What compliance and financial crime leaders told us works when it comes to integration and implementation
The compliance teams getting value from technology are rarely those with the best tools. They are the ones that connect them to their data and processes, and implement them with discipline.
Ask compliance and financial crime leaders what separates the technology that works from the technology that doesn’t, and few start with the product itself. As a senior member of the financial crime team at an e-money firm put it, most solutions are “fairly impressive when they are off the shelf” and “can all do variations of the same things.” What matters is something else:
“The key obviously is then both the ability to integrate within your organisation [and] the quality of the data.”
This view came up repeatedly in our conversations with senior compliance and financial crime leaders, which are a qualitative snapshot rather than a survey of the whole market. The leaders getting the most from their technology described three habits: they design for how tools connect, they are honest about the limits of their data, and they treat implementation as work in its own right, with the supplier close by.
Connection comes first
Most compliance technology stacks are built up over time rather than designed. New tools arrive to meet a new requirement, a new market or a regulatory finding, and each has to fit around what already exists. That is why the leaders we spoke to put integration at the top of the list.
A compliance leader at a large money transfer business described the core challenge as “the integration of new technology into existing systems, and without adding more layers to a system, and also without breaking the existing system down.” Some firms ring-fence a new process to avoid that risk, but the result is that “you don’t get a holistic solution, you get a very piecemeal solution.”
Fast-growing firms face a version of the same problem. A fintech compliance leader pointed to the “eclectic systems and databases” that rapid growth leaves behind. Fixing that is rarely a clean project: “You have to change the parts of the engine while you’re still running it. You cannot really do it in a pit stop.”
Smaller firms are looking for simpler connections. A compliance lead at a small insurer wants one provider for KYC, sanctions and company checks, in place of today’s disjointed set of separate checks.
In our view, this changes the question firms should be asking. It is less about “which tool should we buy?” and more “how should customer information, risk decisions, alerts, investigations and evidence work together?”
Data matters, but perfection is not the goal
Integration and data go together. A financial crime leader at a UK tier-one bank was clear on the dependency: “Ultimately, your AI is only as good as your data,” and the AI “won’t know whether the data in some cases is good or bad.”
Moving between systems is often where that shows up. A compliance manager at a payments firm described moving their KYC refresh to a new vendor and finding that “the data didn’t cross over properly”, which meant cleaning up records so the team was reviewing the right accounts and not wasting time on inactive or dormant ones.
Not everyone accepts that data must be perfect first. The e-money firm leader pushed back: “I get the garbage in, garbage out, but I don’t fully buy that… I don’t think you need perfect data to get better solutions.” What matters, they argued, is being “clear on even the limitations of the data you’re feeding in.”
We think that is the more useful standard. Waiting for perfect data can quickly become a reason never to start. Knowing where your data is weak, and designing controls around it, is what makes a system defensible and trustworthy.
Implementation decides the outcome
The second recurring pattern is that the same product can succeed in one firm and fail in another. The tier-one bank leader measures success by “how easy and successful it’s been to do that project”, and has seen big vendors fall out of favour for reasons that “mainly doesn’t necessarily boil down to the quality of the product.” After projects overrun, some firms simply decide never to use the software again.
What separates the projects that go well, they said, is that success “boils down to understanding what the problem is in the first place and whether your client understands the problem in the first place.”
The gap between promise and delivery can be wide. A financial crime leader at a large payments processor described vendors promising “half the time to [on]board and with a quarter of the costs and then reality sinks in and actually you’ve made a 10% increase or 10% saving.”
Sometimes the fit is wrong from the start. A former compliance leader at a large bank recalled a well-known platform that worked well for dashboards, but “from an actual workflow perspective, it was terrible… after about nine months, we just reverted back to the previous system.”
This is why we see implementation as an organisational capability, rather than a purchase. A tool is only as good as the testing, configuration and ownership around it.
Getting value takes both sides
The firms that do this well test before they commit and keep the supplier close. A compliance manager at a payments firm pointed to testing and simulation they can run “before an alert can go live”, and speaks to their supplier almost every day, so issues are resolved quickly. A senior compliance leader at a payments fintech summed it up: “if you don’t have a good contact person, then it doesn’t really work.”
Buyers can also get more from what they already own. The e-money firm leader described firms that invest heavily up front, “only use part of the functionality,” then start looking for a replacement when “sometimes the problem is you haven’t fully understood what you could get.”
The relationship needs watching over time too. In a crowded RegTech market with frequent acquisitions, the same leader warned that a responsive supplier can change quickly once bought: “overnight your level of service and responsiveness is gone.”
What this means for leaders
Before the next compliance technology decision, it is worth asking:
- Can we state clearly what problem this solves? Does the supplier understand it too?
- How will it connect to what we already run? And who owns the integration once it is live?
- What are the limits of the data it depends on? Are they well documented and understood?
- How will we test it before go-live? Through simulation, parallel running or a phased rollout?
- What happens to support if the vendor is acquired? Is that part of the selection process?
- Are we using what we already have? Before replacing a tool, is the problem the product or how it was set up?
The firms getting most from compliance technology are not always those with the newest tools. They are the ones that treat integration and implementation as the work itself, not as an afterthought to buying.
Continue the conversation
How should compliance teams connect their tools, data and decisions?
It’s one of the themes shaping FinTech Compliance Leaders on 9 March 2027, Shape Finance’s invitation-only forum for senior compliance and financial crime leaders.
