The commercial problem
Four operators, four different starting points
The initial difficulty was not a lack of software. Each business already had a CRM environment. The problem was that a platform, on its own, does not create a retention programme. People can build campaigns, create segments and report opens without having a clear operating model connecting those activities to commercial decisions.
I worked with four separately owned online casino brands with different levels of CRM maturity. One had capable technology but relatively little lifecycle depth. Another carried a large collection of legacy campaigns that was difficult to navigate. Two had reporting that struggled to connect retention activity with revenue. Those are different operational problems. Imposing the same immediate fix on all four would have been a mistake.
My approach was to standardise the questions the teams were answering, not pretend the businesses were identical. What does a new customer need to understand? Which journeys are genuinely active? Where do messages overlap? What counts as a conversion? When a campaign receives attributed revenue, what can that number actually tell us? Until these are agreed, a portfolio report is four sets of incompatible numbers placed next to each other.
This distinction also matters for the word portfolio. The source describes four separate clients and owners, not a single corporate group sharing one player database. A common management framework should never be confused with pooling customer data across independent operators. Each brand retains responsibility for its own permissions, processing, regulatory controls and customer relationships.
Why I kept three different CRM platforms
Two operators used Optimove, one Customer.io and one Fast Track. That allocation is documented in the original case. Rather than making a platform migration the project, the shared programme was designed to function across the systems already in place.
CRM software should be evaluated against the job an operator needs to do. A larger business with mature customer data may value a sophisticated segmentation and orchestration environment. A smaller team may benefit more from a manageable automation setup with well-defined events. A real-time betting or casino operation will have its own integration and service requirements. None of these considerations establishes that one vendor will outperform another across all operators.
A migration can consume months of development and analytics effort without solving unclear ownership, weak campaign logic or poor measurement. My preference is to identify whether the real constraint is technology, data quality, people or operating discipline before prescribing a replacement. If the current platform can execute the necessary journeys reliably and compliantly, building a consistent programme may offer more immediate value than buying a new tool.
There is a limit to standardisation. Different platforms count contacts, opens, conversions and attributed revenue differently unless their event definitions and reporting rules are aligned. I use shared naming, tagging and attribution conventions so the teams can compare performance without pretending their technical environments are identical.
A shared lifecycle structure, not 70 identical messages
The engagement account records a library of 70 lifecycle campaigns per brand, totalling 280 across the four operators. The practical value of a library is that it gives teams a common map of what exists, what is missing and what needs review. It should not mean that four independent operators send the same offers to the same audiences.
The recorded framework covered welcome and onboarding, early customer value, ongoing play, declining activity, lapse, reactivation and VIP handling. I view these as areas of responsibility, not as permission to contact every customer in every state. A particular segment or trigger only makes sense when it reflects verified behaviour, permitted contact and an outcome that matters commercially.
The lifecycle areas covered by the programme
A common lifecycle framework creates consistency while leaving room for brand-specific execution.
Welcome and onboarding
Make account access, product information and the permitted next step clear.
Early customer value
Understand initial experiences and avoid treating every new player alike.
Ongoing engagement
Match useful communication to actual preferences and contact history.
Changing activity
Investigate the reason for a change, including service or product problems.
Lapse and reactivation
Apply consent and safer-gambling eligibility before considering contact.
VIP and customer care
Coordinate appropriate service and protection, not automatic incentive escalation.
The common architecture enables meaningful governance. A CRM manager should be able to identify the owner of a journey, the event that makes someone eligible, the suppression conditions, the message's commercial purpose and the metric that determines whether it deserves to stay live. For a multi-operator review, the same categories should be visible across teams even if their underlying configurations differ.
A campaign library also exposes duplication. Two journeys may compete for the same customer at the same moment. A transactional explanation may be buried under a promotion. A newly launched campaign may inherit an outdated rule from an earlier version. It is better to discover that during routine programme maintenance than after an operator has drawn the wrong conclusion from its monthly results.
Consumer psychology matters, but it is not a substitute for CRM discipline
I use neuromarketing-informed behavioural principles to examine how customers process information, make choices and respond to communications. In a CRM programme, this means improving relevance and comprehension before reaching for more messages. The focus is on relevant communication, understandable choices and a consistent customer experience.
