vertical
iGaming: what we build for casino operators
The studio landing - retention scoring, media-buying agents, and the products behind them.
Read// retention stack
Off-the-shelf retention CRM vs Custom scoring into your CRM
These are not competing products - they solve different halves of the problem. A retention CRM is a delivery and campaign system. A scoring layer is a prioritisation system. The mistake is buying the first expecting it to do the second.
Off-the-shelf platforms segment by rules an analyst writes. That is cohort targeting with a nice interface. It is genuinely useful, and it is a different thing from a model that reads one player's transaction history and returns a decision about that player.
Side by side
Nine dimensions that decide it.
| Dimension | Off-the-shelf retention CRM | Custom scoring into your CRM |
|---|---|---|
| Targeting granularity | Cohort - rules over a handful of attributes. | Per player - a model output per individual. |
| Campaign delivery | Built in. Email, SMS, push, in-app, journeys. | Not included. Feeds the CRM or worklist you already use. |
| Integration with a self-written platform | Often the hard part. Schema assumptions rarely match, and integration can run for months. | An adapter is written against your schema. Mapping exercise, not a migration. |
| Explainability of targeting | The rule is visible, but it is a rule, not a prediction. | Score plus the features that drove it, per player. |
| Learns from outcomes | Only if an analyst manually rewrites the rules. | Retrained on outcome history on a defined cadence. |
| Time to first value | Fast if your stack is mainstream. Slow if it is not. | Data audit plus build - weeks, largely independent of how unusual your platform is. |
| Compliance gating | Usually present but operating on the platform's copy of consent state. | Reads self-exclusion and limit flags from your platform at send time. |
| Ongoing cost | Per contact or per profile. Grows with your player base. | Build cost, then compute. Largely flat as the base grows. |
| What happens if you leave | Segments, journeys, and history live in the vendor. | You own the pipeline and the model. |
How to decide
Pick by situation, not by preference.
A retention CRM is the right buy when
- You have no campaign delivery infrastructure at all - that gap is worth closing first.
- Your platform is a mainstream one the vendor already integrates with.
- Your team is small and needs the journey builder more than it needs better prioritisation.
- Volume is low enough that per-profile pricing is not yet a real cost.
Custom scoring is the right build when
- You already have delivery - a CRM, a dialer, a worklist - and the gap is knowing who to contact first.
- Your platform is self-written and vendor integrations have already stalled once.
- Your retention team ignores the current segments because they do not trust them.
- You have six-plus months of player history and the base is large enough that ranking matters.
- You need to explain to a regulator or an analyst why a specific player was contacted.
Our bias, stated
In most deployments the answer is both. Keep the CRM for delivery, add a scoring layer that decides who goes into which queue. Replacing a working delivery platform in order to add prioritisation is a large project with a small payoff, and we will say so on the first call rather than after signing.
Questions
Follow-up questions.
Can custom scoring feed the retention CRM we already pay for?
Yes, and that is the usual pattern. Scores are written back as attributes or list memberships in your existing platform, so your team keeps the interface it knows and the journeys it has already built. Nothing about the daily workflow has to change.
How long does the adapter take on a self-written platform?
The data audit and adapter phase is typically two to four weeks, covering schema mapping, verification of account-type and consent flags, and time-window correctness. If the audit finds the history cannot support the model, we stop there and say so.
Does this replace our retention analysts?
No. It changes what they spend the day on. Instead of writing and maintaining segment rules, they work a ranked list and design the offers. The judgement stays with them - the model only decides the order.
Still not sure which side you are on?
Thirty minutes, no pitch. We will tell you which of the two your situation actually points to - including when the answer is the one we do not get paid for.
Book a 30-min call