation provides a separate view of those mechanics.

In e-commerce, show discount cost alongside returns, product margin, and delivery subsidies. Compare contribution with and without the promotion; a higher order count is not enough.

Which KPIs measure acquisition, retention, and customer value?

Conversion: name both ends of the funnel

Registration-to-first-deposit conversion = registrants who first deposit within the defined window / eligible registrants × 100. Visitor-to-registration and visitor-to-first-deposit are separate rates. For e-commerce, distinguish session-to-order from customer repeat-purchase conversion.

Specify acquisition cohort and observation window so recent registrants are not compared with customers who have had much longer to convert.

Retention and churn: follow the same people

Let A be last month's active customers and B this month's active customers.

Retention = count of customers in both A and B / count of A × 100.

Churn = count of customers in A but not B / count of A × 100.

These sum to 100% only for that same cohort, activity definition, and pair of windows. Churn here means inactivity in B, not permanent departure.

Constructed example: 1,000 players are active in September. Of those, 600 are active in October. Another 500 new or previously dormant players are active in October. Retention is 60% and churn is 40%. Total actives grew from 1,000 to 1,100, but 110% is an active-base index, not retention.

For onboarding, define a registration or first-purchase cohort and measure return in a specified later window. "Day-30 retention" must say whether it means activity on day 30, during a surrounding interval, or any time up to day 30.

Reactivation: distinguish observed return from campaign impact

Observed reactivation rate = eligible dormant customers who return / eligible dormant customers at assignment × 100. Define dormant, qualifying return, and the outcome window.

Some customers return without a campaign. Treatment rate minus control rate gives the estimated incremental change in percentage points, under a valid experiment. Do not call the treatment's whole return rate campaign-caused. The player retention and churn guide shows how these rates fit into early-life, at-risk, and reactivation programmes.

Acquisition cost and lifetime value: align the economics

CAC = included acquisition costs / new paying customers attributed to those costs. State whether this is marginal or fully loaded, what spend period it covers, and how delayed conversions are handled. Do not automatically assign all retention staff and platform costs to acquisition.

For planning, define contribution LTV as expected cumulative customer contribution over a stated horizon, with discounting and cost assumptions disclosed. Revenue multiplied by an assumed lifespan can conceal changes in retention and margin. Observed cohort contribution is historical; projected lifetime value is a model.

Compare CAC with contribution LTV on compatible cost bases and include payback time. There is no universal LTV:CAC ratio that proves an operator is profitable.

How do you measure incremental CRM campaign ROI?

Use a randomised eligible audience, a concurrent control, a predefined outcome, and the same window. Include everyone assigned, including nonresponders.

Incremental contribution before campaign costs = (mean treatment contribution − mean control contribution) × treatment count.

ROI (%) = (incremental contribution before campaign costs − incremental campaign costs) / incremental campaign costs × 100.

Constructed example: 900 assigned treatment players average €50 of contribution before the evaluated campaign costs. A 100-player control averages €35 on that same basis. Estimated lift is (€50 − €35) × 900 = €13,500. With €5,000 of incremental promotional and execution cost, estimated profit is €8,500 and ROI is 170%.

If the contribution measure already deducts the same promotional costs, do not subtract them twice. Show uncertainty before calling the campaign a winner. The identical logic applies to e-commerce contribution after product and fulfilment costs.

The full CRM campaign ROI guide covers assignment, unequal groups, outliers, theoretical margin, and why adding overlapping campaign lifts can overstate programme value.

Which channel metrics help diagnose performance?

Document what the provider counts, including automated activity and deduplication rules. Do not apply a universal open, click, or unsubscribe threshold across different send types and providers.

Pair reach and response with incremental contribution and player/customer experience. A cheaper channel can still be inefficient if it adds no lift. Attribution after a click can help trace a journey, but it does not establish what would have happened without the message.

How should the KPI set change by business model?

Use the player lifecycle guide and segmentation framework to choose coherent cohorts. Split commercial inactivity from promotional eligibility and player-protection restrictions.

How do you turn the dashboard into a decision?

Start with one question, not every available metric. A useful CRM review connects the observation, explanation, and next action:

1. Observation: cohort retention fell while overall active count rose.
2. Check: confirm the activity definition, acquisition mix, and data completeness.
3. Diagnosis: inspect onboarding by product, market, and acquisition cohort.
4. Action: test a specific journey or fix a confirmed product/payment problem.
5. Decision: compare incremental contribution and guardrails over the agreed window.

For every KPI, record its owner, source, formula, population, period, exclusions, and refresh time. Keep those definitions alongside the dashboard so a changed query cannot silently rewrite the trend.

If finance, CRM, and the platform report different values, reconcile the underlying definitions before setting targets. Player data analytics support addresses that measurement gap; a CRM Growth Audit connects it to lifecycle and campaign decisions.

Key numbers

Regulatory definitions quoted from primary sources, each checked on 30 October 2026.

How to cite this page: EngageHut, "CRM KPIs: how to measure success in online casino, sportsbetting, and e-commerce", engagehut.com, https://engagehut.com/insights/crm-and-kpis-how-to-measure-success-in-online-casino-sportsbetting-and-e-commerce/, updated 30 October 2026.

See the CRM Growth Audit→