Explore the build library
consultance.ai
Deal Copilot
Build libraryControlsAboutCase studiesBlogBook an audit
consultance.ai

Systems

WorkflowsDeal CopilotPrivate knowledgeFinance automationBuild library

Company

About usCase studiesBlogContact

Start a conversation

Free AI auditfounder@consultance.ai

Teams

Investment bankingCFOs & financePrivate equityPrivate creditReal estateProperty managementProfessional servicesOther industriesAll industries

Legal

Privacy policyTerms of service

Our approach

Audit the workflow.Build in your environment.Hand over a system you own.
Build something useful © 2026 consultance.ai
Build library/Insurance Agency Diligence Review

Finance and data

Insurance Agency Diligence Review

For agency acquirers and PE in insurance distribution: break down the book, split growth into lost, rate and new, tie commissions to carrier statements, model the earnout.

Medium setup4 steps10 prompts
Open the setup guide Discuss an implementation
Finance and data

Start with the guided setup

Follow the setup below in your own workspace. For a demonstration of the prompt-copying process, open the optional walkthrough.

Watch the setup walkthrough
Step 1 · setup
Three minutes, four steps, nothing to install by hand

Claude sets it up for you. You just paste.

Never used Claude? It is free and takes 30 seconds to open. Copy the instruction below, paste it into Claude, and it reads this page and walks you through everything, one question at a time.

  1. Tell Claude how to talk to you

    One tap. It changes how much Claude explains, and how slowly it goes. You can change it any time.

  2. Copy your setup instruction

    A short instruction plus a link to this page lands on your clipboard. First copy asks for your email once. That unlocks every button across the whole library.

  3. Open Claude in a new tab

    Free account, no card, 30 seconds. This tab stays open so you can come back.

    Open claude.ai
  4. Paste, send, and answer one question

    Claude reads this page, asks one question about your work, then guides you step by step until your first output is right. If anything looks wrong, tell Claude what you see, and it fixes it with you.

▸Prefer the full prompt instead of the link? (optional)
Click to copy
I am comfortable copy-pasting and following instructions, but I am not a developer.
There is nothing to install for this one and no commands to type: it all happens inside Claude. If any instruction below implies a Terminal, translate it into the equivalent click path for me instead.
- Plain English. Define jargon the first time it appears.
- One step at a time, then wait for me to confirm before the next one.
- Tell me what success looks like at each step, and diagnose any error before moving on.

Follow the instructions below with those rules applied.

If you can browse the web, open and read this page in full first, it has the complete guide and every prompt you will run (the vault is under the-vault anchor): https://consultance.ai/library/insurance-agency-diligence#the-vault . If you cannot open links, tell me and I will paste the page in, do not guess the prompts.

You are going to help me run the Insurance Agency Diligence Review on one agency, in my own Claude, with my own accounts. Go one step at a time and wait for me after each step.

First question, and only this one: where do the seller's files live? A few files I can upload here (the book of business export, carrier statements, the CIM), or a data room folder on my computer?

If I say upload: this is a Claude app job. Tell me to make a private Project, select Claude Opus 5.5 in the model picker, and upload the files to it. Do not give me a Terminal install.
If I say a folder: this is a Claude Code job. Tell me to open Claude Code in that folder, one step at a time, and tell me what success looks like after each step.

Then run the first session drill with me:
1. Paste prompt 01, the shared rules. Ask me my thesis and what worries me, then the agency, the book export, the carrier statements, the deal terms, my thresholds, my rate view and the named sign-off person, one at a time.
2. Run prompt 02 on the two fake agencies and compare every line to the answer key. If one line does not match, stop and tell me why before we touch real data.
3. Then run the first unit, book-breakdown, on my agency's book of business export.
Step 2 · run it on your data

Step 1 set it up. These 10 prompts do the work.

the vault

The 10 prompts

Grab the whole pack as one file, or tap any prompt below to copy it on its own. Placeholders that look like {{THIS}} get swapped for your own numbers — and if you ran Step 1, Claude fills them in for you.

One .md file · all 10 prompts, numbered, in order · nothing left out.
Click to copy
INSURANCE AGENCY DILIGENCE REVIEW: SHARED RULES
Every unit in this pack follows these rules. They are written here once.

