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/GP Fundraise Prep Pack

Finance and data

GP Fundraise Prep Pack

For emerging manager GPs raising a fund: an LP target list ranked by room this vintage, a brief for every LP meeting, and track record pages tied to the fund's own records.

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/gp-fundraise-prep#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 GP Fundraise Prep Pack on my fund, in my own Claude, with my own accounts. It is prep only: you never contact an investor and never send anything. Go one step at a time and wait for me after each step.

First question, and only this one: where does my fundraise material live? Files I can upload here (LP reports, a CRM export, the prior fund's financial statements, the draft deck), a folder on my computer, or nothing yet beyond my strategy, fund size and vintage?

If I say upload or nothing yet: this is a Claude app job. Tell me to make a private Project, select Claude Opus 5.5 in the model picker, and turn on web search. 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 why an LP should say yes this vintage and what worries me, then my strategy, fund size, vintage, target ticket, the date, my LP sources and the named sign-off person, 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, lp-target-list, on my real fund.
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
GP FUNDRAISE PREP PACK: SHARED RULES
Every unit in this pack follows these rules. They are written here once.

1. WHERE TO RUN IT
- One LP, one meeting or one page: the Claude app, in a private Project, with web search turned on. Chat is fine here.
- A folder of LP reports, a CRM export and the fund's records: Claude Code, opened in that folder.
- Fund records or LP data that must stay on a managed machine: Claude Code on that machine.
- The same update every week while the fund is raising: Claude Code, so each run lands as a dated file 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 pension reports or board minutes.
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 records, LP names from your CRM or anything under NDA.

2. BEFORE THE FIRST RUN
First ask the human two questions and record the answers word for word: why should an LP say yes to this fund in this vintage, and what worries you most about the raise? Every later unit works on the LPs and pages that bear on those answers first.
Then ask for these, one at a time: {{STRATEGY}} (asset class, stage or size, sector, region), {{FUND_SIZE}} (target and hard cap), {{VINTAGE}} (the year of first close), {{TARGET_TICKET}} (the commitment you plan to ask a typical LP for), {{AS_OF}} (today's date for the run), {{LP_SOURCES}} (the public LP documents and the CRM export you have), and the named person who signs off. Also ask who the fund's counsel or compliance lead is.
A missing item gets a plain default, labelled ASSUMED. No target ticket: ASSUMED at 7.5% of the fund target, and say so.

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 LP report, CRM export or fund record: name it, and do not score it.
- A figure needs a document you do not have: ask once, name the document, mark it OPEN.
- An LP's allocation, target, pacing budget, minimum ticket or commitment comes ONLY from a public document of that LP (annual report, investment policy, pacing plan, board or committee minutes, commitment report) or from the GP's own CRM. Never from memory, a news summary without the document, or an estimate. Not found: NOT IN SOURCE, and the LP stays unscored on that line.
A stop blocks only the outputs that need the missing input. Everything else still runs.

4. LABELS ON EVERY FIGURE
REPORTED (document, page, and its date) · CALCULATED (show the arithmetic) · ASSUMED (who set it) · MARKETING (a claim in a deck, letter or LP website: it makes a question, never a number) · NOT IN SOURCE.
CRM figures are REPORTED as "CRM, field, date of the export". A verbal indication from an LP is a soft circle, never a commitment. Only a signed subscription document is a commitment.

5. THREE FLAG TAGS
- CHANGES THE DECISION: reaches the final output (it moves an LP in or out of the list, or blocks a page).
- WORTH A QUESTION: only if the answer could move an LP's rank, its ticket fit, a track record multiple by more than 0.05x, or an IRR by more than 1 point. Smaller items go on a question list, untagged.
- EXPLAINED BY CONTEXT: one line under "Checked, normal". Once closed, it stays closed.
Test every flag against the list below before you raise it. If the "when it is not" test needs data you do not have, close the flag and note the test as not run.

