Cost Per FTD and Clipping: What an Operator Can Honestly Measure

Almost every iGaming buyer asks the same question inside the first ten minutes: what is the cost per FTD. It is the correct question, it is the one organic clipping cannot answer directly, and that gap is exactly why the answers on offer tend to be wrong. This page is what the metric actually measures, which link in the chain clipping breaks, what the arithmetic really depends on, and what an operator can measure instead without pretending.

If you have not scoped a campaign yet, start with how clipping works for iGaming operators and what a clipping agency costs. This page assumes you are past both and now have to defend the spend to a team that funds channels on cost per FTD and nothing else.

TL;DR: Cost Per FTD and Clipping in Seven Lines

The argument compressed, before the mechanics.

• FTD means first time deposit. Cost per FTD is spend in a window divided by the new depositors attributed to that spend.
• The vertical buys on it because everything else is worse. Registrations are noisy, lifetime value is slow, and affiliate contracts are already written in FTDs.
• The metric is an attribution product, not a measurement. It only exists where a tracked click connects the impression to the deposit.
• Clips do not carry that click. Most viewers never tap anything, gambling links are restricted on several platforms, and the deposit often happens later on another device.
• So clipping donates its conversions. Under last click, the better a clipping campaign performs, the better your branded search looks.
• Any quoted cost per FTD is an assumed conversion rate wearing a number. Change the assumption slightly and the answer moves by an order of magnitude.
• Measure it at the account level instead. Hold a market, run a defined window, read blended cost per FTD and demand signals against it.

FTD stands for first time deposit, and by extension first time depositor. It is the event where a registered account funds itself for the first time. Cost per FTD is then simple arithmetic: the marketing spend in a period divided by the number of first deposits attributed to that spend.

The reason it sits at the center of iGaming and not, say, retail, is that the gap between a signup and a customer is unusually wide here. A registration can be a bonus hunter, an abandoned know your customer check, a duplicate account, or a curious person who never returns. None of those cost anything to produce, which means a channel optimized on registrations can be made to look excellent while delivering nothing.

A deposit is different. Someone moved money. It is the earliest point in the funnel that correlates with the number the business actually runs on, which is net gaming revenue over the player's lifetime. FTD is not the goal. It is the earliest available proxy for the goal, and the vertical has standardized on it for that reason.

Why the Whole Vertical Buys on This One Number

Three pressures push every acquisition conversation back to the same metric, and it is worth being explicit about them, because they also explain why a clipping proposal gets an unusually hostile reception.

That third point matters more than the other two. A large share of iGaming acquisition runs through affiliates on a cost per acquisition deal, where the acquisition being paid for is literally a first deposit. Once a finance team has a stack of contracts priced that way, every new channel gets held to the same unit whether or not the unit fits.

Clipping does not fit. Not because it performs badly, but because it produces a different kind of evidence. Refusing to admit that up front is how these engagements go wrong.

Every one of those four links is somewhere attribution can leak, and it leaks constantly even for paid media. The difference is that paid media and affiliate traffic are engineered so the chain mostly holds. Trackers, parameters and postbacks exist for no other reason.

Where Clipping Breaks That Chain

A clip is a piece of content in a feed, not a placement with a destination. That single difference cascades through every link.

The last row is the one worth sitting with. It is not that clipping goes unmeasured. It is that clipping's conversions get measured and then assigned to somebody else.

A person sees three clips over a fortnight, remembers the brand, types it into a search engine, clicks the brand's own search ad, registers and deposits. Last click credits paid search. The clipping campaign that manufactured the demand shows nothing, and the harvest channel that intercepted it books a flattering cost per FTD. Under that model, the better a clipping campaign performs, the better your branded search looks and the worse the case for clipping becomes.

This is a general problem for upper funnel spend, not a quirk of clipping. It just bites harder here, because clipping's whole delivery mechanism is designed to look like it is not advertising.

Cost per FTD from a clipping campaign is the cost of a thousand views, multiplied by however many thousands of views it takes to produce one deposit. The first half is a market rate you can negotiate and verify. The second half is a number that does not exist anywhere until after the campaign has run, and cannot be observed cleanly even then, for all the reasons above.

Take a hypothetical operator buying at two dollars per thousand verified views, a figure comfortably inside the range Forbes describes. Hold that media cost completely fixed and vary only the conversion assumption.

The media price never moved. The assumption did all the work, and it swung the answer by a factor of forty. Those four rows are illustrations, not findings, and that is precisely the point: any vendor who hands you a cost per FTD before a campaign has run has picked one of those rows for you and not told you which.

Ask them directly what view to deposit rate they assumed and where it came from. The answer is the whole proposal. If it came from a different operator, in a different market, with a different product, brand recognition, bonus offer and deposit flow, it is not a benchmark. It is a borrowed guess.

What You Can Measure Instead, Honestly

Giving up on channel level cost per FTD is not the same as giving up on measurement. It means moving the measurement up a level, from the channel to the account, where the number is still real.

