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/LP Due Diligence Questionnaire Desk

Finance and data

LP Due Diligence Questionnaire Desk

For GP investor relations and fund CFOs: match a new LP questionnaire to your past answers, draft the rest, flag every stale or changed answer, and stop any figure that does not tie.

Medium setup4 steps9 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/lp-ddq-desk#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 LP Due Diligence Questionnaire Desk on one new questionnaire, 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 questionnaire, my past DDQ answers and the fund documents live? Files I can upload here, or a 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 new questionnaire, my last answered DDQ and the latest fund report. 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 what worries me about this questionnaire and what the LP is focused on, then the fund, the LP, the as-of date, the due date, the answer library, the fund documents and the two sign-off people, one at a time.
2. Run prompt 02 on the fake sample 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, the answer match, on my real questionnaire and my last answered one.
Step 2 · run it on your data

Step 1 set it up. These 9 prompts do the work.

the vault

The 9 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 9 prompts, numbered, in order · nothing left out.
Click to copy
LP DDQ DESK: SHARED RULES
Every unit in this pack follows these rules. They are written here once.

1. WHERE TO RUN IT
- One questionnaire and a few past answers: the Claude app, in a private Project. Upload the files. Chat is fine here.
- A folder of past DDQs, fund reports and policies: Claude Code, opened in that folder.
- Fund documents that must stay on a managed machine: Claude Code on that machine.
- Every questionnaire in a fundraise: Claude Code, so each LP's draft, change log and open items land 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 past answers.
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 fund documents or an LP's questionnaire.

2. BEFORE THE FIRST RUN
First ask the human: what worries you about this questionnaire, and what is this LP most focused on? Record both answers word for word. Units draft and check the questions that bear on them first.
Then ask for these, one at a time: {{FUND}} (the fund being raised and its predecessor), {{LP_NAME}}, {{QUESTIONNAIRE}} (the new DDQ file), {{AS_OF_DATE}} (the date the LP wants answers as of; default: the last quarter end), {{DUE_DATE}}, {{ANSWER_LIBRARY}} (the last fund's DDQ response and recent LP responses, each with its as-of date), {{FUND_DOCUMENTS}} (latest quarterly report, audited financial statements, LPA, Form ADV if you file one, policies, team roster), and the named sign-off people: the CFO for every figure, the compliance officer for every policy, legal and regulatory answer. Team, committee and process answers are drafted with the COO and signed by the compliance officer.
A missing item gets a plain default, 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 or unreadable file: name it, and do not use it.
- An answer needs a document or a confirmation you do not have: ask once, name the document and its owner, mark the answer OPEN. Never fill it from the last fund's answer.
A stop blocks only the answers that need the missing input. Everything else still runs.

4. LABELS ON EVERY FIGURE AND DATED FACT
REPORTED (document, page, and its as-of date) · CALCULATED (show the arithmetic) · ASSUMED (who set it) · MARKETING (a claim from a pitchbook or deck, such as "top quartile" or "best in class": it makes a question for the owner, never an answer) · NOT IN SOURCE.
A past answer is not a source. It points to where the fact came from. The fact is current only if a document dated after the past answer's as-of date says so.

5. THREE FLAG TAGS
Here "the decision" is what the packet tells the LP.
- CHANGES THE DECISION: the answer must change before it goes out, or a figure cannot go out (a stale fact, a change since the last fund, a figure that does not tie).
- WORTH A QUESTION: only if the answer could move a figure in the draft by more than 0.5 points of IRR, 0.05x of a multiple or 1% of AUM, or change a name, date or policy the answer states. Smaller points are not flags: they go on the owner's list.
- EXPLAINED BY CONTEXT: one line under "Checked, normal". Once closed, it stays closed.
Each question also gets a status (REUSE, REFRESH, PARTIAL, NEW, OPEN) from the answer-match unit. An OPEN row gets no flag tag: a missing document or confirmation is a question for its owner, listed on the open items, never a verdict.
Test every flag against the list below before you raise it.

