The quarterly report nobody enjoys building
Your compliance lead needs to show the board how customers are faring. Which measures matter, and what they mean, is your firm's decision. Getting the numbers is the painful part: renewal retention by product, how often clients with vulnerability markers were contacted, claims declined, complaints by theme, cancellations in the first weeks, fees charged across customer groups.
So someone runs a dozen exports from the broking system, pulls the complaints log from a spreadsheet, asks the claims team for their figures, and spends days in Excel lining them up. By the time the report reaches the board, the data is two months old, and next quarter the whole thing is done again.
The data exists, but not in one place
Broking systems are built for transacting business, not for outcome reporting. Complaints are often in a separate log. Vulnerability flags, where they exist, may be free-text notes. Claims outcomes come from insurers. None of it joins up without manual work, and every manual step is a chance for the numbers to change between quarters for reasons that have nothing to do with customers.
- Exports differ each time depending on who runs them.
- Customer groups, such as product, channel or vulnerability, are defined inconsistently.
- Complaints and claims data are not linked to the policy record.
- Nobody can trace a figure in the board report back to its source.
What manual monitoring costs
The main cost is that the monitoring is too slow and too irregular to act on. If a problem shows up in a quarterly spreadsheet, it has been happening for months. The second cost is skilled compliance time spent on data wrangling. The third is confidence: figures that cannot be traced back are hard to defend when questioned.
A pipeline your compliance team defines
- We agree with your compliance lead the measures you want to monitor and how each is defined, written down in plain English.
- Data is pulled regularly from your broking system, complaints log, claims records and any call or note system, through APIs or scheduled exports.
- Records are joined on client and policy, and customer groups are applied consistently from one definition.
- Where vulnerability information exists only as free text, a model can suggest flags for a person to confirm, never changing records by itself.
- Each measure is calculated the same way each period and shown on a dashboard, with breakdowns by product, channel and customer group.
- Every figure can be traced to the records behind it, so a reviewer can click through to the cases.
- A board pack is produced from the same data, with space for the compliance lead's commentary.
We do not decide what good outcomes look like or whether the data shows them. That is your firm's judgement. We make the data consistent, regular and traceable.
Example measures and their sources
| Measure (your definition) | Source | Breakdown |
|---|---|---|
| Renewal retention | Broking system | Product, channel |
| Complaints by theme | Complaints log | Product, customer group |
| Claims declined | Claims records | Product, insurer |
| Early cancellations | Broking system | Product, channel |
| Contact with vulnerable customers | Notes and flags | Handler, product |
The table shows the kind of measures firms often choose. Yours may differ entirely.
Is your outcome reporting like this?
- The board report is built in Excel from several exports.
- Figures change depending on who ran the export.
- Complaints and claims data are not linked to policies.
- Your compliance lead spends days on data rather than on review.