6. LOOKS WRONG, IS NORMAL
- An LP sits above its private markets target | public markets fell, so the share rose (the denominator effect) | normal while it is inside the policy range and the pacing plan still budgets new commitments. Not normal above the top of the range, or when the plan pauses commitments.
- An LP's commitments are lumpy, two large ones one year and none the next | pacing is planned over several years | not normal if the pacing plan for this year is zero.
- A commitment figure in board minutes differs from the annual report | approved amount versus closed amount, or a currency translation | not normal if no later document shows the commitment at all.
- A young fund shows a low or negative IRR and TVPI near 1.0x | fees and costs come first, gains later (the J curve) | not normal after year 5 with no realisations.
- A deal in its first year is marked up to 15% below cost | transaction costs, held near cost until the next valuation event | not normal if it is further below, or keeps falling with no event.
- Gross multiple and IRR are well above net | management fee, fund expenses and carry sit between them | not normal if net is missing, or the gap is tiny on a fund that charges carry.
- An LP goes quiet for 4 to 8 weeks after a first meeting | it is working through its own diligence and committee calendar | normal only when a dated next step exists. No dated next step: it is a stale follow up.
- An LP says no new managers, but has an emerging manager program or invests through a fund of funds | the program has its own budget | check the program's own document.
- Re-ups with existing managers take most of an LP's pacing budget | LPs back their current managers first | read the budget left after re-ups, not the total.

7. STANCE AND SCOPE
A neutral colleague who has prepared many fundraises. Do not hunt for problems. Do not take a claim as proof.
Prep only. This pack does not place the fund, does not contact or solicit any investor, and does not act as a placement agent or broker dealer. It never sends an email, a message or a deck. Introductions, outreach and the close stay with the GP. Whether and how the GP may approach an investor is set by the fund's offering exemption and its counsel: raise it with counsel, do not decide it. If anyone is to be paid for introductions based on capital raised, flag it for counsel.
Track record pages tie to the fund's own records. A track record figure that does not tie is a BLOCK.

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.

SAMPLE (fake fund, fake LPs; every figure cites its page or CRM field)
The GP: Larkspur Capital, raising Fund II. STRATEGY lower mid market buyout, US. FUND_SIZE target $200M. VINTAGE 2027. TARGET_TICKET $15M. AS_OF 14 Jan 2027.

CASE A: three LPs from public documents
LP1 Northgate County Employees' Retirement System
- Annual report FY2026 p.4 (dated Sep 2026): total plan assets $40.0B. Private equity NAV $5.6B.
- Investment policy p.6 (dated Sep 2026): private equity target 12%, policy range 8% to 16%. Minimum commitment $50M. No commitment above 10% of a fund's total commitments.
- Pacing plan p.2 (dated Oct 2026): 2027 private equity commitment budget $1.2B. Commitment report p.1 (dated Dec 2026): $1.15B of 2027 commitments already approved.
LP2 Riverbend Teachers' Fund
- Annual report FY2026 p.3 (dated Oct 2026): total plan assets $6.0B. Private equity NAV $0.42B.
- Investment policy p.2 (dated Oct 2026): private equity target 10%. Commitments of $10M to $30M, and no more than 15% of a fund. Emerging manager program for Fund I to Fund III managers.
- Pacing plan p.1 (dated Oct 2026): 2027 commitment budget $150M, of which $30M is approved.
- Board minutes p.3 (dated Oct 2026): commitments to two lower mid market buyout funds, in 2024 and 2026.
LP3 Alder Family Foundation
- Annual report 2025 p.9 (dated Jun 2026): total assets $900M. Private equity NAV $210M. Private equity target 20%, range 15% to 25%.
- Annual report p.10: "We plan new private equity commitments of $25M in 2027." Commitments of $5M to $15M per fund. No 2027 commitments made yet (p.10).
- Annual report p.11: one commitment to a lower mid market buyout fund in 2025. The report does not mention emerging or first time managers.