1. WHERE TO RUN IT
- One agency, with a book export and a few statements: the Claude app, in a private Project. Chat is fine here.
- A data room folder of exports and carrier statements: Claude Code, opened in that folder.
- Seller files that must stay on a managed machine: Claude Code on that machine.
- The same tie out every month until close, or after close for the earnout: Claude Code, so each run lands as dated files you can compare.
Model: select Claude Opus 5.5 in the model picker. Use Claude Sonnet 5 only for the first bulk read of a large folder of carrier statements.
Privacy: your materials go only to your own Claude account, never to us. Use Team or Enterprise, or turn model training off, before you add client lists, premiums or anything under NDA.

2. BEFORE THE FIRST RUN
First ask the human: what is your thesis on this agency, and what worries you most? Record both answers word for word. Every later unit reports on them first.
Then ask for these, one at a time:
{{AGENCY}}: the agency's name and the fiscal years to review (default: the last 2 full years).
{{BOOK_EXPORT}}: the book of business export from the agency management system, one row per policy: client ID, carrier, line, producer, written premium, commission, new or renewal, effective date.
{{CARRIER_STATEMENTS}}: the carrier commission statements for the months to tie (default: the last 3 months of the latest year), and any agency bill invoices.
{{DEAL_TERMS}}: the CIM or teaser claims, the price and the earnout clause as written.
{{THRESHOLDS}}: concentration line (default 30% of the latest recurring base for one producer or one carrier, ASSUMED; house accounts are a group, not a producer), question line (default: an answer that could move the latest recurring base by more than 5%, ASSUMED).
{{RATE_VIEW}}: the renewal rate changes to model for the earnout year (default: 0% and the rate change measured last year, ASSUMED).
And the named person who signs off.
A missing item gets the plain default above, labelled ASSUMED.

3. STOP RATHER THAN GUESS
- A file path or folder listing pasted into chat: this is a Claude Code job. Say so and stop.
- A truncated export, or one missing a column the unit needs: name the column, and do not score that part.
- A figure needs a document you do not have: ask once, name the document, mark it OPEN.
A stop blocks only the outputs that need the missing input. Everything else still runs.
Never invent a carrier name, a commission rate or a market benchmark. Use only the sample and the human's own files.

4. LABELS ON EVERY FIGURE
REPORTED (document and page) · CALCULATED (show the arithmetic) · ASSUMED (who set it) · MARKETING (a claim in the CIM, teaser or seller letter: it makes a question, never a number) · NOT IN SOURCE.
Deal terms quoted as written (the price, the earnout clause) are REPORTED with their page, even inside the CIM, and the purchase agreement confirms them. A MARKETING figure feeds a calculation only "on the seller's own figures". The book export and the carrier statements outrank the CIM. When they disagree, the gap is a question for the seller.

5. THREE FLAG TAGS
- CHANGES THE DECISION: reaches the IC memo's view.
- WORTH A QUESTION: only if the answer could move recurring commission by more than the question line in {{THRESHOLDS}}. Smaller items go on the seller question list, untagged.
- EXPLAINED BY CONTEXT: one line under "Checked, normal". Once closed, it stays closed.
Test every pattern against the list below before you raise it. A pattern that matches the list goes straight to "Checked, normal" and is not a flag.
A BLOCK is not a flag. It is a tie out that failed. It stops the IC verdict, never the analysis. Only a carrier statement or a corrected book clears it, and the named person signs off the clearance.

6. LOOKS WRONG, IS NORMAL
- Revenue with one carrier drops sharply | the carrier left a line or a state, and the agency rewrote the clients with another carrier | not normal if the same client IDs do not reappear with another carrier.
- Contingent or profit sharing commission swings from year to year | it follows carrier loss ratios and growth targets | not normal if a high year sits in the recurring base or the earnout base.
- One or two months carry a large share of the year | renewals cluster on common dates such as 1 January and 1 July | not normal if a cluster month is annualised as a run rate.
- A producer's book falls to zero | the producer left and the accounts were reassigned inside the agency under a non solicit | not normal if the accounts left the agency.
- Policy count falls while client count holds | policies combined into a package | not normal if client count falls too.
- Revenue grows while client count falls | premiums rose on renewal | the revenue is real and closes as normal. A claim that calls it retention or new client growth is a separate test in retention-history.
- Commission on a client falls when it moves carrier | each carrier pays on its own schedule | not normal if many clients move to lower paying carriers at once.

7. STANCE
A neutral colleague who has done agency deals many times. Do not hunt for problems. Do not take a claim as proof.
This is not legal advice. Producer agreements, non solicits, licensing and carrier appointments go on a list of questions for counsel. Never judge whether a clause is enforceable.
Never contact the agency's producers, clients or carriers unless the seller has agreed.

