Measuring affiliate value beyond clicks: FTD, NGR, LTV

An affiliate's value is the net revenue their referred players produce over time, minus what you paid to get them. Clicks and registrations tell you about reach. FTDs, deposits and NGR, tracked by cohort over months, tell you about value. The work is in measuring those consistently and being honest about which numbers are still moving.

This guide sets out the measures, how to compare like with like, and how to share the results with affiliates. It contains no industry benchmarks, because none we have seen are published with a method we could check. Where programme terms are cited, they were read in October 2026.

Why clicks are the wrong unit

A click costs the affiliate effort and costs you nothing. It says little about whether a person will register, deposit or stay. Two affiliates can send the same number of clicks and produce entirely different businesses: one sends visitors who are already looking to sign up, the other sends curious browsers.

Clicks are still useful for one thing: diagnosing the funnel. If clicks are healthy and registrations fall, something has changed between the link and your sign-up page, such as a broken landing page, a geo-block or a tracking fault.

The measures, in order

Two ratios matter more than any single count: registration to FTD, and FTD to qualifying player. A conversion rate that drops sharply at either step is a quality signal worth investigating before you look at volume.

From deposits to NGR

Deposits are not revenue. NGR is what is left after winnings and your deductions, and each programme defines those differently. In the eight programme definitions we read, all deduct winnings, bonuses or promotions, and gaming taxes or duties; most also deduct chargebacks and payment fees.

For measuring value, use the same NGR definition you pay revenue share on. If your internal finance view uses a different number from the affiliate's statement, the affiliate will eventually notice, and every value conversation after that starts from suspicion.

LTV: an estimate, labelled as one

LTV is the total net revenue a player produces over their whole relationship with the brand. You only know it for certain when the relationship ends. Before then it is a projection, and should be shown as one.

Two horizons are worth separating:

• Player LTV, the value of the player to you, however long they stay.
• The commission horizon, the period over which you pay the affiliate. This is set by your terms. Entain drops revenue share on the third anniversary of a customer's registration (clause 9.5); Rank pays until the fifth anniversary unless agreed otherwise (clause 7.1); 888 caps poker revenue share at 24 months (clause 5.8).

Comparing the two tells you whether a deal is priced sensibly. If most of a player's value arrives after commission stops, a revenue share affiliate is earning less from that player than the headline rate suggests.

Cohorts: compare like with like

A cohort is a group of players who joined in the same period. Grouping by affiliate and month of registration is the simplest way to compare value fairly, because it stops older, more mature players flattering one affiliate's numbers.

For example, take two affiliates who each sent 100 FTDs in January. Look at each group's cumulative NGR at one, three, six and twelve months after joining. One group might start strong and fade, the other start slowly and keep depositing. Monthly totals blur that difference; a cohort table shows it.

Keep cohort comparisons honest:

• Compare the same GEO and product. A sports cohort and a casino cohort behave differently.
• Compare cohorts of the same age. A three-month-old cohort cannot be judged against a twelve-month-old one.
• Note the cohort size. Small groups swing on one or two large players.
• Mark cohorts still inside a deposit or qualification window as incomplete.

Quality versus volume

Volume is easy to count and easy to reward. Quality takes longer to show and is what pays for the programme. The signals of quality are the ones that survive time and checks:

• FTDs that go on to qualify, rather than stopping at the minimum deposit
• Players still active in later months of the cohort
• Low rates of chargebacks, closed accounts and bonus abuse flags
• NGR that is positive across the cohort, not driven by one player

Programme terms show how much weight operators put on this. FDJ United (clause 6.7) and LeoVegas (clause 5.3.3) can withhold CPA for accounts identified as bonus abuse, suspended, closed for fraud or self-excluded. If quality problems change what you pay, they should change how you measure value too.

Affiliate view

We'd welcome being judged on quality, as long as we can see the same quality signals you see. If our players are flagged, we need to know which ones and why, or we can't fix the source.

Data freshness: when is a number final?

Most value figures move after you first report them. Know when each one settles:

• Deposit windows. A registration cohort is not complete until players have had the full window to deposit. bet365 allows six months, 888 allows 90 days for CPA trackers and LeoVegas 60 days.
• Fraud reviews. 888 may withhold commission during a fraud review of up to 180 days (clause 5.19).
• Chargebacks. Super Partners deducts chargebacks in the month they are charged, so an earlier month's value can fall later.
• Tracking loss. Safari caps cookies set by scripts to seven days (WebKit, ITP 2.1). Players who return after that may not be credited to anyone, so a cohort can undercount an affiliate's real contribution if you rely only on browser-side tracking.

The practical rule: show each number with the date it was calculated and whether it can still change.

Sharing value data honestly with affiliates

Affiliates cannot improve what they cannot see. Sharing value data also helps you, because a partner who sees their own cohort quality can shift their traffic towards what works.

What to share:

• Funnel counts per tracking link or campaign, so the affiliate can see which source performs
• Qualification outcomes, including how many players were withheld and the reason category
• Cohort NGR at fixed ages, with incomplete cohorts marked
• Any change to how you calculate a figure, before it appears in a statement

What not to do:

• Don't quote an LTV figure as fact when it's a projection
• Don't compare an affiliate to an undisclosed "programme average" they cannot check
• Don't change a definition mid-period without saying so

Operator view

Sharing cohort data does reveal something about our player economics. We can share it per affiliate, for their own players, without exposing the whole book.

What to do next

1. Pick one NGR definition and use it for both commission and value measurement.
2. Build a cohort view by affiliate, registration month, GEO and product.
3. Track the two key ratios: registration to FTD, and FTD to qualifying player.
4. Date every figure and mark which can still change.
5. Agree with each affiliate what value data they will see and how often. See What to report to affiliates.