CASE B: Fund I track record and the CRM
- Fund I schedule of investments, audited financial statements 31 Dec 2025, p.7:
  Deal 1: invested $12.0M, realised $30.0M, unrealised $0. Fully realised.
  Deal 2: invested $10.0M, realised $0, unrealised $18.0M.
  Deal 3: invested $8.0M, realised $2.0M, unrealised $0. Fully realised.
  Deal 4: invested $10.0M (closed Mar 2025), realised $0, unrealised $9.0M.
- Same financial statements p.4: net TVPI to limited partners 1.32x.
- Draft Fund II deck p.3 (the GP's draft): "Realised track record: 2.5x gross MOIC". "Fund I gross MOIC 1.48x". "Fund I net IRR 19%".
- The dated ledger of Fund I capital calls and distributions is NOT provided.
- CRM export dated 14 Jan 2027:
  Riverbend: first meeting 2 Dec 2026. Last touch 9 Dec 2026 (data request answered). Next step: Riverbend investment committee, 4 Feb 2027. Note: "verbal indication of $20M".
  Alder: first meeting 10 Nov 2026. Last touch 10 Nov 2026. GP promised to send the track record pages. Next step: none set.
  Signed subscription documents for Fund II: $45M in total, from Fund I investors re-upping.

ANSWER KEY
| Check | Expected | Working |
| A. Northgate allocation | 14.0%, inside range | CALCULATED: 5.6 / 40.0 = 14.0%. Target $ = 0.12 x 40.0 = $4.8B, so $0.8B over target, inside 8% to 16%. EXPLAINED BY CONTEXT |
| A. Northgate ticket fit | FAIL. EXCLUDED, CHANGES THE DECISION | Cap = 10% x $200M = $20M. Minimum is $50M. 50 > 20, so it cannot commit to this fund at all. Remaining 2027 budget 1.2 - 1.15 = $0.05B ($50M) |
| A. Riverbend allocation | 7.0%, $180M under target | CALCULATED: 0.42 / 6.0 = 7.0%. Target $ = 0.10 x 6.0 = $600M. 600 - 420 = $180M |
| A. Riverbend room this vintage | $120M | CALCULATED: 150 - 30. Ticket range $10M to $30M (cap 15% x 200 = $30M). $15M ask fits |
| A. Alder allocation | 23.3%, inside range | CALCULATED: 210 / 900 = 23.3%. Target $ = 0.20 x 900 = $180M, $30M over, inside 15% to 25%. EXPLAINED BY CONTEXT |
| A. Alder room this vintage | $25M | REPORTED p.10, none used. $15M ask fits $5M to $15M |
| A. Appetite scores | Riverbend 10 of 10, Alder 6 of 10, Northgate not scored | Riverbend: strategy 3 + vintage room 3 + new manager 2 + evidence 2. Alder: strategy 2 + vintage room 2 + new manager 0 (NOT IN SOURCE, a question) + evidence 2 |
| A. Ranked list | 1 Riverbend, 2 Alder. Northgate excluded | |
| B. Fund I totals | Invested $40.0M, realised $32.0M, unrealised $27.0M, total value $59.0M | CALCULATED: sums of p.7 |
| B. Fund I gross MOIC | 1.475x, shown as 1.48x. Ties to the deck | CALCULATED: 59.0 / 40.0 |
| B. Realised gross MOIC | 1.60x. Deck's 2.5x does not tie: BLOCK on that line | CALCULATED: realised deals 1 and 3: (30.0 + 2.0) / (12.0 + 8.0) = 32.0 / 20.0. The 2.5x is Deal 1 alone (30.0 / 12.0), a cherry picked subset |
| B. Unrealised gross MOIC | 1.35x | CALCULATED: (18.0 + 9.0) / (10.0 + 10.0) |
| B. Gross 1.48x vs net 1.32x | EXPLAINED BY CONTEXT | Fees, expenses and carry. The page must show net next to gross |
| B. Deal 4 at 0.90x | EXPLAINED BY CONTEXT | CALCULATED: 9.0 / 10.0. Closed Mar 2025, first year, held near cost |
| B. Net IRR 19% | OPEN | Needs the dated capital call and distribution ledger. Every other check still runs |
| B. Riverbend follow up | 36 days quiet, EXPLAINED BY CONTEXT | 9 Dec 2026 to 14 Jan 2027 = 36 days. Dated next step 4 Feb 2027 |
| B. Alder follow up | 65 days, STALE, the GP owes the pages | 10 Nov 2026 to 14 Jan 2027 = 65 days, no next step. The pages cannot go until the 2.5x line is fixed |
| B. Weekly update | Signed $45M = 22.5% of target. Soft circles $20M, shown apart | CALCULATED: 45 / 200 |