8. HANDOFF (every unit ends with this)
Read (and not read) · Outputs · Checks: pass, fail or not run · Open items, with the document needed · Next unit · Sign-off needed from the named person.
Click to copy
Run the pack on this sample before you use real data. Compare your result to the answer key. If one figure does not match, stop and find out why first. One document is missing on purpose, so you see the pack stop instead of guess. Run units 04 to 10 on both cases in order.

SAMPLE (fake agencies, fake carriers; every figure cites its page)
The sample carries only the splits it lists. Any other split is NOT IN SOURCE, not an open item. The one missing document is in Case B.
Thresholds for both cases: concentration line 30%, question line 5%. Rate view: Case A 0% and 15%; Case B 0% and 3%. New client commission in the earnout year: the latest year's figure repeats, ASSUMED by the human.

CASE A: Tallow Creek Insurance (fake)
- Book export p.1: recurring commission FY2024 $2,000,000 from 1,000 clients; FY2025 $2,105,000. No contingent commission in either year.
- Book export p.2: 880 clients are on the book in both years. Their commission: FY2024 $1,700,000, FY2025 $1,955,000. Their written premium: FY2024 $14.0M, FY2025 $16.1M, same policies, same carriers.
- Book export p.2: 120 clients left; their FY2024 commission was $300,000. 60 new clients; their FY2025 commission is $150,000.
- Book export p.3, FY2025 by producer: P1 $800,000, P2 $505,000, P3 $500,000, house accounts $300,000.
- Book export p.3, FY2025 by line: commercial $1,263,000, personal $842,000.
- CIM p.4: "Revenue retention of 98%. Growth is driven by new clients."
- CIM p.6, earnout: "$800,000 is paid if FY2026 recurring commission, excluding contingent and profit sharing commission, is $2,100,000 or more. Nothing is paid below that."
- Producer agreement for P1, p.4: a non solicit for 2 years after departure.
- Commission tie, book vs carrier statement (statement p.1 for each carrier and month):
  Carrier A: Oct 2025 book $61,200, statement $61,200. Nov book $58,400, statement $58,400. Dec book $70,100, statement $66,300.
  Carrier B: Oct book $44,000, statement $44,000. Nov book $42,700, statement $42,700. Dec book $47,900, statement $47,900.
  No statement shows a reconciling item for Carrier A in December.

CASE B: Larkspur Insurance (fake)
- Book export p.1: recurring commission FY2024 $1,500,000 from 1,000 clients; FY2025 $1,560,000. Contingent commission (carrier profit sharing statements p.1): FY2023 $95,000, FY2024 $40,000, FY2025 $120,000.
- CIM p.3: "Total revenue FY2025 $1,680,000."
- CIM p.4: "Gross revenue retention of 94%."
- Book export p.2: 930 clients are on the book in both years. Their commission: FY2024 $1,410,000, FY2025 $1,452,000. 70 clients left; their FY2024 commission was $90,000. 50 new clients; their FY2025 commission is $108,000.
- Book export p.3, by carrier: FY2024 Carrier D $300,000, Carrier E $200,000, all others $1,000,000. FY2025 Carrier D $120,000, Carrier E $376,000, all others $1,064,000. No other single carrier is above 10% in either year.
- Carrier D notice p.1, dated 1 Mar 2025: Carrier D stops writing homeowners in the state from 1 Apr 2025.
- Book export p.4: 40 client IDs moved from Carrier D to Carrier E at renewal. Their commission: FY2024 at Carrier D $180,000, FY2025 at Carrier E $176,000. All 40 are among the 930 retained clients.
- Book export p.3, FY2025 by producer: Q1 $450,000, Q3 $420,000, Q4 $390,000, house accounts $300,000. Producer Q2 had $150,000 in FY2024 and left on 31 Dec 2024.
- Book export p.4: every one of Q2's FY2024 clients was reassigned to Q1 or house accounts and is among the 930 retained clients. Producer agreement for Q2, p.3: a non solicit for 2 years.
- Book export p.3, FY2025 by line: commercial $1,092,000, personal $468,000.
- Book export p.5: January 2025 commission $312,000. CIM p.5: "Most commercial accounts renew on 1 January."
- CIM p.7, earnout: "$500,000 is paid if FY2026 recurring commission, excluding contingent and profit sharing commission, is $1,600,000 or more. Nothing is paid below that."
- Commission tie, book vs carrier statement:
  Carrier D: Oct 2025 book $9,800, statement $9,800. Nov book $10,100, statement $10,100. Dec book $9,600, statement $9,600.
  Carrier E: Oct book $30,500, statement $30,500. Nov book $31,200, the Carrier E statement for November 2025 is NOT provided. Dec book $33,900, statement $33,900.

