Insights
Saiga Systems publishes practical thinking on the operational layer behind CRM, revenue systems, and customer communications.
Each paper examines a problem we see repeatedly in mid-market and regulated environments: where control breaks down, what it costs, and what a stronger operating model looks like in practice.
Consumer Duty: evidence, not assertion
The FCA has said plainly that board reports asserting good customer outcomes, without the data to evidence them, will no longer stand. Two of the four outcomes are evidenced almost entirely from customer communications data, and in multi-brand groups that data is fragmented across the estates each acquisition brought with it. We have set out the problem, and a two to three week review that answers it, in full.
Why Your CRM Isn't the Problem
When CRM underperforms, the platform gets the blame. The operating model is almost always the real cause.
When CRM results disappoint, the instinct is to question the platform: the wrong system, time to migrate. It is almost always the wrong question, and platform migrations undertaken to solve operating model problems are among the most expensive mistakes a business can make. The operating model, the ownership, governance, and accountability around the platform, is the part that was never formally anyone's job to build. No migration fixes a problem the platform did not cause.
The Cost of Under-Activated CRM
Most platforms run at a fraction of their capability. The gap is quiet, and that is exactly what makes it expensive.
Most mid-market businesses run their CRM at a fraction of its capability. The platform is live, campaigns go out, the reporting looks acceptable, and beneath that surface sits a large, invisible gap between what the system does and what it was bought to do. The cost does not appear on a dashboard, because dashboards measure activity rather than capability. Surfacing it takes a deliberate exercise, and that exercise is almost always where the real conversation about CRM value begins.
AI on Top of Bad Data
AI in CRM is a sequencing problem before it is a technology opportunity. Applied in the wrong order, it scales what is already broken.
There is a version of AI-powered CRM that delivers real value, and it depends on foundations most businesses do not yet have. AI is an amplifier: applied to sound foundations it extends what works, and applied to weak ones it scales the flaws with equal efficiency. The useful question is not how to add AI to a CRM, but what a CRM would need to look like for AI to help. Answering it honestly tends to surface several months of unglamorous foundational work, and that work is far more valuable than the AI conversation it enables.
Who Is Governing Your AI
AI is already acting on your customers, usually with no oversight. Readiness gets it switched on. Governance keeps it safe.
Readiness determines whether AI should be switched on. Governance determines whether it stays safe once it is. AI rarely enters a CRM function through a single governed decision; it arrives incrementally, a drafting assistant here, a predictive score there, until a set of systems is acting on customers that no one formally owns, inventories, or oversees. A system acting continuously, at a scale no human reviews in full, needs deliberate oversight. Where that oversight is absent, the business is not running its AI. It is being run by it.
The CONTROL Scorecard
See where your CRM and revenue systems stand against the seven CONTROL dimensions. Two minutes, scored instantly, results on screen.
The GUARD Scorecard
AI is entering customer operations faster than the controls around it. See how governed your use of it actually is, in two minutes.