To overview
Why your onboarding process takes too long – and why a new tool alone won't help

Why your onboarding process takes too long – and why a new tool alone won't help

May 27, 2026 · 6 min read

Many financial institutions that come to us have the same story: "Our customer onboarding takes too long." The compliance team is overloaded, sales is complaining because customers are leaving, and the whole setup is expensive and inefficient. The supposed solution is usually found quickly: a new KYC/KYB tool is needed! 

In theory, this isn't a bad idea, as many tools offer measurable benefits, from improved user experience and automated register queries to structured data collection. In reality, however, we repeatedly observe the same pattern: The tool is implemented, expectations are high. Six months later, the problems remain the same, only the bottom line has worsened.  

The reason is simple, but inconvenient: A tool automates what already exists. If the process is unclear, ambiguity is automated. If the data model is incorrectly set up, incorrect data is collected more quickly. The real problem lies deeper and has multiple dimensions.

What do I actually need from the tool? The requirements problem 

Buying a tool before knowing your own requirements is like buying the answer to a question you never asked. Which customers am I onboarding – with what level of complexity, what legal structures, in which markets? What risk classes do I have? Where should the process be in three years? These questions sound obvious, but in practice, they are surprisingly rarely asked before a purchase decision is made. 

What we experience more often instead is that someone from purchasing has heard about a tool that is good for "something to do with identities", and the implementation begins.  

What the tool should deliver in the specific regulatory and operational context remains unclear. This stems less from negligence and more from a fundamentally unanswered question: Who in the organization is actually qualified to define these requirements, and who is authorized to make the purchasing decision? Operations, IT, Compliance, Product, and Sales each bring a different perspective – without the structured involvement of all departments affected by the introduction of a new tool, the end result is a purchase that misses the mark. 

Who does what – and who is responsible for what? The division of labor between the obligated party and the provider

A tool relieves the responsible party of operational work, but not of responsibility. Therefore, for every relevant process step, it must be absolutely clear: Who provides the technical specifications? Who executes them? Who approves them? Who escalates them? Who monitors them? 

If this performance threshold isn't defined before implementation, chaos will ensue. The tool collects data, but no one has determined what happens to it – which decision logics apply, which cases are automatically approved, and which end up for manual review by the KYC analyst. 

When responsibilities are unclear, a process that was supposed to be automated ultimately generates vast amounts of manual work. A tool can execute tasks, but it cannot make decisions – that responsibility always remains with the person responsible.  

What data do I actually need to collect? The data model problem

When we ask four departments in a project which data should be collected, we get four different answers. Compliance wants to play it safe and "collect everything!" This makes the UX team break out in a cold sweat. But Sales would also like a complete customer overview, and Customer Support would be happy to have a second email address to wish the customer a happy birthday. 

"Collecting everything" sounds like the safest option, but it isn't. Data points that are neither required by regulations nor support risk scoring lengthen the onboarding process without adding any value and cost conversions.  

The line between regulatory requirement and overcompliance is thin. Which data must be collected for which customer type, risk class, and market: these are not IT questions, but rather regulatory and technical decisions made by the obliged entity. If these decisions are left to the tool or not even made in the first place, gaps, duplicate data collection, and poor data quality result. 

The correct approach is therefore to consistently question every data point before introducing a new tool: Is this really a mandatory field? Are we collecting this because we have to – or because we've always done it this way?  

What happens when the happy path doesn't work? Decision logic and customer outreach 

The tool provider's demo puts a smile on everyone's face: Customer data flows in seamlessly, the process works flawlessly, and the customer is approved in no time. Reality, however, is often less rosy: Data is incomplete, reviews reveal discrepancies – someone has to contact the customer. The crucial question, which is far too rarely asked, is: How exactly does this customer outreach work? Who contacts the customer using which tool, and how does the information the customer provides get back into the same system?

Often, a scattershot approach is used here – one piece of information is missing, but the customer is expected to provide all the details again. If only a copy of an ID is missing, instead of sending a link to the entire portal, a link to a single upload field is provided; everything is kept as simple and streamlined as possible. This ensures the information goes directly into the case file and not into the email inbox of a KYC analyst who's currently on vacation – information stored in various locations will inevitably cause problems during the next audit. 

Those who haven't defined in advance what happens when information is missing – how often do we remind the customer, what risks do we take if we continue the onboarding process with incomplete data – end up improvising in individual cases. This slows down the process and makes it expensive and unpredictable. 

What flows where – and in what quality? The integration problem 

A pattern we frequently observe in projects: The new tool has been implemented, but not all employees are using it. The new colleague is unfamiliar with the system, accesses the old contact in the CRM, and writes an email. The customer's response doesn't end up in the case file, and the source of truth becomes unclear—a problem arises at the latest when auditors want to see a complete case history. This is particularly critical with re-KYC and customer lifecycle management: If customer data is maintained in multiple locations, eventually no one knows which data point takes precedence. 

A new tool won't solve a data silo problem if it creates a new one itself. How does the tool connect to existing CRM systems? What data is transmitted in real time, and what data is retrieved? What information flows into which downstream systems, and in what format? These questions must be answered before implementation—not during, and not after. 

What needs to come before the tool – and how consalty supports this 

A tool is a crucial component for making compliance processes more efficient and customer-friendly. But before its implementation, the foundations must be in place, and this step is often skipped in practice. Thinking through requirements, clarifying the service interface with the provider, defining the data model from a business perspective, establishing decision and escalation logic, and setting up integration smoothly: these are not nice-to-haves, but prerequisites without which even the best tool cannot do its job. 

Institutions that commission us have often already gone through one or two tool changes. The pattern is always the same: without a sound process and governance foundation, the old problems are carried over to the new system. Creating this foundation and then selecting the right tool is what we at consalty help obligated parties with – from process design and data modeling to TOM development, tool selection, and its implementation. 

Do you have questions about process and tool selection for your onboarding? Feel free to contact us for a free initial consultation via [email address/phone number] This email address is protected against spambots! JavaScript must be enabled to view it. or via LinkedIn.

FAQ

Why does KYC/KYB onboarding take too long despite the new tool?

Because a tool automates what already exists: An unclear process remains unclear, and a poorly designed data model only collects incorrect data faster. Typical causes include undefined requirements, an unclear division of labor between the institution and the provider, an overloaded data model, a lack of rules for querying the customer, and weak integration with existing systems.

What should be clarified before implementing a KYC/KYB tool?

Five key points: defining your own requirements based on customer types, legal forms, markets, and risk classes; defining the service level with the provider; establishing a professionally defined data model; implementing decision-making and escalation logic, including customer contact in case of missing data; and integrating with CRM and downstream systems. A tool relieves operational workload, while decisions and regulatory responsibility remain with the obligated entity.

Should as much customer data as possible be collected during onboarding?

No. Data points that are neither required by regulations nor support risk scoring lengthen the onboarding process and cost conversions. The responsible entity decides, based on technical and regulatory criteria, which data is needed for which customer type, risk class, and market; this is not an IT issue. Therefore, every data point must be rigorously reviewed before any tool is implemented.

How should missing customer documents be requested during onboarding?

Targeted approach: If only a copy of an ID is missing, the customer receives a link to a specific upload field instead of the entire portal. This ensures the information goes directly into the case file and not into an analyst's email inbox, where it would be missing at the next review. It should be determined beforehand how often the customer will be reminded and what level of risk is acceptable in the event of incomplete data.

Note: This article does not constitute legal or regulatory advice. All information is current as of May 2026.

also interesting

Do you have a specific problem in mind?