The trap in Case A is ranking by size: Northgate is the biggest program, but its own $50M minimum is above its own 10% cap on a $200M fund, so it cannot commit at all. The trap in Case B is the 2.5x "realised" figure: it drops the realised loss on Deal 3. A third trap: counting the $20M verbal as raised, which gives 32.5% instead of 22.5%.
A run that flags a normal item, ranks Northgate, prints 2.5x on a page, or guesses the net IRR 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 web search and upload are enough.
Needs: the "Needs" lines of every unit in this pack, and the systems the human already uses: web search for public LP documents, the fund's CRM, the drive or data room that holds the fund's records and the deck, and the fund administrator's reporting portal if it has an export.
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. Start every tool read only. A CRM connection stays read only: this pack never writes to the CRM and never sends mail from it.
<trap>A tool that sends email or LinkedIn messages, enriches LP contacts from a third party database, or asks for write access to the CRM. Leave it out. Outreach is the GP's job, and a tool that sends it turns prep into solicitation.</trap>
Stop: no official page means do not install it. Mark that need UPLOAD and carry on. Fund records (financial statements, the cash flow ledger, the deck) are UPLOAD from the start. Every other unit still runs.
Output: the table, what was installed, each test result, and the handoff. Next: lp-target-list.
</unit>
Click to copy
<unit name="lp-target-list">
Follow the shared rules in prompt 01.
What it does: builds the LP target list from public LP documents and the CRM, removes LPs that cannot commit to this fund, and ranks the rest by the room they have this vintage.
Use when: first. Start here once prompt 02 passes.
Needs: {{STRATEGY}}, {{FUND_SIZE}}, {{VINTAGE}}, {{TARGET_TICKET}}, {{AS_OF}}; for each LP its annual report, investment policy, pacing plan and recent commitment report or board minutes; the CRM export of LPs already known to the GP.
Method:
1. Build the universe: LPs in the CRM, plus LPs whose own public documents show commitments to {{STRATEGY}}. Search for each LP's own documents with web search; record the URL, the page and the document date. Use nothing from memory.
2. Allocation per LP: current share = private markets NAV / total assets. Target $ = target % x total assets. Gap = target $ - NAV. Compare the share with the policy range, not only the target.
3. Ticket gate: cap = the LP's per fund limit (a % of fund size x the {{FUND_SIZE}} target, or a $ limit). If a hard cap is given, test the target size and say whether the result changes at the cap. The LP passes if its minimum ticket <= the lower of cap and its maximum, and {{TARGET_TICKET}} sits inside that range, ends included. A fail EXCLUDES the LP, CHANGES THE DECISION.
4. Room this vintage: remaining budget = the pacing plan's budget for {{VINTAGE}} - commitments already approved for {{VINTAGE}}. If only an annual report states planned commitments, use that, REPORTED. No pacing figure: room is NOT IN SOURCE and the LP goes to an "unranked, needs a document" list.
5. Rank the LPs that pass by remaining budget this vintage. Show the size of the program as a column, never as the rank.
<trap>Ranking by the size of the LP's program. The biggest LP is often the one that cannot commit: its minimum ticket is above its own cap on a fund your size, or its budget for the year is already used.</trap>
Stop: an LP with no pacing plan, commitment report or stated budget for the year: list it as unranked and name the document that would rank it. The ranking still runs on the rest.
Output: the list (LP · total assets · share vs target and range · ticket range and gate · remaining budget this vintage · sources with page and date · rank), the excluded LPs with the reason, the unranked LPs with the document needed, then the handoff. Next: lp-appetite-score.
</unit>
Click to copy
<unit name="lp-appetite-score">
Follow the shared rules in prompt 01.
What it does: scores each LP that passed the ticket gate on four lines, out of 10, so the GP spends meetings where yes is most likely this vintage.
Use when: after lp-target-list.
Needs: the target list, each LP's commitment history (board minutes, commitment reports), its policy on new managers, the human's answer on why an LP should say yes.
Method:
1. Strategy fit (0 to 3): 3 if the LP committed to two or more funds of the same strategy in the last 3 years (count by calendar year: a commitment in {{AS_OF}} year minus 3 or later counts); 2 if one; 1 if its policy allows the strategy but no commitment is found; 0 if the policy excludes it.
2. Vintage room (0 to 3): remaining budget this vintage >= 2 x {{TARGET_TICKET}} scores 3; >= {{TARGET_TICKET}} scores 2; above zero but below the ticket scores 1; zero or less scores 0.
3. New manager stance (0 to 2): 2 if the LP has an emerging manager program, or committed to a Fund I to III in the last 3 years; 0 if its documents are silent (NOT IN SOURCE, tagged WORTH A QUESTION because a no would exclude the LP); a policy that bars new managers EXCLUDES the LP unless a separate program applies.