6. LOOKS WRONG, IS NORMAL
- A policy dated two or more years ago | policies are reviewed on a cycle and kept when nothing changed | normal when a review log shows an on-cycle review; not normal if the policy says annual review and the log has no entry in the 13 months before the questionnaire's as-of date.
- Net IRR with the subscription line is higher than without it | the line delays capital calls, which lifts IRR; its interest lowers TVPI a little | report both, as the LP asks; not normal if the gap reverses sign with no reason.
- A figure dated earlier than the questionnaire's as-of date | quarterly reports come out weeks after quarter end, and Form ADV assets are stated once a year | state the figure's own date; not normal if a newer issued document has the figure.
- Gross multiple well above net TVPI | fees, carry and fund expenses sit between them | not normal if net is above gross.
- DPI near zero in the first years of a fund | nothing is sold yet | not normal past the investment period with no realisations planned.
- Two AUM figures in one packet | regulatory AUM includes uncalled commitments, NAV does not | normal if each is named and dated; not normal if two figures with the same name and date differ.
- The same question worded differently by each LP | every LP sends its own version | answer the fact the same way each time.
- Unrealised marks moved between the last quarterly report and the audited year end | audit adjustments | normal if the audited figure is used and the change is explained.

7. STANCE
A neutral colleague who has run many fundraises. The job is an accurate, consistent answer the CFO and compliance officer can sign. Never improve a number, never soften a change, never put deck language into an answer. Do not hunt for problems either: a dated policy with a review log is fine.
Your compliance officer checks how performance is presented against the rules that apply to your firm. This pack does not decide that.

8. HANDOFF (every unit ends with this)
Read (and not read) · Outputs · Checks: pass, fail or not run · Open items, with the document and owner needed · Next unit · Sign-off needed from the CFO or the compliance officer.
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.

SAMPLE (fake firm; every figure cites its document and page)
Firm: Northgate Capital, raising Fund IV. Predecessor: Fund III.
New questionnaire: Harlow County Pension DDQ, received 2 Nov 2026, due 20 Nov 2026. p.1: "Answer as of 30 Sep 2026."

CASE A: the answer match
Harlow questions:
- H1 (p.2): "State the firm's regulatory assets under management."
- H2 (p.2): "List the members of the investment committee."
- H3 (p.3): "Describe the key person provision and any changes since the predecessor fund."
- H4 (p.3): "Describe your responsible investment policy and when it was last reviewed."
- H5 (p.4): "Describe any litigation or legal proceedings against the firm in the last five years."
Past answer library: Fund III DDQ response, as of 31 Mar 2025 (PAST):
- PAST p.3: "Regulatory AUM of $1.40B as of 31 Dec 2024."
- PAST p.9: "The investment committee is Ana Ruiz, Ben Osei and Carla Diaz."
- PAST p.6: "Key persons are Ana Ruiz, Ben Osei and Carla Diaz."
- PAST p.22: "Our responsible investment policy, dated 15 Jan 2024, is reviewed annually."
- PAST p.25: "There has been no litigation against the firm."
Current documents:
- Form ADV Part 1A, filed 27 Mar 2026, p.2: regulatory assets under management $1.85B as of 31 Dec 2025.
- Team roster, 30 Sep 2026, p.1: "Ben Osei left the firm on 30 Jun 2026. Investment committee from 1 Jul 2026: Ana Ruiz, Carla Diaz, Dev Patel."
- Fund III LPA, p.41: key persons Ana Ruiz, Ben Osei and Carla Diaz; a key person event occurs if any two cease to devote substantially all of their business time to the fund. No amendment to the key persons has been made (side letter and amendment index, p.1).
- Fund IV draft LPA, p.38: key persons Ana Ruiz, Carla Diaz and Dev Patel; a key person event occurs if any two cease to devote substantially all of their business time.
- Responsible investment policy, dated 15 Jan 2024, p.1. Review log, p.4: "Reviewed 20 Jan 2026 by the compliance officer. No changes."
- The compliance officer's litigation and regulatory confirmation as of 30 Sep 2026 is NOT provided.

CASE B: the track record exhibit
- H6 (p.5): "Provide Fund III net IRR with and without the subscription line, TVPI, DPI and gross multiple, with value creation attribution for the largest realised deal."
- Fund III quarterly report, as of 30 Jun 2026 (QR). The Q3 2026 report is not yet issued.
  QR p.4: paid-in capital $400M, cumulative distributions $220M, NAV $380M.
  QR p.5: net IRR 14.2% with the subscription line, 12.9% without it.
  QR p.7, deal schedule (cost, then realised plus unrealised gross value): Alder $100M, $300M (realised). Birch $200M, $240M. Cedar $50M, $100M.
  QR p.8, Alder: entry EBITDA $20M at 8.0x, entry net debt $60M. Exit EBITDA $30M at 11.0x, exit net debt $30M.
- IR team's draft exhibit (DRAFT p.1): net IRR 14.2%, TVPI 1.50x, DPI 0.55x, gross multiple 2.07x.

