A private equity buyer signs an exclusivity letter on a mid-market business. Somewhere in the next three weeks, someone has to tell the buyer what insurance the target really carries, what it has cost, what it has paid out, what stops working at closing and what the buyer will have to buy on day one. That someone is usually a broker’s deal practice, and the tool is usually a spreadsheet.
This post describes that work as it is done today, before any product is involved. It is the baseline InsightDueDiligence is measured against, so it is worth being precise about where the time goes and where the mistakes come from.
What insurance due diligence is for
Insurance due diligence answers four questions for a buyer:
- What does the target carry? Every line, carrier, limit, retention, premium, term and trigger, and how it is structured: standalone, on the seller’s master program, through a captive or self-insured, with collateral posted or premium financed.
- Is it enough? Missing lines and thin limits against what a business of that sector and size normally buys.
- What has happened? Five years of loss runs: frequency and severity by line and year, open reserves, large losses, the trend, and the litigation and contingent exposures the loss runs do not show.
- What changes at closing? Claims-made cover that needs a tail, the sellers’ D&O run-off, change-of-control clauses that cancel or convert a policy, subsidiaries no policy names, and what the target’s program will cost once it stands alone.
The answers feed three places: the purchase price and the purchase agreement (indemnities, special indemnities, who funds collateral), the buyer’s program from day one, and the representations and warranties insurance placement, where every known issue is a likely exclusion.
How the work actually goes
40–200 documents"] --> B["Analyst downloads
policies, schedule,
loss runs"] B --> C["Retypes limits, retentions,
premiums into a
spreadsheet"] C --> D["Reads loss runs
line by line"] D --> E["Skims legal letters
and Phase I for
exposures"] E --> F["Drafts a report
from the spreadsheet"] F --> G["Partner review
the night before"] G --> H["Report to the buyer;
R&W team works
from its own notes"] style C fill:#fff4e2,stroke:#f4a03a style G fill:#fdecec,stroke:#e57373 style H fill:#fdecec,stroke:#e57373
Most of the hours go into the third and fourth boxes. A schedule of insurance lists the program, but the facts that matter at closing — the retroactive date, the change-of-control clause, the named insured — are in the declarations and the conditions of each policy, not the schedule. Loss runs arrive as carrier exports in different layouts, often for fewer years than asked. The legal letter and the environmental report describe exposures that the insurance program may or may not respond to, and nobody connects them to the program unless they think to.
Where it goes wrong
| What is missed | Why it is missed | What it costs the buyer |
|---|---|---|
| The tail | The schedule says “D&O, $10M”. The trigger and the retroactive date are in the declarations. | Claims for pre-closing acts, made after closing, are not covered. |
| The parent program | Four lines sit on the seller’s master program and simply end at closing. | A standalone program bought in a hurry, at a price nobody modelled. |
| The unnamed subsidiary | The named-insured schedule lists three entities; the corporate hierarchy has four. | An operating entity with no cover, sometimes in another country. |
| The clause that acts at closing | “Terminates automatically on a change of control” sits in the conditions. | A cyber policy that ends the day the deal closes. |
| Collateral | Letters of credit posted to insurers are rarely on the schedule at all. | A purchase agreement silent on who funds or releases them. |
| The exposure outside the program | A product recall or a contamination finding sits in a legal letter, not the loss runs. | An uninsured cost that should have been priced or indemnified. |
None of these are exotic. They are the ordinary consequence of reading forty documents by hand under a deadline, and of a report that is a summary of a spreadsheet rather than a set of findings each tied to the page it came from.
The second problem: the hand-off
Insurance diligence and the R&W placement are usually run by different people, sometimes in different firms. The R&W underwriter will exclude what the buyer already knows; the insurance report is one of the places the buyer learns it. In practice the R&W team reads the report as a PDF and re-types what it needs. Findings drift between the two, and nobody can say which version a given exclusion came from.
became which exclusion
What a better version needs
Reading this list back, the requirements for doing it well are not about speed for its own sake:
- Every policy read into fields with the page each came from, so a reviewer can check a limit in one click instead of re-opening the PDF.
- Figures computed by an engine from what was read — totals, trends, gaps against a written benchmark, the cost after closing — so they are the same every time.
- Findings, not prose: each one categorised, graded, cited and reviewed by a person before it reaches the report.
- Something that reads the clauses and the corporate hierarchy, because that is where the closing surprises are.
- A signed hand-off of the reviewed findings to the R&W placement, so the exclusion and the finding are the same record.
- A record of who decided what, because a diligence report is advice a buyer relies on.
That is the shape of InsightDueDiligence: agents that do the reading and the drafting, engines that do the arithmetic, and people who review every finding and sign every decision. The rest of this blog shows how, on an illustrative book of six deal targets and one vendor review.
Questions we get asked
Is this an insurance product or a broker tool?
A broker tool. InsightDueDiligence is used by a broker’s deal practice to review a target’s existing insurance program. It does not issue, price or bind insurance.
Does it replace the diligence team?
No. It takes the reading, the re-typing and the first draft off the team. Every finding is reviewed by a person before it reaches a report, and every action the agents prepare waits for someone with the authority to take it.
Are the companies on this site real?
No. Every target, carrier, person and figure on this site is invented for illustration and labelled so.