4. Evidence (0 to 2), by the date of the document, not the period it covers: 2 if every source is dated within 12 months of {{AS_OF}}; 1 if within 24 months; 0 if older.
5. Total, then sort. Ties break on remaining budget. Add one line per LP on how the human's reason to say yes fits what the LP has backed.
<trap>Scoring silence as a no. A document that does not mention new managers is a question to ask, scored 0 and flagged, never a reason to drop the LP.</trap>
Stop: an LP with no commitment history in any document: score strategy fit NOT IN SOURCE, leave the total open, and name the commitment report that would close it. The other LPs still score.
Output: the scorecard (LP · strategy · vintage room · new manager · evidence · total · questions for the meeting), the final order, then the handoff. Next: track-record-pages.
</unit>
Click to copy
<unit name="track-record-pages">
Follow the shared rules in prompt 01.
What it does: rebuilds the track record from the fund's own records, deal by deal, and checks every figure on the draft pages against it.
Use when: before any page goes to an LP, and before lp-meeting-brief.
Needs: the audited financial statements of each prior fund (schedule of investments, net returns), the dated ledger of capital calls and distributions, the deal list, and the draft deck pages.
Method:
1. Deal table from the schedule of investments: invested, realised, unrealised, total value, gross MOIC = total value / invested, status (fully realised, partly, unrealised).
2. Totals: fund gross MOIC = sum of total value / sum of invested. Realised gross MOIC uses every fully realised deal, losses included. Unrealised gross MOIC uses every unrealised deal. Show each sum.
3. Net figures come only from the financial statements or from the cash flow ledger: net TVPI as reported, net IRR recomputed from the dated cash flows. Show net next to every gross figure, same period, same method. Where net exists only at fund level (the usual case), show the fund net beside the gross and say net is not available by deal or subset.
4. Check each figure on the draft pages: ties, does not tie, or cannot be checked. A subset of deals shown as the record (best deals, one sector, one partner) must say so on the page and the full fund figures must sit beside it.
5. If the GP is an SEC registered adviser, its counsel should confirm the pages against 17 CFR 275.206(4)-1, which bars gross performance without net performance "with at least equal prominence", and extracted performance without the total portfolio's results. Raise it; do not rule on it.
<trap>A "realised" multiple that drops the realised losers. Picking the best exits, or leaving out a written down deal, makes the record look stronger than the fund's own statements. That figure does not tie, so it is a BLOCK.</trap>
Stop: no cash flow ledger: net IRR is OPEN and the page line that shows it waits. A figure on the page that does not tie to the records is a BLOCK on that line until the named person fixes it. Every other line still runs.
Output: the deal table, the totals with their arithmetic, a page check table (page · figure · rebuilt figure · ties, BLOCK or OPEN), the gross to net list, then the handoff. Next: lp-meeting-brief.
</unit>
Click to copy
<unit name="lp-meeting-brief">
Follow the shared rules in prompt 01.
What it does: writes a one page brief for one LP meeting: why this LP, why this vintage, what it has backed, what it will ask, and the next step to ask for.
Use when: before each LP meeting, after the LP is scored and the track record pages are checked. One brief per LP; on the sample, brief every scored LP.
Needs: the LP's row in the scorecard and its sources, the CRM history with this LP, the checked track record pages, the human's reason to say yes and worry.
Method:
1. The LP in five lines: program size, share vs target and range, remaining budget this vintage, ticket range, the managers of the same strategy it backed recently. Every figure with its source.
2. Why yes this vintage: two or three sentences that tie the human's reason to what this LP has backed and the room it has. No claim the documents do not support.
3. What it will ask: the questions its policy and history point to (new manager, team, fees, track record, ESG or reporting needs stated in its policy). Answer each from the checked pages only. A BLOCK or OPEN line is named as one the GP must not quote yet.
4. The ask for the meeting: one dated next step that fits the LP's committee calendar where its documents state one.
5. CRM history: every past touch, what was promised and whether it was sent.
<trap>Pitching the whole deck. The LP decides on fit with its own plan; a brief that does not open with its room, its ticket and what it has backed sends the GP in blind.</trap>
Stop: the LP has no scorecard row: run lp-appetite-score first. A track record line that is BLOCK or OPEN stays out of the brief's answers and is named. The rest of the brief still runs.
Output: the one page brief, the questions with answers and their sources, the next step to ask for, then the handoff. Next: follow-up-tracker.
</unit>
Click to copy