ANSWER KEY
| Check | Expected | Working |
| A. Recurring growth | +5.25% | CALCULATED: 2,105,000 / 2,000,000 - 1 (book p.1) |
| A. Client retention | 88.0% | CALCULATED: 880 / 1,000 (book p.2) |
| A. Gross revenue retention | 85.0% | CALCULATED: 1,700,000 / 2,000,000 (book p.2) |
| A. Seller's 98% | Ties as net revenue retention, 97.75% | CALCULATED: 1,955,000 / 2,000,000. It includes the rate rise on the clients who stayed |
| A. Rate on retained clients | +15.0% | CALCULATED: 1,955,000 / 1,700,000 - 1; premium 16.1 / 14.0 - 1 agrees (book p.2) |
| A. Bridge | 2,000,000 - 300,000 lost + 255,000 rate + 150,000 new = 2,105,000 | CALCULATED (book p.1, p.2) |
| A. "Driven by new clients" | CONTRADICTED, CHANGES THE DECISION | New clients add 150,000, lost clients take 300,000. The rate rise, 255,000, is the larger driver |
| A. Producer P1 | 38.0%, above the 30% line, WORTH A QUESTION | CALCULATED: 800,000 / 2,105,000. P2 24.0%, P3 23.8%, house 14.3%. Question for counsel: does P1's 2 year non solicit (agreement p.4) hold, and does P1 sign on at close |
| A. Lines | Commercial 60.0%, personal 40.0% | CALCULATED (book p.3) |
| A. Tie | Oct and Nov tie for both carriers. Dec Carrier B ties. Dec Carrier A: BLOCK, difference $3,800 | CALCULATED: 70,100 - 66,300. No statement shows a reconciling item |
| A. Earnout base year | 2,105,000 x 0.85 = 1,789,250 | CALCULATED, gross retention, not 98% |
| A. Earnout at 0% | 1,789,250 + 150,000 = 1,939,250. Not paid, short by 160,750 | CALCULATED |
| A. Earnout at 15% | 1,789,250 x 1.15 + 150,000 = 2,207,637.50. Paid | CALCULATED |
| A. Rate the earnout needs | +9.0% | CALCULATED: (2,100,000 - 150,000) / 1,789,250 - 1 = 8.98% |
| A. Earnout read | CHANGES THE DECISION | It pays only if renewal rates rise at least 9.0% again. The seller's 98% hides that |
| A. Verdict | BLOCK on the IC verdict until Dec Carrier A ties. The analysis is complete. View: reprice on the 85.0% gross retention; the earnout is a bet on rates | If nobody acts, the memo goes to IC marked BLOCK with no price view |
| B. Recurring growth | +4.0% | CALCULATED: 1,560,000 / 1,500,000 - 1 (book p.1) |
| B. Total revenue | 1,680,000 ties: 1,560,000 recurring + 120,000 contingent | CALCULATED (book p.1, CIM p.3) |
| B. Gross revenue retention | 94.0%, SUPPORTED | CALCULATED: 1,410,000 / 1,500,000 (book p.2, CIM p.4) |
| B. Client retention | 93.0% | CALCULATED: 930 / 1,000 |
| B. Rate on retained clients | +3.0% | CALCULATED: 1,452,000 / 1,410,000 - 1 = 2.98% |
| B. Bridge | 1,500,000 - 90,000 lost + 42,000 rate + 108,000 new = 1,560,000 | CALCULATED |
| B. Carrier D drop | EXPLAINED BY CONTEXT | Carrier D left homeowners (notice p.1). The 40 clients reappear at Carrier E (book p.4): Carrier E rose 176,000, the same as the rewrites |
| B. Contingent swing | EXPLAINED BY CONTEXT | 95,000, 40,000, 120,000 (statements p.1). Kept out of the recurring base, and the earnout clause excludes it (CIM p.7) |
| B. January cluster | EXPLAINED BY CONTEXT | 312,000 is 20.0% of the year, 2.4 times the monthly average of 130,000. 1 January renewals (CIM p.5). Never annualise January |
| B. Producer Q2 to zero | EXPLAINED BY CONTEXT | Q2 left 31 Dec 2024. Every Q2 client was reassigned and retained (book p.4). Counsel question only: Q2's non solicit (agreement p.3) |
| B. Concentration | None above 30%. Carrier E 24.1%, Carrier D 7.7%. Q1 28.8%, Q3 26.9%, Q4 25.0%, house 19.2% | CALCULATED on 1,560,000 |
| B. Lines | Commercial 70.0%, personal 30.0% | CALCULATED (book p.3) |
| B. Tie | All months tie except Nov Carrier E: OPEN, needs the Carrier E November 2025 statement | Every other check still runs |
| B. Earnout base year | 1,560,000 x 0.94 = 1,466,400 | CALCULATED, recurring only |
| B. Earnout at 0% | 1,466,400 + 108,000 = 1,574,400. Not paid, short by 25,600 | CALCULATED |
| B. Earnout at 3% | 1,466,400 x 1.03 + 108,000 = 1,618,392. Paid | CALCULATED |
| B. Rate the earnout needs | +1.7% | CALCULATED: (1,600,000 - 108,000) / 1,466,400 - 1 = 1.746%, so 1.7% at one decimal |
| B. Verdict | No flag changes the decision. One open item: the Nov Carrier E statement. The earnout sits near the line | |