Reduce cognitive load before adding another incentive
Nielsen Norman Group distinguishes necessary mental effort from avoidable cognitive load. A message that promises one thing and opens a page with three different actions makes the customer do extra work. So does an account notification whose wording conflicts with an earlier email. A clear subject, a relevant reason for the contact, accurate conditions and one understandable next step can reduce confusion without concealing any important information.
I would treat the benefit of that clarity as a hypothesis to validate with downstream behaviour, not a guaranteed increase in deposits. A higher open rate does not automatically mean a customer understood the offer or that the relationship became more valuable.
Use relevance rather than decorative personalisation
A customer's name in a subject line is easy to implement. It tells me very little about whether the message should be sent. Lifecycle state, product preference, recent service issues, expressed choices and prior communications are more useful questions to investigate, provided the data is accurate and processing is lawful.
Research into email personalisation shows why message framing can influence attention. In practice, I focus on whether a message is relevant to the customer and whether it leads to a meaningful next action, rather than optimising open rates in isolation.
Apply Cialdini's principles with evidence and restraint
Robert Cialdini's work on authority, consistency and social proof is useful when explaining trust. Genuine licensing information, understandable terms and a consistent message across channels can help customers assess an operator. Authentic, verifiable customer information may provide social proof when appropriate. None justifies inventing testimonials, presenting fabricated popularity or implying that others' gambling behaviour is a reason to place a bet.
I am interested in behavioural science because it helps identify a better question. Is the message understandable? Is this genuinely relevant? Is the customer making an informed choice? Those principles help me shape communication that customers can understand and trust.
The operational debt that a single-platform report can miss
One of the most useful findings came from the operational audit. I identified 26 finished campaigns left unpublished in draft across two brands and five campaigns at another brand that referenced an SMS fallback that was not connected. These are exactly the execution problems a shared CRM operating model should uncover.
A draft campaign that is never activated cannot do the job it was designed for. A fallback channel that is missing creates a gap between the journey diagram and what the customer actually receives. Both problems can persist while dashboard reporting still looks busy. They also show why campaign governance should include execution status and data dependencies, not only engagement charts.
The account also refers to a group of poorly performing onboarding emails beneath a much stronger individual subject-line result. That difference should prompt an investigation of audience eligibility, intent, deliverability, message relevance and sample size. It does not automatically mean the highest-open subject line should be copied across every programme.
What I would put into a cross-brand quality review
These are practical audit questions informed by the historical findings, not a verbatim record of the client's unpublished processes.
- Which campaigns are live, paused, drafted or no longer relevant?
- Do triggers, fallbacks, permission checks and suppression rules actually work?
- Are journeys competing for the same customer's attention?
- Can analytics reproduce the event and attribution logic used in the report?
- Which messages have a clear service purpose, and which merely add contact volume?
A shared reporting framework makes these defects easier to discuss across operators, but it does not remove the need for local accountability. Each business must still own its data contracts, campaign approvals and compliance decisions. A central review can identify a problem. It should not silently change another operator's customer experience.
Performance across four casino operators
The monthly portfolio overview brought the four brands into one commercial reporting framework. The figures show the different levels of engagement, conversion, churn and CRM-attributed revenue across the operators, creating a clearer basis for management decisions.
| Operator | Platform | Open rate | Click rate | Conversion | Monthly churn | Attributed revenue |
|---|---|---|---|---|---|---|
| Operator A | Optimove | 38.4% | 8.3% | 6.1% | 4.2% | $284k |
| Operator B | Optimove | 33.6% | 7.8% | 4.8% | 5.6% | $201k |
| Operator C | Customer.io | 29.1% | 6.8% | 3.4% | 7.8% | $158k |
| Operator D | Fast Track | 24.7% | 7.5% | 2.2% | 9.9% | $92k |
The four operators generated a combined $735k in monthly CRM-attributed revenue. The portfolio's average open rate was 31.5%, alongside substantial differences in conversion and churn performance between brands.