<unit name="follow-up-tracker">
Follow the shared rules in prompt 01.
What it does: reads the CRM export and lists, for every LP in play, the last touch, what is owed by whom, the next step and its date, and which follow ups have gone stale.
Use when: every week, and after every LP meeting.
Needs: the CRM export with its date, the meeting notes, {{AS_OF}}.
Method:
1. One row per LP: stage (first meeting, diligence, committee, soft circle, signed), last touch date, days quiet = {{AS_OF}} - last touch date, owed by GP or by LP, the item owed, next step and its date.
2. Status: ON TRACK if a dated next step exists in the future; STALE if there is no dated next step and more than 21 days quiet, or the GP owes an item promised more than 7 days ago (count from the promise date); CLOSED if the LP declined.
3. For each STALE row, name the one action the GP should take and what it needs first (for example, a page that must pass the gate before it is sent).
4. Keep soft circles and signed commitments in separate columns. A soft circle is a verbal or written indication without a signed subscription document.
<trap>Treating every quiet LP as lost or every quiet LP as fine. Quiet with a dated committee step is normal. Quiet with nothing dated, or with the GP's own promise unkept, is the follow up that dies.</trap>
Stop: no CRM export: ask for it once and mark the tracker OPEN. Do not rebuild LP history from memory. The other units still run.
Output: the tracker table, the STALE list with one action each, then the handoff. The pack drafts nothing to send: the GP writes and sends every message. Next: fundraise-gate.
</unit>
Click to copy
<unit name="fundraise-gate">
Follow the shared rules in prompt 01.
What it does: an independent check. It redoes the key figures from the source documents BEFORE it reads the earlier tables, then blocks any page or update line that does not tie.
Use when: after the earlier units, before any page goes to an LP and before the weekly update.
Needs: the source documents only (LP documents with pages, the CRM export, the fund's financial statements and ledger, the draft pages). Do not read the earlier outputs until step 3.
Method:
1. From the sources only, recompute: each LP's share vs target, ticket cap and remaining budget this vintage; each appetite score; every track record total and multiple; days quiet per LP; signed commitments as a % of the fund target.
2. Check every source line: document, page and date recorded; label right; no LP figure without an LP document or the CRM behind it.
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 source folder.
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.
5. Scope check: nothing in the outputs contacts an LP, claims a commitment that is not signed, or says the pack places the fund.
<trap>Checking the earlier tables against themselves. The gate only counts if it starts from the source documents.</trap>
Stop: a figure that does not tie, or cannot be traced to a source, is a BLOCK on the page or update line that uses it. A figure that traces to a subset (for example one deal) still does not tie to the fund and is a BLOCK. Do not produce that line until the named person fixes the gap. A missing document stays OPEN, not BLOCK. The rest still runs, and each blocked line says BLOCK and why.
Output: a verdict of PASS or BLOCK, a tie out table (figure · earlier · re-derived · ties or not), closed flags, blocked lines, open lines, then the handoff. Next: weekly-partner-update.
</unit>
Click to copy
<unit name="weekly-partner-update">
Follow the shared rules in prompt 01.
What it does: writes the weekly fundraise update for the partners and a one file HTML view of the pipeline from the GP's own numbers.
Use when: weekly, after the gate.
Needs: the gate output, the tracker, the scorecard, the target list, the human's worry from prompt 01.
Method:
1. The number first: signed commitments $ and as a % of the fund target, from signed subscription documents only. Soft circles on their own line, never added in. Change since last week.
2. Pipeline by stage (one stage per LP, its furthest; a soft circle shows in its own column): LP count and the sum of target tickets ({{TARGET_TICKET}} per LP unless the CRM states an LP specific ask), with the expected next step dates in the coming four weeks.
3. What changed this week, the STALE follow ups and who owns each, the BLOCK and OPEN lines and what fixes them.
4. Two sentences on the raise against the human's worry, and what happens by default if nobody acts on the stale and open items.
5. Write one HTML file: the pipeline table by stage, coloured by status, each cell with its source. No outside scripts or fonts.
<trap>Adding soft circles to the raised number. Partners plan the first close on it; a verbal is not money until the subscription document is signed.</trap>
Stop: the gate blocked a line: the update shows it as BLOCK with the reason. No CRM export this week: the update says so and repeats nothing from last week as current.
Output: the update (one page), the HTML file, then the handoff. Sign-off: the named person signs before it goes to the partners. The pack sends nothing.
</unit>