The trap is Case A. The plausible answer is "98% retention and new client growth, so the earnout pays comfortably". It is wrong because 98% is net of a 15% rate rise. Gross revenue retention is 85.0%, and the earnout pays only if rates rise at least 9.0% again.
The second trap is Case B: putting the $120,000 contingent in the earnout base makes FY2026 look like it clears the line. The clause excludes it.
A run that flags any Case B normal item, or guesses the November Carrier E statement, has failed.
Click to copy
<unit name="tools-setup">
Follow the shared rules in prompt 01.
What it does: suggests tools for the jobs in this pack, then installs only the ones the human approves, on the human's own accounts.
Use when: after prompt 02 passes, before the first real run. Skip it if upload is enough, which it is for most single deals.
Needs: the "Needs" lines of every unit in this pack, and the systems the human already uses: their virtual data room, their accounting system, their spreadsheet tool, and (after close) the agency management system and carrier portals.
Method:
1. See what is already connected. In Claude Code, run `claude mcp list`. In the Claude app, look up the current steps in Anthropic's own help pages with your web search, then ask the human what their Claude already shows as connected.
2. For each need, search the web yourself for an MCP server or connector published by the vendor the human already uses, or listed by Anthropic. Prefer: official publisher, read only access, sign-in by OAuth, updated in the last 6 months.
3. Show one table: need · tool · publisher · official page URL · read only or read and write · how far it reaches (one folder, one account, or everything the human can open) · what it costs on the human's own plan (write "not stated" if the page does not say) · last update (write "no date shown" if none).
4. For each tool the human approves, open its official page and follow its own steps. A server with a command or URL: copy the command word for word, never from memory. In Claude Code the shape is `claude mcp add --transport http <name> <url>` for a remote server, or `claude mcp add <name> -- <command>` for a local one. Use `--scope project` when the tool is for this folder only. A connector added in the Claude app: give the human the steps from the official help page, as written there. No official page or steps: mark it ASK VENDOR and use UPLOAD for that need.
5. Install one at a time. Make one small read only test call and show the result. Then go to the next tool.
Rules: use only the human's accounts, keys and credits. Never ask for a key in chat. Use the tool's own sign-in, or an environment variable the human sets. Start every tool read only.
<trap>A server that is not from the vendor, asks for write access to the agency management system, or sends client lists to a third party for "processing". Leave it out, even when it is free. Before close, the seller's systems are the seller's: use the exports they give you.</trap>
Stop: no official page means do not install it. Mark that need UPLOAD and carry on. Producer agreements, carrier notices and the CIM are documents, so they are UPLOAD from the start. Every other unit still runs.
Output: the table, what was installed, each test result, and the handoff. Next: book-breakdown.
</unit>
Click to copy
<unit name="book-breakdown">
Follow the shared rules in prompt 01.
What it does: splits the agency's commission revenue by carrier, by line and by producer, and separates the recurring base from contingent commission and fees. This is the base every later unit uses.
Use when: first, as soon as the book of business export arrives.
Needs: {{BOOK_EXPORT}} for each year, the CIM's revenue figures, any contingent or profit sharing statements.
Method:
1. Check the export: one row per policy, the columns from prompt 01, a total per year. Sum it and compare with the CIM's revenue for the same year. Any gap is a line in the table, with its cause if a document shows one.
2. Split revenue into recurring commission (new and renewal), contingent or profit sharing commission, and fees. Recurring base = the sum of new and renewal commission. Contingent and fees are shown, never added to the base.
3. Group the recurring base by carrier, by line and by producer, for each year the export covers. A split the files do not carry is NOT IN SOURCE; it becomes OPEN only when a later unit needs it to reach a view. Share = group / recurring base, to one decimal. List every group, including house accounts and "all other carriers". Keep the carrier names exactly as the export writes them.
4. Do the same for the prior year, so retention-history and concentration can compare.
<trap>Building the base on the CIM's total revenue. The total often includes contingent commission and fees, which do not renew on their own. The base is recurring commission from the export.</trap>
Stop: no export, or one without the carrier, line or producer column: mark that split OPEN and name the column. A split with the columns present still runs.
Output: a revenue table (recurring, contingent, fees, total, CIM total, gap), three split tables (carrier, line, producer) for each year, then the handoff. Next: retention-history.