I evaluate the financial figures alongside customer behaviour, because opens alone do not reflect commercial performance. Email privacy features can also affect how opens are recorded. This is why my portfolio reviews combine engagement data with conversion, churn, CRM-attributed revenue and customer value.
The results made the differences between brands visible. Operator A achieved the highest open rate, lowest churn and highest CRM-attributed revenue, while operator D had more work to do across those measures. I used this comparison to identify where a deeper review of lifecycle coverage, communication quality and execution would be most valuable.
I would use the spread to select the next diagnostic question. Why does one operator have a stronger repeat-activity profile? Is its customer base different? Are permissions and deliverability comparable? How much value remains after bonuses? Only after answering those questions would I consider scaling a campaign pattern across the other brands.
How I connect CRM-attributed revenue to commercial decisions
Attribution asks which customer activity a reporting system associates with a campaign. Incrementality asks what additional activity happened because that campaign existed. Those are different questions, and the distinction is central to the commercial credibility of CRM.
A customer might receive an email and deposit the next day even if the email had never arrived. A mobile message and an embedded message might both be associated with the same deposit. An incentive might increase the number of deposits while reducing net contribution once the bonus cost is deducted. None of those possibilities can be resolved from platform-attributed revenue alone.
The portfolio recorded $735k in monthly CRM-attributed revenue. I use that commercial view alongside acquisition mix, customer retention, incentive costs and campaign performance to identify where the CRM programme can contribute more value.
My preferred measurement approach is to define the eligible customer group before a journey begins, use a properly designed holdout where lawful and operationally suitable, and compare a relevant outcome across a fixed window. For a retention programme, that may include customer activity, net gaming revenue, bonus costs, contribution, retention and player-protection outcomes. A smaller campaign with a genuine incremental effect can be more valuable than a larger one that simply collects credit for existing demand.
What this changed in the way I manage CRM
The programme needs an owner, not just a software administrator
Automation tools execute decisions. Someone still has to resolve competing priorities, define success, check that campaigns reached the intended audience and decide when to stop doing something. Fractional CRM leadership can be useful when an operator already has platform access and an execution team but needs that commercial layer of management.
Comparable structure is valuable only when the definitions match
A common campaign naming convention helps teams find their work. A shared lifecycle map helps them see missing journeys. Neither makes revenue attribution comparable by itself. I want common definitions alongside common labels, plus a record of where platform differences prevent an exact comparison.
Behavioural relevance beats sending for the sake of activity
The objective of retention work should not be to maximise messages or encourage more gambling from every customer. Appropriate contact requires consent, a legitimate reason and respect for responsible-gambling controls. In regulated markets, suppression, self-exclusion and indicators of harm must take priority over commercial scoring. As one jurisdictional example, the UK Gambling Commission requires applicable remote operators to provide marketing preferences by product and channel, with opt-out defaults. The anonymised source does not identify these operators as UK licensees.
The dashboard should improve the next decision
A board can make decisions from a clean view of campaign health, cohort contribution and uncertainty. It should not need to admire a wall of open rates. The point of the cross-brand model was to make the operation easier to manage, with consistent reporting and clearer quality checks. Whether a new campaign produced incremental value remains a further measurement question, not an assumption I would ask an operator to accept.
When a shared CRM operating model is worth building
This approach is useful for operators with a working CRM platform but a fragmented lifecycle programme, small teams working without consistent priorities, or independent brands that need the same management discipline without sharing their customer data. It is less useful when the product is unreliable, registration or payments are fundamentally broken, or the operator cannot produce trustworthy events. In those cases I would fix the underlying journey or data contract before introducing another sequence.
There is a commercial decision underneath all of this: where can the next hour of CRM work make a measurable difference? Sometimes the answer is a new lifecycle journey. Just as often, it is repairing a failed trigger, removing a redundant send, resolving a service issue or setting up a valid control group. That is the difference between running campaigns and running a programme.
For operators and multi-brand teams
Make your CRM reporting useful to the people running the business
If your platform is capable but your lifecycle programme is difficult to manage, I can assess your campaign structure, event definitions, operating responsibilities and attributed-versus-incremental performance. We can start with a short review of the bottlenecks and agree what needs attention first.
Discuss a CRM programme review