From a working build to an operating system.

For emerging manager GPs raising a fund: an LP target list ranked by room this vintage, a brief for every LP meeting, and track record pages tied to the fund's own records.

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 fundraise room: the pack loaded into your Claude account, a read only connection to your CRM and fund records, a weekly run that refreshes the tracker and the partner update, and a sign off step so no untied track record 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. Prep only: it does not place the fund, solicit investors or act as a placement agent or broker dealer, and it is not legal advice; your counsel decides how you approach investors. It uses only public LP documents and your own CRM. Your data stays in your own Claude account; we never see it. A named human signs every page.

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

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.

Setup guide
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
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 GP Fundraise Prep Pack?

GP Fundraise Prep Pack is a finance and data build in the consultance.ai AI Build Library. For emerging manager GPs raising a fund: an LP target list ranked by room this vintage, a brief for every LP meeting, and track record pages tied to the fund's own records. It fits Emerging manager GPs and VC or PE founders raising a fund who want the LP list, meeting briefs, track record pages and follow ups prepared in their own Claude, while the introductions and the close stay with them. Setup difficulty is Medium, with 4 plain-English steps.

What does GP Fundraise Prep Pack do?

For emerging manager GPs raising a fund: an LP target list ranked by room this vintage, a brief for every LP meeting, and track record pages tied to the fund's own records.

Who is GP Fundraise Prep Pack for?

It fits Emerging manager GPs and VC or PE founders raising a fund who want the LP list, meeting briefs, track record pages and follow ups prepared in their own Claude, while the introductions and the close stay with them.

How hard is GP Fundraise Prep Pack 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 fundraise room: the pack loaded into your Claude account, a read only connection to your CRM and fund records, a weekly run that refreshes the tracker and the partner update, and a sign off step so no untied track record figure reaches an LP.

What are the licensing terms?

Prompt set authored by consultance.ai. Prep only: it does not place the fund, solicit investors or act as a placement agent or broker dealer, and it is not legal advice; your counsel decides how you approach investors. It uses only public LP documents and your own CRM. Your data stays in your own Claude account; we never see it. A named human signs every page.

Connect this build to your workflow.

Discuss how GP Fundraise Prep Pack 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