ANSWER KEY
| Check | Expected | Working |
| H1 AUM | REFRESH, CHANGES THE DECISION. Answer: $1.85B as of 31 Dec 2025 | PAST p.3 is dated 31 Dec 2024. ADV p.2 is newer. CALCULATED: 1.85 / 1.40 - 1 = 32.1% change. The ADV date is earlier than 30 Sep 2026: normal (annual figure), state the date |
| H2 Investment committee | REFRESH, CHANGES THE DECISION. Ruiz, Diaz, Patel | Roster p.1: Osei left 30 Jun 2026 |
| H3 Key person | REFRESH, changed since last fund, CHANGES THE DECISION | Osei replaced by Patel (LPA p.38). Fund III: 1 of 3 named key persons has ceased; the trigger is 2 (Fund III LPA p.41, no amendment), so no key person event; one more departure would trigger one. Disclose the departure and the change |
| H4 Policy | REUSE (or PARTIAL), EXPLAINED BY CONTEXT. Add "last reviewed 20 Jan 2026" | Policy dated 15 Jan 2024, but review log p.4 shows an on-cycle review, no changes |
| H5 Litigation | OPEN, owner: compliance officer | Needs the confirmation as of 30 Sep 2026. "No litigation" from PAST p.25 is not reused. Every other answer still runs |
| TVPI | 1.50x, ties | CALCULATED: (220 + 380) / 400 = 1.50 |
| DPI | 0.55x, ties | CALCULATED: 220 / 400 = 0.55. RVPI 380 / 400 = 0.95 |
| Net IRR | 14.2% with line, 12.9% without, EXPLAINED BY CONTEXT | QR p.5. The line delays calls, so IRR with it is higher |
| As of date | 30 Jun 2026, EXPLAINED BY CONTEXT | Q3 report not yet issued. State "as of 30 Jun 2026" on the exhibit |
| Gross multiple | 1.83x. Draft 2.07x does not tie: BLOCK on that line, CHANGES THE DECISION. The draft also omits the 12.9% without the line and the with or without labels: add them from QR p.5 | CALCULATED: value 300 + 240 + 100 = 640; cost 100 + 200 + 50 = 350; 640 / 350 = 1.83. The draft averaged deal multiples: (3.0 + 1.2 + 2.0) / 3 = 2.07 |
| Alder attribution | EBITDA growth $80M, multiple $90M, debt paydown $30M, total $200M | Entry equity 20 x 8.0 - 60 = 100. Exit equity 30 x 11.0 - 30 = 300. EBITDA growth (30 - 20) x 8.0 = 80. Multiple (11.0 - 8.0) x 30 = 90. Debt 60 - 30 = 30. 80 + 90 + 30 = 200 = 300 - 100 |
| Packet | Not ready to send. Open: H5 confirmation (compliance officer); gross multiple corrected to 1.83x and signed (CFO). Everything else drafted | By default nothing goes to Harlow until both close; due 20 Nov 2026 |