</unit>
Click to copy
<unit name="retention-history">
Follow the shared rules in prompt 01.
What it does: measures client retention and gross revenue retention, and bridges last year's recurring revenue to this year's: lost clients, rate on the clients who stayed, new clients.
Use when: after book-breakdown, and before any view on price or earnout.
Needs: the book export for both years, matched on client ID; the CIM's retention and growth claims.
Method:
1. Match clients across the two years by client ID. Count retained, lost and new.
2. Client retention = retained clients / last year's clients. Gross revenue retention = last year's commission from retained clients / last year's recurring base. Net revenue retention = this year's commission from retained clients / last year's recurring base.
3. Rate on retained clients = this year's commission from retained clients / their last year's commission - 1. Check it against their written premium change if the export has premium.
4. Bridge: last year's base - lost + rate effect + new = this year's base. It must tie to the dollar. Show every line.
5. Test each CIM claim against the bridge. Name which measure the seller's retention figure is, by re-deriving it. A figure that ties only as a different measure (net sold as retention) is CONTRADICTED. A CONTRADICTED claim that the price, the thesis or the earnout rests on is CHANGES THE DECISION.
6. Read the months: a month above twice the monthly average is checked against the renewal dates before anyone calls it a run rate.
<trap>Taking "revenue retention" as gross. A figure built on this year's revenue from last year's clients includes the rate rise. In a rising rate market it can sit near 100% while clients leave.</trap>
Stop: no client ID, or IDs that change between years: the bridge is OPEN. Name the field needed. The breakdown and the tie out still run.
Output: a retention table (client, gross revenue, net revenue), the bridge, a verdict on each claim (SUPPORTED, CONTRADICTED, OPEN) with its tag, then the handoff. Next: concentration.
</unit>
Click to copy
<unit name="concentration">
Follow the shared rules in prompt 01.
What it does: shows how much of the recurring base sits with one producer or one carrier, tests each move between years against the normal list, and writes the questions for counsel.
Use when: after retention-history.
Needs: the producer and carrier splits from book-breakdown for both years, the retained client list, producer agreements, carrier notices.
Method:
1. Mark every producer and every carrier above the concentration line in {{THRESHOLDS}}. For each, the revenue at risk = its share x the recurring base.
2. For each carrier whose revenue fell by more than the question line: look for a carrier notice and for the same client IDs at another carrier. Same clients, new carrier = rewrite, EXPLAINED BY CONTEXT.
3. For each producer whose book fell or went to zero: check whether the client IDs stayed in the agency under another producer or house accounts. Stayed = EXPLAINED BY CONTEXT.
4. For each producer above the line, and each departed producer: write the questions for counsel (the non solicit and its term, whether the producer signs on at close, licence status). Do not judge any clause.
<trap>Reading a carrier drop or a departed producer as lost revenue before you match the client IDs. Most move inside the agency, and the retention bridge already counts the ones that left.</trap>
Stop: no producer agreement for a producer above the line: mark it OPEN for counsel and name the agreement. The shares still run.
Output: a concentration table (name · share · revenue at risk · above the line or not · tag), the "Checked, normal" lines, the counsel question list, then the handoff. Next: commission-tie.
</unit>
Click to copy
<unit name="commission-tie">
Follow the shared rules in prompt 01.
What it does: ties the commission booked per carrier per month to the carrier's commission statement, and stops on any month that does not tie.
Use when: after concentration, and again each month until close.
Needs: {{CARRIER_STATEMENTS}} for the months to tie, and the book export's commission by carrier by month.
Method:
1. One row per carrier per month: book commission, statement commission, difference = book - statement.
2. A row ties when the difference is zero to the dollar.
3. A row with a difference ties only if a carrier statement shows the reconciling item (for example, a line booked in the next month's statement). Name the statement and page.
4. Any other difference is a BLOCK on that carrier month. Do not explain it, net it against another month, or call it rounding.
5. A carrier month with no statement is OPEN, not BLOCK. Name the statement needed.
<trap>Calling a small difference immaterial, or netting it against another month that happens to offset it. The tie is the integrity check on the whole book. A number that does not tie is not interpreted.</trap>
Stop: a BLOCK stops the IC verdict until a carrier statement or a corrected book clears it and the named person signs off. Every other unit still runs, and the memo prints the block. A missing statement is OPEN and blocks only that row.
