Skip to main content
Curvestone AI
Definition

What drives the ongoing cost of automated compliance checking?

Updated

Why the cost question comes after the first quarter

On calls with mortgage networks, lenders and brokers, the first pricing question is rarely about the start. It is about what happens next. The principal of a UK mortgage brokerage put it this way: "And then once we got past the first quarter, what are the ongoing costs going to be?"

It is the right question. The first few weeks are easy to plan. The ongoing cost lands in next year's budget, and it moves with later decisions: how many cases you send, which checks you switch on and how far you roll the tool out.

The four things that drive the ongoing cost

How many cases you check

Volume is the first driver. A firm that checks every submitted case sends more cases than a firm that checks a sample. So start by deciding how many cases you want checked, and at which stage.

How deep each check goes

Checks come in different depths, and not every case needs the deepest one.

  • Admin check. Confirms every document is there and every figure agrees, so the lender does not send the case back.
  • Pre-submission soft check. Asks fewer questions than the full review and gives the adviser a readiness view before the case goes in.
  • Full file review. Grades the whole file against your standard. Every grade shows the document and page it came from.

A firm might run the admin check on every case and keep the full review for a risk-based selection. The mix you choose is a cost decision as much as a compliance one.

Which checks and tools you need

A mortgage advice check is one scope. Adding other checks or tools widens it. Start with the checks your file reviews already cover, then add where your MI shows the gaps.

Your setup effort for the route you choose

Some cost sits with you, not the vendor. A CRM integration needs your systems and people to be involved. Drag and drop needs no integration project: someone exports the case and drops it in, and the check runs. Count the internal time for each route you consider.

How to work out your cost per file today

Before you compare any quote, put a number on what one manually checked file costs you now, using your own figures.

  1. Time how long a reviewer takes to check one file, from opening it to recording the outcome.
  2. Multiply that time by the reviewer's loaded hourly cost: salary, employer costs and overheads.
  3. Add the time spent on rework: chasing missing documents and re-checking cases that come back.
  4. Add the time a supervisor spends reviewing the reviewer's work, spread across the files checked.
  5. Divide the monthly total by the number of files checked that month.

That gives you a cost per file checked. Then multiply it by the number of files you would check if cost were not the limit. The gap between the two numbers is the honest comparison for any automated tool.

What a sample leaves unchecked

The FCA's Consumer Duty guidance says "a key part of the Duty is that firms assess, test, understand and are able to evidence the outcomes their customers are receiving". It also accepts that some monitoring, "such as distributional analysis or file reviews will be based on a risk-based sample".

A sample is a fair response to cost, but its size is usually set by reviewer hours rather than risk. When the cost per file falls, the same budget checks more files, and the sample can follow risk instead.

The guidance also says firms "will need to identify relevant sources of data" to assess customer outcomes. Results from a consistent check on every case give you that data: which advisers fail which checks, how often and on what kind of case. A small sample cannot show that pattern.

Questions to ask any vendor about ongoing cost

  1. What does the ongoing cost move with: cases, depth of check, the checks you switch on, or something else?
  2. Can you run a lighter check on most cases and the full review on a selection?
  3. What changes the cost after the first quarter, and how much notice do you get?
  4. What setup work falls on your team for each route, and what does the vendor do?
  5. Can you start without an integration project?

Curvestone's ongoing cost follows the same drivers: how many cases you check, how deep, and which checks and tools you use. Drag and drop needs no integration project. To work through it against your own cost per file, book a call or a demo.

Questions

Frequently asked questions

What drives the ongoing cost of automated compliance checking?
Four things: how many cases you check, how deep each check goes, which checks and tools you need, and the setup effort for the route you choose. Volume and depth usually matter most. Running a light admin check on every case and a full review on a selection is a different cost choice from fully reviewing everything.
How do we work out what manual file checking costs us today?
Time how long a reviewer takes per file and multiply it by their loaded hourly cost. Add rework, such as chasing missing documents, and the supervisor time spent on quality checks. Divide the monthly total by the number of files checked. That cost per file is the number to compare with any automated option.
Does every case need the full compliance review?
No. An admin check confirms that every document is there and every figure agrees before submission. A pre-submission soft check asks fewer questions and gives the adviser a readiness view. The full file review grades the whole file against your standard. Many firms run a lighter check widely and keep the full review for a selection.
Do we need an integration project before we start?
No. Drag and drop needs no integration project: someone exports the case, drops it in and the check runs. A CRM integration removes that manual step later, but it is not a precondition. Count your own team's setup time for whichever route you choose, because that effort is part of the ongoing cost.
Sources
  1. 01FG22/5 Final non-Handbook guidance for firms on the Consumer Duty (FCA)
Related reading
Dawid Kotur
Written by

Dawid Kotur

CEO and co-founder, Curvestone

Dawid co-founded Curvestone in 2024 after a decade working at the intersection of financial services and applied machine learning. He writes about the strategic direction of regulated-industry AI, the FCA's evolving approach to model risk, and the operational changes UK lenders are making in response to Consumer Duty. He sits on the FCA Smart Data Accelerator advisory cohort.

LinkedIn

Compliance that thinksahead. Automatically.

Join mortgage networks, lenders, and legal firms using Curvestone to review cases at scale.