Cost per reactivated player: the reactivation ROI maths

Most "reactivation ROI" content stops at a sentence: compare cost per reactivation with your cost of acquisition. That's the right comparison. But the number is easy to get wrong in three ways, and each one makes a campaign look better than it was:

1. Counting players who would have come back anyway. Without a holdout, every organic returner is credited to the campaign.
2. Leaving out bonus cost. Call cost is visible on an invoice. Bonus cost sits in a different budget, but it's usually the bigger number.
3. Ignoring unanswered attempts. Most dials don't reach a person. On per-hour labour they cost as much as a conversation. On per-minute AI billing they cost little, but not always nothing.

This guide works through the full calculation once, with real Dial Milo prices and clearly made-up campaign numbers, so you can rerun it with your own.

The cost per reactivated player formula

Cost per reactivated player = (call cost + bonus cost) ÷ incremental reactivated players

Incremental reactivated players = depositors in called group − (called group size × holdout deposit rate)

Reactivation ROI = (incremental players × NGR per reactivated player − total cost) ÷ total cost

Define "reactivated" before the campaign starts: a real-money deposit within a fixed window, for example 30 days after the first call attempt. Use the same window for the holdout. The definitions behind each term are in iGaming retention KPIs.

What a call actually costs

An outbound Dial Milo call has two parts, as set out in how calls are priced:

• The phone line, priced for the exact number dialled and your caller ID country, billed per started minute. A 61-second call is billed as two minutes. A UK mobile called from a UK number starts at 8.5¢ a minute. A Netherlands mobile is 10¢. A US mobile or landline starts at 4¢. A few destinations add a fixed fee per call.
• The AI, at 9.57¢ for each minute Milo is talking with someone.

So a UK mobile works out at roughly 18¢ a minute all-in. Calls that never connect, such as ring-outs or switched-off phones, have no minute charge. Voicemail is different. The carrier's voicemail answers the call, so it connects. Milo detects the voicemail greeting and hangs up without leaving a message, but that started minute still counts. You can price any specific number at /rates.

To keep the example on the cautious side, we round every call up to whole minutes for both line and AI:

Worked example: 10,000 lapsed UK players

All campaign figures below are illustrative. Prices are Dial Milo's as of November 2026. The arithmetic is shown so you can swap in your own numbers.

1. 1

   Remove everyone you may not call

   Say you have 10,000 players with no deposit in 60 days. 4,000 (40%) have phone-marketing consent for casino under the UKGC's per-channel rule and pass your self-exclusion, harm and age filters. The rules behind that filter are in UK Gambling Commission rules for marketing calls.

2. 2

   Hold out 10%

   400 eligible players are set aside at random and not contacted by this campaign. 3,600 are called.

3. 3

   Dial twice, four hours apart

   First attempt: 792 answer (22%). The other 2,808 get a second try four hours later, and 337 more answer (12%). That's 1,129 players reached, from 6,408 attempts. Of the 5,279 unanswered attempts, 2,903 (55%) reach voicemail and 2,376 ring out.

4. 4

   Count the offers and deposits

   169 of the 1,129 reached players accept the offer (15%), and 105 of them deposit and claim it within 30 days (62%). Across the whole called group, including people who declined or weren't reached, 216 players deposit within 30 days: 6.0%.

5. 5

   Check the holdout

   10 of the 400 held-out players deposit in the same 30 days: 2.5%. Without the campaign you'd expect 3,600 × 2.5% = 90 of the called group to deposit anyway. Incremental reactivated players = 216 − 90 = 126.

The costs

The result

Two things stand out. The naive figure understates the real cost by 42%, because 90 of the 216 depositors were coming back anyway. And calls are only 30% of total cost. The bonus is the other 70%.

Sensitivity: what moves the number

Holding call cost at $1,136.60 and bonus claims at 105, here's cost per incremental reactivated player as the incremental lift and the average bonus change:

Read across a row and the bonus does most of the work. Read down a column and targeting does. The call itself barely registers: with a $10 bonus at 3.5 points of lift, halving the call cost would only take cost per player from $17.35 to about $12.84.

That suggests where to spend effort:

• Call the segments with the highest lift, not the biggest segments. Measure lift per segment, by dormancy age and past value, and stop calling the segments where the called group does no better than its holdout. The dormant player segmentation guide and template help you split the list.
• Test a smaller offer, or none. A check-in call with no bonus may bring some players back on its own. A script without an offer, A/B tested against one with an offer, tells you how much of the lift the bonus is actually buying. Dial Milo lets you A/B test up to five scripts side by side and compares cost per depositor once deposit data is connected.
• Stay inside the bonus rules. In Great Britain, wagering requirements have been capped at 10x since January 2026, which changes the real cost of a given headline bonus. Price offers on what they actually cost, not their face value.

Set the ceiling before you dial

A cost per reactivated player only means something next to what a reactivated player is worth. Work out the ceiling before the campaign, not after:

Maximum cost per reactivated player = NGR per reactivated player over your window × the share of it you're willing to spend

With the illustrative $110 of 90-day NGR and a rule that a campaign may spend at most half of what it brings back, the ceiling is $55. The worked example's $29.85 clears it comfortably. The 2-point, $50-bonus corner of the sensitivity table, at $88.70, doesn't.

The same numbers give you a break-even lift. The campaign costs $3,761.60 and each incremental player brings back $110, so you need 3,761.60 ÷ 110 ≈ 34 incremental players to break even. Out of 3,600 called, that's a lift of under one percentage point over the holdout. A segment that can't beat its holdout by even that much shouldn't be called again with the same offer.

Two costs are easy to leave out of the total:

• Follow-up messages. If an accepted offer triggers an SMS and an email with the bonus code from your journey tool, those sends belong in the campaign's cost.
• Team time. Building the list, approving the script and reviewing safer-gambling flags takes hours. It's usually small next to bonus cost, but count it for the first campaign so you know.

The same campaign with a human team

For comparison, the reactivation ROI calculator puts AI and a human call centre side by side. A human agent is paid for every dial, answered or not. Using the calculator's starting assumptions, which are placeholders to replace with your own, of $25 per agent hour and 30 dials per hour:

6,408 dials ÷ 30 per hour = 213.6 agent hours × $25 = $5,340

With the same $2,625 of bonuses and the same 126 incremental players, that's $7,965 in total, or $63.21 per incremental reactivated player against $29.85. The difference is almost all the 5,279 unanswered attempts, which are close to free on per-minute billing and cost a full slot of agent time on per-hour billing.

A human team still wins in places: VIP players, complex complaints, anyone who needs a judgement call. A sensible split is AI for the long tail of lapsed players and your own hosts for the top tier.

Run it on your own list

1. 1

   Price your actual numbers

   Check your markets and caller ID countries at /rates, then load the list into the spend calculator to see attempts, minutes and cost before you spend anything.

2. 2

   Set a budget cap

   Give the campaign a total and a daily budget. Dial Milo forecasts what the rest of the list will cost, emails you at 80% of the budget, and stops calling when a budget is reached. Top-ups start at $10 with no subscription.

3. 3

   Keep a holdout

   Hold back 10% of every segment in your CRM before the list reaches the dialler. Without it, you can't compute the incremental number.

4. 4

   Wait for the full window

   Read 30-day results after 30 days have passed for the last player called, not the first.

If you'd rather hear the call before running numbers, the browser demo lets Milo call you first. When you're ready, sign up and price your first list. The wider strategy, from segments to offers, is in the player reactivation guide.