Output: a tie out table (carrier · month · book · statement · difference · TIES, BLOCK or OPEN · source), then the handoff. Next: earnout-model.
</unit>
Click to copy
<unit name="earnout-model">
Follow the shared rules in prompt 01.
What it does: models the earnout year from gross revenue retention, the rate view and new business, and finds the renewal rate change the earnout needs.
Use when: after retention-history, whenever the deal has an earnout or a deferred payment tied to revenue.
Needs: the earnout clause as written, the latest recurring base, the gross revenue retention from retention-history, the new client figure, {{RATE_VIEW}}.
Method:
1. Read the clause: the measure (recurring only, or with contingent), the threshold, the payment, any partial payment. Quote it with its page.
2. Earnout year revenue = latest recurring base x gross revenue retention x (1 + rate change) + new business. Run it once for each rate in {{RATE_VIEW}}.
3. Paid or not at each rate, and the gap to the threshold.
4. Rate the earnout needs = (threshold - new business) / (latest base x gross revenue retention) - 1.
5. Tag CHANGES THE DECISION when the seller's case for the earnout rests on a retention figure that includes rate, or on a base that includes what the clause excludes. Otherwise the needed rate is a pricing fact for the memo, with no tag.
<trap>Applying the seller's net retention and then adding new business. Net retention already carries last year's rate rise, so the model assumes the rise repeats without saying so.</trap>
Stop: no earnout clause: mark the model OPEN and name the purchase agreement section. If the bridge is OPEN, run the model on the seller's own figures and label every line that way.
Output: an earnout table (rate · year revenue · threshold · paid or not · gap), the needed rate, the tag, then the handoff. Next: diligence-gate.
</unit>
Click to copy
<unit name="diligence-gate">
Follow the shared rules in prompt 01.
What it does: an independent check. It redoes the key figures from the export, the statements and the clause BEFORE it reads the earlier tables, then blocks the IC verdict if they do not tie.
Use when: after all earlier units, before the memo.
Needs: the raw book export, the carrier statements, the earnout clause and the CIM pages. Do not read the earlier outputs until step 3.
Method:
1. From the raw files only, recompute: the recurring base for each year, client and gross revenue retention, the bridge, every share above the line, every tie out row, the earnout at each rate and the needed rate.
In one chat session, do the recompute in a fresh message before you scroll back to the earlier tables. In Claude Code, run the gate as a separate session on the raw files folder.
2. Check every source line: document and page recorded, label right.
3. Now compare with the earlier tables. Any figure that differs: show both workings and the cause.
4. Check every flag against the normal list in prompt 01 again. Close any that match with one "Checked, normal" line.
<trap>Checking the earlier tables against themselves. The gate only counts if it starts from the raw files.</trap>
Stop: a re-derived figure that does not tie and cannot be traced is a BLOCK on the memo line that uses it. A tie out BLOCK from commission-tie stays a BLOCK here until a statement resolves it. The rest of the memo still runs.
Output: a verdict of PASS or BLOCK, a tie table (figure · earlier · re-derived · ties or not), closed flags, blocked lines, then the handoff. Next: ic-memo.
</unit>
Click to copy
<unit name="ic-memo">
Follow the shared rules in prompt 01.
What it does: writes the diligence memo for the investment committee, and a one file HTML view of the tie out and the earnout table from the human's own numbers.
Use when: last, after the gate.
Needs: the gate output, every earlier table, the human's thesis and worry from prompt 01.
Method:
1. Open with the view in two sentences, then what it means for the thesis and the worry.
2. Sections: recurring base and splits, retention bridge, concentration, tie out status, earnout read. Only CHANGES THE DECISION items reach the view. WORTH A QUESTION items become the seller question list. EXPLAINED BY CONTEXT items go in a short "Checked, normal" list.
3. A BLOCK prints at the top: which carrier month, the difference, the statement needed. With a BLOCK the memo gives no price view, and it says so.
4. End with the open items, the questions for counsel (marked "not legal advice"), and what happens by default if nobody acts.
5. Write one HTML file with the tie out table and the earnout table, colour by TIES, BLOCK, OPEN. No outside scripts or fonts.
<trap>A memo that lists everything found. The committee needs the two or three things that move the price, and what to ask about them.</trap>
Stop: a BLOCK from the gate: the memo still runs in full and shows the block and the default. An OPEN item is listed with the document needed.
Output: the memo, the seller question list, the counsel question list, the HTML file, then the handoff. Sign-off: the named person signs before it leaves the deal team.
</unit>