The geo holdout is the strongest instrument available and the one most operators skip, usually because holding a market out feels like leaving money on the table. It is worth the cost once. It is the only method here that produces a causal read rather than a correlation, and it answers the question the finance team is actually asking, which is whether the account gets more depositors when this line item exists.

Be honest about its limits too. Markets are never perfectly matched, seasonality does not respect your test design, and a regulatory change or a competitor's campaign in one market ruins the comparison. If you cannot hold a market, holding weeks is the fallback: run, pause, run, and read the shape. That is weaker, and it is still better than an attributed number invented from an assumed conversion rate.

Blended cost per FTD is the one that actually gets budget renewed. It is unglamorous, it gives no channel a trophy, and it cannot be gamed by moving credit around, because there is no credit to move. It is one division across the entire account.

What to Settle Before Anything Goes Live

Most of the argument about measurement happens after the campaign, which is far too late. Nearly all of it can be settled in the scoping call.

One item deserves its own paragraph. If a clipping vendor guarantees a cost per FTD, ask how they intend to deliver it, because there are only three routes. They can control the conversion path, which organic distribution by definition does not. They can buy traffic that converts on paper, which is the failure mode view verification exists to catch. Or they can claim last click credit for demand they did not create, which is free money for them and a distorted picture for you.

A guarantee here is not confidence. It is a statement about which of those three is happening. Our wider list of clipping agency red flags covers the rest of the pattern.

When Cost Per FTD Should Rule Clipping Out

Sometimes the honest answer is that this channel is not for you, and it is cheaper to hear it now than after a quarter of arguing about a report.

• Your model only funds deposit attributable channels. If there is no internal appetite for a market test, clipping will fail review regardless of how it performs. Do not buy it and then litigate the measurement.
• You need depositors this month against a hard efficiency target. Harvest channels move faster. Clipping builds the demand those channels harvest, on a slower clock.
• You cannot hold anything flat. No held market, no quiet week, constant promotional changes. You will never be able to read the result, so you are buying on faith.
• Your brand name is ambiguous or hard to spell. The fallback path from a clip is search or direct entry. If that path is broken, the demand a clip creates leaks to whoever owns the name.
• Your compliance position cannot survive it. Geo control and licensing constraints come first. Our iGaming clipping guide covers what the brief has to say before any of this measurement discussion matters.

None of that means clipping does not work in this vertical. It means the metric your organization runs on cannot see it directly, and you have to decide whether you are willing to measure it a different way. That decision is a real one and it belongs to you, not to a vendor's dashboard.

Frequently Asked Questions

What does FTD mean in iGaming marketing?

FTD stands for first time deposit, and it is also used to describe the player who makes one, the first time depositor. It marks the moment a registered account funds itself with real money for the first time. Cost per FTD is the marketing spend in a period divided by the number of first deposits attributed to it, and it is the standard efficiency unit across the vertical because a registration on its own costs nothing to produce and predicts very little.

Can you measure cost per FTD from a clipping campaign?

Not directly, and anyone claiming otherwise is describing an assumption rather than a measurement. Cost per FTD is reconstructed from a chain that starts with a tracked click, and clips do not carry one. What you can measure is the effect on the whole account: run the campaign in defined markets over a defined window with everything else held flat, then read blended cost per FTD, registrations by market, branded search and direct traffic against a held out comparison.

What is a good cost per FTD?

There is no portable answer, and a vendor who offers one without asking about your product is guessing. The figure that matters is whether the cost of acquiring a depositor sits comfortably below what that depositor is worth to you over their lifetime, and that varies by market, license, game mix, bonus structure and retention. A number borrowed from another operator's account tells you about their economics, not yours.

Why do clipping campaigns get no credit in last click attribution?

Because a clip almost never sits at the end of the journey. Someone watches clips in a feed, remembers the brand, and later searches for it or types the domain directly, so the final click belongs to branded search or direct traffic. Last click models award the whole conversion to that final step. The practical consequence is that a clipping campaign which is working will often show up as an improvement in your search and direct numbers rather than in its own.

Should an operator pay a clipping agency on a cost per FTD basis?

It sounds like the safest possible deal and it usually is not. To guarantee a cost per FTD a supplier has to either control the conversion path, which organic distribution does not, buy traffic that converts on paper, or take last click credit for demand it did not create. Billing on verified views keeps the unit auditable, keeps the incentive on real distribution, and leaves the deposit level question where it belongs, which is in your own account level measurement.

How do you run a geo holdout test for a clipping campaign?

Pick two comparable sets of markets, run clipping in one and nothing in the other, and hold every other line of spend flat in both for the duration. Run long enough to cover the lag between a view and a deposit rather than reading it in the first week. Then compare first deposits and blended cost per FTD across the two groups. It is imperfect, since markets never match exactly and seasonality does not cooperate, but it is the only method here that produces a causal read instead of a correlation.

Written by

Arian Saffar

Founder of ClipUp. Runs clipping campaigns for brands, podcasts, studios, games and labels.