Who has already decided to leave? COD reads the early bend in a player's engagement and flags the onset, while the player is still reachable and the save is still possible.
15–30%
churn reduction via earlier intervention
Flagged at onset
not after the silence
Regulated-ready · Pseudonymised inputs · DPA available
Ranges are modelled estimates; actual results vary by operator data and maturity.
The industry counts days of inactivity: seven for regulars, fourteen for casual players, thirty before a player is filed as lost. By day one of silence, the decision to leave is already weeks old. The behavioural shifts that precede disengagement are visible long before the silence; standard reporting just isn't built to read them. COD reads the week before the silence, while intervention can still change the outcome.
Behavioural signals flow in. COD flags the players whose engagement is beginning to slip — and routes them to your retention team while the window is open.
The decline starts while the dashboard still says active. COD reads the bend, not the silence.
Retention signals answer one question at a time: this player, this week. Your month still gets planned on last month's actuals. The CRM calendar, the bonus budget, the VIP coverage, all committed before anyone knows what the base will look like when the money lands.
COD's player-level signals roll up into active-player forecasts at brand, market, or product level. The same model that flags who is drifting tells you how many will still be there in 30 days, so the month is planned on where the base is going, not where it was.
See the play: How many active players will you have in 30 days?
Not a churn score in a dashboard. A flag on the players slipping now — in time to do something about it.
Each player whose engagement is bending downward is flagged at onset, on the player record your team already works.
The signal fires when the decline begins — not weeks later when the player has already gone quiet.
Delivered into the CRM, retention automation, or alerting your team already uses — no new tool to check.
Live in weeks, not quarters — here's the path
Test whether COD's onset signals are accurate and useful against your own player base.
Explore the Pilot →Calibrate churn-onset detection from regulated-market priors before you have your own engagement history.
Explore Cold-Start →The exact scope is tailored to each operator's stage, data, and goals.
Book a 20-minute discovery call — we'll show you what churn-onset detection looks like against your data.
Related reading
HumanGraph may use cookies or similar technologies to support core website functionality and understand site usage. Essential cookies are always active and cannot be disabled.
Optional cookies (analytics and marketing) will only be activated with your explicit consent. You can change or withdraw your consent at any time.
Essential
Required for the website to function.
By clicking "Save Preferences" or "Accept All", you consent to the use of optional cookies as described in our Privacy and Cookies pages.