Research paper · Retention
Retention Cohorts After the First Deposit
Retention measures how many first-time depositors are active again in a later period. It is a diagnostic for content and product quality, never a target to reach by any means, because the same activity that counts as retention can also be a sign of harm.
Three definitions to fix first
Google Analytics' cohort exploration uses three settings that work well as a checklist for any tool. Cohort inclusion "Defines the initial condition a user has to meet to be included in a cohort." Return criteria "Defines the returning condition a user must meet to be included in a cohort." Cohort granularity "Defines the initial and returning cohort time frame."
For an organic acquisition funnel, a sensible starting set is:
- Inclusion: a first-time deposit, attributed to organic search under your written attribution rule.
- Return: a meaningful action you define, such as signing in and placing a bet or game round, not merely opening an email.
- Granularity: weekly for the first months, monthly after that.
Standard, rolling and cumulative views
The same cohort can produce three different retention curves. Google describes them this way:
| Type | Google's definition | Question it answers |
|---|---|---|
| Standard | "Each cell includes all cohort users who meet the return criteria for that individual period" | How many were active in week 4? |
| Rolling | "Each cell includes all cohort users who meet the return criteria for that period as well as all previous periods" | How many were active every week so far? |
| Cumulative | "Each cell includes all cohort users who meet the return criteria in any period in the exploration" | How many came back at least once? |
The site's model uses a single retained-account stage set at 61% of first-time depositors. It is an illustrative input with no fixed period, chosen only to finish the arithmetic. A real report must name its period and calculation type, or the number cannot be read.
Harm rules come before retention goals
Retention work in gambling sits under duties that do not apply in most other sectors. In Great Britain, the Gambling Commission's remote customer interaction rules require licensees to monitor activity from account opening and to use indicators including spend, patterns of spend, time spent gambling and customer-led contact. They state that "Licensees must take appropriate action in a timely manner when they have identified the risk of harm," and that licensees must prevent marketing and the take-up of new bonus offers where strong indicators of harm are identified. The main provision has been in force since September 12, 2022, and the marketing paragraph since February 12, 2023.
Self-exclusion schemes add another layer. Australia's BetStop register, launched in August 2023, lets people self-exclude from licensed online and phone wagering for a minimum of three months up to a lifetime. Once someone registers with BetStop, the government says licensed providers cannot let them place a bet, open a new betting account or send them marketing messages. In the UK, GAMSTOP is a free tool that lets people exclude themselves from "all licensed gambling websites and apps," with minimum periods of six months, one year or five years. In Australia, the National Consumer Protection Framework's measures include deposit limits and activity statements.
In practice this means:
- Exclude self-excluded and harm-flagged accounts from every retention campaign list before it is sent.
- Report safer gambling measures, such as deposit limit use and time-out use, beside the retention curve.
- Never build a model that predicts who is likely to deposit again in order to target people showing signs of harm.
Linking retention back to content
Organic content can only influence retention indirectly: clear rules pages and honest explanations attract people who know what they signed up for. To see that link, compare retention curves of cohorts grouped by landing intent, as set out in matching landing pages to search intent. A cohort that arrived through a rules explainer and stayed is better evidence of useful content than a cohort that arrived through an offer page.
Limits of this paper
This paper uses no operator data and gives no expected retention rates. Rules are summarized as published on the dates shown and are not legal advice.
Questions
What counts as active for a retention report?
Define it in writing. Signing in and placing a bet or game round is a common choice; opening a marketing email is not activity in the account.
Should cohorts start at registration or first deposit?
For an acquisition funnel, start at first deposit, so retention follows the stage before it. Registration cohorts answer a different question.
Can retention numbers be used to choose who gets bonus offers?
Only within the rules. In Great Britain, strong indicators of harm require marketing and new bonus offers to stop, and self-excluded people should be removed from every list.
Sources
Each source was read on October 27, 2026. How we choose sources is set out on the sources and reading list page.
- Google Analytics Help, [GA4] Cohort exploration, no date shown.
- Gambling Commission, LCCP social responsibility code provision 3.4.3: remote customer interaction, no date shown.
- Victorian Gambling and Casino Control Commission, BetStop: national self-exclusion register has launched, updated December 9, 2025.
- Victorian Gambling and Casino Control Commission, National Consumer Protection Framework, updated June 12, 2026.
- GAMSTOP, What is Gamstop Online?, no date shown.
- Australian Government, BetStop: the National Self-Exclusion Register, no date shown.