The trap is Case A. The plausible answer is "H1 to H5 all match the Fund III response word for word, so reuse all five". It is wrong because the AUM, the committee and the key persons have changed, and "no litigation" was true only as of 31 Mar 2025.
A run that flags H4, the IRR gap or the 30 Jun date, or that reuses or guesses the litigation answer, 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.
Needs: the "Needs" lines of every unit in this pack, and the systems the human already uses: the shared drive or document store that holds past DDQs and policies, the virtual data room for the fund, any DDQ or RFP answer library software the firm already licenses, the fund administrator's portal, and the CRM where LP requests are logged.
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 data room, 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. Start every tool read only. A data room connector must reach only this fund's room, never every room the human can open.
<trap>A tool that can write to the data room or send answers to an LP portal. Leave write access off. Nothing leaves the firm until the CFO and compliance officer sign.</trap>
Stop: no official page means do not install it. Mark that need UPLOAD and carry on. Every other unit still runs.
Output: the table, what was installed, each test result, and the handoff. Next: answer-match.
</unit>
Click to copy
<unit name="answer-match">
Follow the shared rules in prompt 01.
What it does: points at one new questionnaire and your last answered ones, matches every question to the closest past answer, and says which can be reused, which need a refresh, which are new and which are open.
Use when: first, the day the questionnaire arrives.
Needs: {{QUESTIONNAIRE}}, {{ANSWER_LIBRARY}} with each response's as-of date, {{FUND_DOCUMENTS}}, {{AS_OF_DATE}}.
Method:
1. One row per question: its ID and page, the topic, and any document it asks for. Note the LP's as-of date and due date.
2. Find the closest past answer by meaning, not wording. Record the past document, page and its as-of date. If two past answers to different LPs give different facts for the same question, name both.
3. List the decaying facts in that past answer: AUM or fund size, any person's name, headcount, committees, key persons, a policy name or date, a performance figure, a service provider, litigation or regulatory matters, any as-of date.
4. For each decaying fact, find a current document dated after the past answer's as-of date. Status: REUSE (every fact is confirmed by a current document or a review log, even if a fact the question asks for, such as a review date, is added straight from that document), REFRESH (a fact has changed or is out of date: give the current value and its source), PARTIAL (the past answer covers only part), NEW (no past answer), OPEN (needs a confirmation no document gives: name it and the owner).
5. Put the questions tied to the human's worry and the LP's focus at the top.
<trap>Reusing an answer because the question text matches. A word for word match with an old AUM, a departed partner or "no litigation" from last year is the most common way a wrong answer reaches an LP.</trap>
Stop: no answer library: every row is NEW and the next unit drafts them all. No current document for a decaying fact: that row is OPEN with the document and owner named. Every other row still runs.
Output: the match table (question · page · topic · past answer, document, page, as of · status · decaying facts · current source · owner · tag), counts by status, then the handoff. Next: ilpa-draft.
</unit>
Click to copy
<unit name="ilpa-draft">
Follow the shared rules in prompt 01.
What it does: drafts every REFRESH, PARTIAL and NEW answer from the fund documents, in the LP's own format, mapped to the ILPA DDQ 2.0 topics when the LP uses that template.
Use when: after answer-match.
Needs: the match table, {{FUND_DOCUMENTS}}, and the human's copy of the ILPA DDQ 2.0 if the LP uses it (ILPA's page, ilpa.org/due-diligence-questionnaire, offers it as PDF and DOCX with a user guide).
Method:
1. Map each question to the ILPA topic it falls under, using the section headings in the human's copy, for example Succession Planning / Key Persons, Track Record, Accounting / Valuation, Legal. Where the LP uses its own form, keep its numbering and note the ILPA topic beside it.
2. Draft from current documents only. Every sentence that states a fact cites the document and page. Every dated fact carries its own as-of date.
3. Answer the question asked, in plain words, at the length the form allows. Where the question asks for a document, name the file that answers it.
4. Name the owner for each answer: the CFO for figures, the compliance officer for policies, legal and regulatory answers, the COO for team and committee answers, the deal team for strategy.
5. Where ILPA asks about change (for example the key person provision "including context into any changes since the predecessor fund"), the answer states the change.
<trap>Filling a gap with a sentence from the last fund's answer, or with deck language such as "top quartile". A sentence with no current source is either OPEN or MARKETING, never an answer.</trap>
Stop: a question whose source document is missing gets the line "[OPEN: needs <document> from <owner>]" and no draft. The rest still drafts.
Output: the draft answers in the LP's order (question · draft · sources · as of · owner · status), then the handoff. Next: change-log.
</unit>
Click to copy
<unit name="change-log">
Follow the shared rules in prompt 01.
What it does: compares the new draft with the last fund's response, line by line, and lists every change so the CFO and compliance officer see them in one place before they sign.
Use when: after ilpa-draft, before the exhibits.
Needs: the draft answers, the last fund's response, the predecessor and new LPA.
Method:
1. For each answer: SAME, UPDATED (a routine refresh, such as AUM or a date) or CHANGED (substance: team, committee, key persons, terms, a policy, a service provider, strategy, credit facility use, litigation).
2. Key person count: from the predecessor LPA and its amendment index (never from a past answer), list the named key persons, check for any amendment or approved substitution, mark who has ceased, and compare with the trigger. Say whether an event has occurred and how many more departures would trigger one.
3. Trace each CHANGED fact through every answer that touches it (team bios, committee, key persons, succession, deal attribution, references). The change must read the same everywhere.
4. Note answers ILPA asks to explain over time, such as valuation policy changes in the last five years.
