Diligence Workspace ยท Tidewater
An engagement at scoping before the agents have run
An illustrative carve-out at Scoping. The data room is empty, the agents have not run, and the workspace says so rather than guessing.

A buyer is carving a professional services business out of a listed parent. Every policy the business relies on is part of the parent’s global program and names the parent as the insured. The engagement has just been opened; the data room is empty. This post is about the start of an engagement, when the honest answer to most questions is “not yet known”. Every name and figure here is illustrative.

Use case card
EngagementProject Beacon — Tidewater Analytics Group Inc. (illustrative), DIL-2026-0006
DealIllustrative $88M EV; revenue $58M; 260 employees. Buyer Arlo Street Equity, seller a listed parent, Coastline Holdings plc (both illustrative)
StageScoping: no documents, no runs
ActorsDeal broker (owner), practice leader
Agents involvedNone yet; all nine once the data room has documents
Decisions and authorityThe seller request, the stage moves and every review are the broker’s
Systems touchedInsightPulse (parent, hierarchy) as soon as Analyze runs
OutcomeA scoped engagement that knows what to ask for, and a workspace that says what it does not yet know

What a global program does to a carve-out

flowchart TB PARENT["Listed parent
global program"] --> GL["GL"] PARENT --> PL["Professional liability
claims-made"] PARENT --> DO["D&O $25M
claims-made, run-off clause"] GL --> NI["Named insured:
the parent and its subsidiaries"] PL --> NI DO --> NI NI --> CLOSE["At closing the business
stops being a subsidiary"] CLOSE --> NONE["No policy responds
for the business"] NONE --> NEED["Standalone program,
tails for claims-made lines,
run-off for the D&O"] style NONE fill:#fdecec,stroke:#e57373 style NEED fill:#e7f8f1,stroke:#10b981

Two consequences follow. The business needs a standalone program from the day of closing, priced at its own scale rather than its parent’s; and every claims-made line needs either a tail on the parent’s policy or a retro date preserved on the new one. The report agent’s illustrative uplift for lines leaving a parent program is the largest in the benchmark table for that reason.

At scoping, the workspace says what it does not know

The workspace does not fill an empty data room with plausible text. The hero says the agents have not run; every zone says what it is waiting for. “Move to Data room” is the next step, and the stage machine refuses Analysis until there are documents.

Step by step

Project Beacon, the first two weeks Illustrative; days from the engagement opening.
1
Engagement openedDeal brokerDay 1

Target, client, seller, due date and the Pulse lookup key are set. Stage: Scoping.

2
First requestDeal brokerDay 1

Ask the seller for the global program’s schedule of insurance with the business’s allocation, the policies naming the business, five years of loss runs for the business alone, and any intercompany indemnities.

3
Documents arriveData roomDay 4

The schedule and the declarations are read into page-level pieces. The engagement moves to Data room, then to Analysis once documents exist.

4
AnalyzeAll nine agentsDay 4

The program agent marks lines on the parent program; the tail agent reads the D&O’s run-off clause and the professional liability retro date; the subsidiary agent compares Pulse’s hierarchy with the named insureds; the data room agent prepares the request for what is still missing.

5
Report agentCostDay 4

Estimates the standalone program from the allocated premiums with the parent-program uplift, and the tails.

6
ControlsSupervisor agentDay 10

If loss runs cover only the parent’s book, or fewer than five years for the business, the loss-run control warns; a report due within a week with gaps open refers to a practice leader.

The engagement’s first states

stateDiagram-v2 [*] --> Scoping Scoping --> DataRoom: data room created DataRoom --> DataRoom: documents arrive DataRoom --> Analysis: at least one document Analysis --> Review: Analyze has produced findings note right of Scoping: The agents have not run.
Nothing is guessed.

Who asks whom

sequenceDiagram participant B as Deal broker participant DD as InsightDueDiligence participant S as Seller (parent) participant P as InsightPulse B->>DD: Open engagement (scoping) B->>S: Request: allocation, policies, loss runs S-->>DD: Documents into the data room B->>DD: Analyze DD->>P: Parent and hierarchy P-->>DD: Parent, subsidiaries (as of) DD-->>B: Findings: lines on the parent program, tails, run-off

Why show an empty engagement at all?

Because the first thing a deal practice needs to trust is that the tool will not make things up. An engagement with no documents shows zero findings and says why. Every figure on the other engagements in this blog exists because a document or a Pulse fact put it there.

On the deal desk

The first request to the seller is the most valuable hour of a carve-out. The playbook’s data room questions make it the same request every time.

For IT & architecture

The stage machine’s blockers are enforced by the API, not just the screen: no Analysis without documents, no Review without findings.

For practice leaders

The standalone program is usually the largest number a carve-out buyer has not modelled. It appears as soon as the policies are read.

Questions we get asked

Can the agents use the parent’s public filings?

They use what InsightPulse knows about the parent and its hierarchy, cited with an as-of date, and what is in the data room. Nothing else.

Keep reading