From a working build to an operating system.

For agency acquirers and PE in insurance distribution: break down the book, split growth into lost, rate and new, tie commissions to carrier statements, model the earnout.

Path A · free

Run the build independently

The setup rail and every prompt above are free and stay free. The cost is your time, and the risk of wiring it wrong on live data.

Back to the prompts
Path B · done with you

Connect it to your operation

We would deliver a private deal room: the review loaded into your Claude account, read only access to the data room exports and carrier statements, a monthly tie out until close and through the earnout, and a sign off step so no untied month reaches IC.

Book a build call
data safety

Before you use live numbers

  • • Run last quarter's numbers first. Live data is not a test bed.
  • • Nothing here uploads to us. It runs in your own Claude account, on your own machine.
  • • A named human reviews and signs every output before it reaches a board, lender, or client.
  • • Mask account numbers and names to the minimum the task needs.
the fine print

Straight answers on ownership

Review authored by consultance.ai. Not legal advice: producer agreements, non solicits, licensing and carrier appointments are listed as questions for your counsel. It uses only the seller's files and your own accounts. Your deal data stays in your own Claude account; we never see it. A named human owns the IC view.

Bring this workflow into your operation.

Book a build callBack to the library

Want this wired into your stack instead of running it yourself? That is our AI deal desk and finance automation service.

Newsletter

New builds, as they ship

One short note when a new build ships: what it does, who it is for, and how to run it in your own Claude.

Free. Three short welcome emails over your first three days, then at most one email a week. Unsubscribe in one click.

More like this

Related builds

All builds
Finance and data
Finance and data

AI Data Residency Check for CFOs

A go/no-go pack for CFOs putting AI on the books: a one-page data residency standard, a vendor questionnaire, and a board memo. Together they pin down where your data goes before any pilot.

Setup guide
Finance and data
Finance and data

Nightly Chart Review for Your Watchlist

For people who trade their own watchlist: 12 prompts and a free engine that review every chart after the close, then backtest and walk forward test your rule. No TradingView account needed.

Setup guide
Finance and data
Finance and data

Morning Portfolio Risk Review in Claude

For family offices and active investors running their own book: a before-the-open review of exposure, stops and pre-trade sizing in your own Claude, with a free local tool and 13 prompts.

Setup guide
Get the free kitBook a call

Share with your workflow owner.

Bring the relevant team into the conversation with the build and setup guide.

Forward by email
in one line

What is Insurance Agency Diligence Review?

Insurance Agency Diligence Review is a finance and data build in the consultance.ai AI Build Library. For agency acquirers and PE in insurance distribution: break down the book, split growth into lost, rate and new, tie commissions to carrier statements, model the earnout. It fits agency acquirers, MGA buyers, and PE and corp dev teams in insurance distribution who want to test an agency's book and renewals on the seller's own files before they set the price. Setup difficulty is Medium, with 4 plain-English steps.

What does Insurance Agency Diligence Review do?

For agency acquirers and PE in insurance distribution: break down the book, split growth into lost, rate and new, tie commissions to carrier statements, model the earnout.

Who is Insurance Agency Diligence Review for?

It fits agency acquirers, MGA buyers, and PE and corp dev teams in insurance distribution who want to test an agency's book and renewals on the seller's own files before they set the price.

How hard is Insurance Agency Diligence Review to set up?

Medium to set up — one guided setup instruction covering 4 plain-English steps, plus 10 ready-to-run prompts on the resource page.

How would consultance.ai build this out?

We would deliver a private deal room: the review loaded into your Claude account, read only access to the data room exports and carrier statements, a monthly tie out until close and through the earnout, and a sign off step so no untied month reaches IC.

What are the licensing terms?

Review authored by consultance.ai. Not legal advice: producer agreements, non solicits, licensing and carrier appointments are listed as questions for your counsel. It uses only the seller's files and your own accounts. Your deal data stays in your own Claude account; we never see it. A named human owns the IC view.

Connect this build to your workflow.

Discuss how Insurance Agency Diligence Review fits your systems, controls, and team. We define the integration, review process, and deployment requirements with you.

This build comes from our AI consulting and AI implementation practice — see the full AI in finance guide and how we work with CFO teams.

Book an audit