← Back to Work
Process file · Finance Finance

What’s inside: step-by-step flow · output checking

Small in-policy motor claims settle on their own track

The question this file answersWhat share of our claims are simple enough that a clerk adds nothing — and still wait behind the hard ones?

Fits: insurers where low-value claims wait behind complex files in one queue.

Typical day

What the desk looks like today

Typical, from the finance playbook. Depending on the line, an insurer's claims desk takes in anywhere from hundreds to thousands of claims each month. The same clerk opens every first notification — form, photos, policy number — whether it is a cracked wing mirror or an injury. The simple ones queue behind the ones that need investigation.

What changes

What Monday looks like after

By mid-morning the adjusters are working the files that need them — injury, disputed cover, a fraud flag — while the wing-mirror claims have already been checked and prepared for payment. You can see how many claims took the straight-through door this week, and every one that was held back, with the reason.

Typical, not a measured client result. Every figure here comes from the playbook source named below.

~35–55%

of routine claims handling — disputed and complex files stay human

Before: low-value, well-documented claims wait behind complex files because the same clerk opens every notification. After: industry studies (McKinsey, Accenture, 2023) put 35–55% of routine claims activity within reach — the straight-through door stays narrow and every hold-back is logged.

Where this number comes from

McKinsey “Insurance 2030” (2023) puts 50–60% of claims activities as automatable; Accenture (2023) reports 40–50% shorter cycle time for standard claims with AI triage; the finance playbook uses 35–55%. An industry range for routine claims, not our measurement and not a loss-ratio promise.

What we install

What we put in front of the systems you already run

Your policy and claims systems stay — SAP, Oracle, FIS or the core you run. This is the same triage build as our claims-triage file, with the straight-through door added:

  1. the first notification, the photos and the policy are read and classified as they arrive from the portal or the mailbox
  2. the policy record is pulled and coverage, excess and your thresholds are compared with it
  3. a claim that is in-policy, under your limit and shows no fraud flag is prepared for settlement through the claims system’s API, and everything else goes to an adjuster with a coverage summary attached.

You set the limit; we do not.

What stays human — and what this will not do

Injury. Total loss. Coverage disputes. Fraud flags. The angry customer on the phone.

What can go wrong — and what we do about it

A straight-through door is only as safe as the rules behind it — a threshold set too high pays claims that should have been questioned, so we start narrow and widen it with your claims director. Photo assessment is a suggestion, not an estimate; a damaged photo or an unusual vehicle stops the file. If the claims core has no API, settlement is prepared as a draft your team confirms, not posted. Accenture’s 40–50% is cycle time for standard claims; complex files are not in it.

What it costs to get there

The path: free 60-second estimate → free 20-minute review → paid audit of this one process (€1.5–3K, typically two weeks) → pilot with your people in the loop (€10–20K, weeks, not quarters). No transformation programme. Prices are public, on the services page →

Scoped in the audit — the playbook has no estimate for this exact desk.

This is about you if…
What does this mean in euros?

That depends on your volumes and wage costs — this page will not invent the number. The free 60-second estimate runs that calculation from your answers, with every multiplier sourced.

Get your free savings estimate 60 seconds · no sales call Or write first → Map a finance process like this one — free, 60 seconds →

Not a named Aperanda client. Process file · Finance.

Short process file. Same build as its parent file; the playbook has no separate volume or benchmark for this desk.

All process files