<trap>Updating the committee answer and leaving the departed partner in the team bios, the key person answer or the deals he led. One change, told five ways, is a consistency failure the LP will find.</trap>
Stop: no predecessor response to compare: mark the log NOT RUN, say every CHANGED item will need the owner's own list, and continue.
Output: the change log (answer · last fund · now · SAME, UPDATED or CHANGED · every answer it touches · owner · tag), the key person count, then the handoff. The log is internal: the LP sees the changes in the answers themselves. Next: track-record-exhibits.
</unit>
Click to copy
<unit name="track-record-exhibits">
Follow the shared rules in prompt 01.
What it does: builds the fund level track record table and the deal level value creation attribution from the fund documents, with every figure's working shown.
Use when: any question on returns, track record, value creation or exclusions.
Needs: the latest quarterly report or audited financial statements for each fund in the record, the deal schedule, entry and exit figures for the deals the LP asks about.
Method:
1. Fund level from one document per fund, with its as-of date: paid-in (PI), cumulative distributions (D), NAV. TVPI = (D + NAV) / PI. DPI = D / PI. RVPI = NAV / PI. Net IRR as reported, with and without the subscription line where the document gives both.
2. Gross multiple, pooled: total value of all deals / total cost of all deals. Never the average of deal multiples.
3. Deal attribution, convention stated on the exhibit: EBITDA growth = (exit EBITDA - entry EBITDA) x entry multiple. Multiple change = (exit multiple - entry multiple) x exit EBITDA. Net debt change = entry net debt - exit net debt. Check the three sum to exit equity - entry equity, and that entry equity ties to the deal's cost.
4. For each fund the LP asks about, every investment is in the exhibit, or listed under exclusions with the reason. "Largest" deal means largest by gross value unless the LP says otherwise. If entry equity does not tie to cost, the attribution for that deal is OPEN for the CFO.
5. One as-of date per exhibit, printed on it.
<trap>Averaging deal multiples. It overweights small deals and nearly always overstates the gross multiple. The pooled figure is the only one that ties to the fund.</trap>
Stop: no fund report for a fund: no row for it, OPEN with the document named. IRR without the subscription line not in any document: mark it OPEN for the CFO and do not estimate it. The other rows still build.
Output: TABLE A fund level (fund · as of · PI · D · NAV · TVPI · DPI · RVPI · net IRR with and without the line · source), TABLE B deal schedule and pooled gross multiple, TABLE C attribution with the arithmetic, the exclusions list, then the handoff. Next: tie-out-gate.
</unit>
Click to copy
<unit name="tie-out-gate">
Follow the shared rules in prompt 01.
What it does: an independent check. It recomputes every figure and dated fact from the fund documents BEFORE it reads the draft, then blocks any answer whose figure does not tie.
Use when: after all earlier units, before the cover letter. Again after any edit to a figure.
Needs: {{FUND_DOCUMENTS}} and the LP's questions. Do not read the draft answers or exhibits until step 3.
Method:
1. From the fund documents only, recompute: every multiple, ratio and growth figure, the pooled gross multiple, each attribution line, and list every dated fact with its latest source and date.
In one chat session, do this in a fresh message before you scroll back to the draft. In Claude Code, run the gate as a separate session on the documents folder.
2. Check each draft source line: the document exists, the page holds the figure, the as-of date is the latest issued.
3. Now compare with every draft: the answers and exhibits from the earlier units and any draft the human's team supplied. A figure ties if it matches at the decimals shown. Any other difference: show both workings and the cause. A figure the LP asked for that the draft leaves out, or a figure missing its label (for example with or without the line), is a GAP: add it from the source; the CFO signs the added figure, and the row is not ready until then. It is a BLOCK only if no document gives it.
4. Check every flag against the normal list in prompt 01 again. Close any that match, with one "Checked, normal" line.
<trap>Checking the draft against the pitchbook, or against itself. A figure that matches the deck but not the fund report does not tie.</trap>
Stop: a figure that does not tie is a BLOCK on every answer that uses it. It does not go to the LP until the re-derived figure replaces it and the CFO signs. The rest of the packet still runs, and each blocked line says BLOCK and why.
Output: a verdict of PASS or BLOCK, the tie out table (answer · figure · draft · re-derived · ties, GAP or BLOCK · cause), closed flags, blocked lines, then the handoff. Next: cover-letter-and-open-items.
</unit>
Click to copy
<unit name="cover-letter-and-open-items">
Follow the shared rules in prompt 01.
What it does: writes the cover letter to the LP, the open items list and the sign-off sheet for the CFO and compliance officer.
Use when: last, after the gate.
Needs: the gate output, the match table, the change log, {{LP_NAME}}, {{DUE_DATE}}, the human's worry from prompt 01.
Method:
1. Sign-off sheet: one row per answer: status · tag · owner · gate result · signed (blank). Every CHANGED and BLOCK row first.
2. Open items list: item · what is needed · owner · date needed to make {{DUE_DATE}}. Say what happens by default if nobody acts: those answers do not go out.
3. Cover letter, plain and short: what is enclosed, the as-of date of each set of figures, where the LP will find the changes since the last fund, the items to follow and when, and the contact for questions. No claims about performance beyond the exhibits.
4. Readiness line: READY TO SEND only when every question's row is signed, nothing is OPEN and the gate passed. On the send date, check once more that no newer report, filing or roster has been issued since the gate ran. Otherwise NOT READY, with three counts: unsigned rows, OPEN rows, BLOCK rows.
<trap>A cover letter that calls the packet complete while answers are open or blocked. Say what is to follow. The LP trusts the packet more for it.</trap>
Stop: a BLOCK or OPEN row stays out of the sent packet and on the open items list. The letter and the rest of the packet still draft.
Output: the sign-off sheet, the open items list, the cover letter, the readiness line, then the handoff. Sign-off: the CFO and the compliance officer sign before anything goes to the LP.
</unit>

From a working build to an operating system.

For GP investor relations and fund CFOs: match a new LP questionnaire to your past answers, draft the rest, flag every stale or changed answer, and stop any figure that does not tie.

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 DDQ desk: your answer library loaded into your Claude account, read only connectors to your data room and fund administrator, a refresh of every dated fact each quarter, and a sign off step so no stale answer or untied figure reaches an LP.

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

Prompt set authored by consultance.ai. The ILPA DDQ 2.0 is published by ILPA; no affiliation is implied, and the reader uses their own copy. This is drafting support, not legal or compliance advice. Your fund data stays in your own Claude account; we never see it. The CFO and compliance officer own every answer.

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

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.

Setup guide
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

AI Portfolio Manager Prompt Pack

9 Claude prompts that write the investment policy statement, set the asset allocation, run the liquidity check, and rebalance: the portfolio manager work you'd usually pay an advisor $300 an hour for.

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 LP Due Diligence Questionnaire Desk?

LP Due Diligence Questionnaire Desk is a finance and data build in the consultance.ai AI Build Library. For GP investor relations and fund CFOs: match a new LP questionnaire to your past answers, draft the rest, flag every stale or changed answer, and stop any figure that does not tie. It fits GP heads of investor relations, fund CFOs, COOs and compliance officers who answer LP due diligence questionnaires during a fundraise and need every answer current, consistent and tied to the fund documents. Setup difficulty is Medium, with 4 plain-English steps.

What does LP Due Diligence Questionnaire Desk do?

For GP investor relations and fund CFOs: match a new LP questionnaire to your past answers, draft the rest, flag every stale or changed answer, and stop any figure that does not tie.

Who is LP Due Diligence Questionnaire Desk for?

It fits GP heads of investor relations, fund CFOs, COOs and compliance officers who answer LP due diligence questionnaires during a fundraise and need every answer current, consistent and tied to the fund documents.

How hard is LP Due Diligence Questionnaire Desk to set up?

Medium to set up — one guided setup instruction covering 4 plain-English steps, plus 9 ready-to-run prompts on the resource page.

How would consultance.ai build this out?

We would deliver a private DDQ desk: your answer library loaded into your Claude account, read only connectors to your data room and fund administrator, a refresh of every dated fact each quarter, and a sign off step so no stale answer or untied figure reaches an LP.

What are the licensing terms?

Prompt set authored by consultance.ai. The ILPA DDQ 2.0 is published by ILPA; no affiliation is implied, and the reader uses their own copy. This is drafting support, not legal or compliance advice. Your fund data stays in your own Claude account; we never see it. The CFO and compliance officer own every answer.

Connect this build to your workflow.

Discuss how LP Due Diligence Questionnaire Desk 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