HomeLibraryServicesCase studiesBlogAbout
consultance.ai
Book a discovery call →

Services

  • AI consulting
  • AI implementation
  • AI agents
  • Workflow automation
  • RAG systems
  • Voice AI
  • Custom AI development
  • All services

Library

  • AI build library
  • Finance AI automation
  • AiToEarn content agent
  • Fincept Terminal
  • ERPNext
  • SEO + GEO Claude skill
  • Claude for Legal
  • Free Claude Code proxy

Resources

  • Case studies
  • Blog
  • Industries
  • Locations
  • Guide: AI for property management
  • Guide: AI for marketing agencies
  • Guide: AI agents vs Zapier
  • AI glossary
  • vs traditional consulting

Company

  • About
  • Book a call
  • Contact
  • Privacy
  • Terms

© 2026 consultance.ai · AI, implemented.

audit → build → deploy

← Libraryconsultance.ai
Book a build call
Finance and data

Allocator Excel Desk

For allocators, hedge PMs, and PE operating partners: demo the desk on a public sample you can show in a meeting, then run the same gate on the internal workbook that stays on your machine. A local Python desk (openpyxl + pandas) writes DPI, TVPI, IRR, Kaplan Schoar PME, and a PASS or BLOCK back into the file. Hedge mode flags name concentration. The browser tool is the public demo only. Not a Burgiss or Aladdin replacement.

Free — runs in your own ClaudeMedium setup · 4 steps8 ready-to-run prompts+ live interactive tool
Set it up free — takes 3 minutes ↓Or have us wire it in →
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. 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. 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. 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. 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)
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.

I want to set up the Allocator Excel Desk. Demo it on public sample data I can show in a meeting. Run it on an internal workbook that must stay on my machine. Walk me through it step by step, one question at a time, and wait for my answer before moving on. I am an allocator, a hedge PM, or a PE operating partner, not a developer. Define every term once. Keep me to one action per message. Pin Claude Opus 5.

## What this produces
A public demo that is safe to screenshot, and a Desk Output sheet inside my own .xlsx for the internal run: DPI, TVPI, IRR, Kaplan Schoar PME, a residual check, and a PASS or BLOCK. Hedge mode writes name weights and a concentration BLOCK. Default tools are openpyxl and pandas on my machine. This does not pull Burgiss or Aladdin. The internal book does not go in the browser tool.

## Step one, ask me this first, nothing else
"Do you want the public demo first (safe to show), or are you scoring an internal workbook that must stay on this machine?"

Wait for my answer. Then continue, one step per message.

## Step two, get the desk script onto my machine
desk.py is a file I download. Claude does not install it and it is not on my machine yet. Give me this block, then wait until I confirm the file landed. Do not move to a scoring command before I have it.

```
mkdir -p ~/ExcelDesk
cd ~/ExcelDesk
curl -O https://consultance.ai/library/freebies/allocator-excel-desk/desk.py
ls -l desk.py
```

The last line should show a file named desk.py of roughly 11 KB. Every command after this one runs from that folder. If I open a fresh terminal window later, tell me to run `cd ~/ExcelDesk` before anything else.

If I would rather click than type, open https://consultance.ai/library/freebies/allocator-excel-desk/desk.py in a browser. It saves to my Downloads folder and I move it into ~/ExcelDesk.

## Step three, the public demo
From the same folder:

```
python3 desk.py book.xlsx --sample --mode cashflows
```

`--sample` writes a public stub workbook named book.xlsx and scores it. Nothing of mine touches it.

Success looks like this, and the final line names the file it wrote:

```
mode=cashflows verdict=BLOCK
  calls: 150.0000
  dists: 60.0000
  nav: 110.0000
  residual: 20.0000
  tvpi: 1.1333
  pme: 0.8959
  irr: 0.0417
  verdict: BLOCK
wrote Desk Output -> book.xlsx
```

BLOCK is the correct answer on the sample. The stub NAV of 110 does not tie to calls minus dists of 90, so the desk refuses to report TVPI. Explain that to me before we move on, because it is the whole point of the tool.

The browser version at /library/freebies/allocator-excel-desk-tool.html is the version with nothing to install, for a live meeting. Do not let me paste a live blotter or capital account into it.

## Step four, the internal run
Same folder, my file:

```
python3 desk.py {{MY FILE}}.xlsx --mode cashflows
python3 desk.py {{MY FILE}}.xlsx --mode blotter
```

Cashflows sheet needs date, type (CALL, DIST, NAV), amount, and exactly one NAV row at the as of date. Blotter sheet needs ticker, qty, price. If my workbook has no Index sheet with date and level columns, add `--public-cagr 0.08` and tell me plainly that a flat 8 percent path is a stub, not a licensed series. Open Desk Output on the laptop. I may load only that sheet into Claude.

## When it fails, these are the real errors
Match my error text to one of these. Do not guess a fifth.

1. `can't open file ... desk.py: [Errno 2] No such file or directory`
   I am in the wrong folder. Have me run `cd ~/ExcelDesk` then `ls -l desk.py`, then run the command again.

2. `ModuleNotFoundError: No module named 'pandas'`, or the same line naming openpyxl
   The two libraries are missing. First try `pip install openpyxl pandas`. If that returns `error: externally-managed-environment`, my Python is managed by the system or by Homebrew and will not take a global install. Use a local environment inside the same folder:

```
cd ~/ExcelDesk
python3 -m venv .venv
source .venv/bin/activate
pip install openpyxl pandas
```

   Then run the scoring command again. Every new terminal needs the `source .venv/bin/activate` line first, and I should see (.venv) at the start of my prompt when it is on.

3. `Missing workbook: book.xlsx`
   Nothing by that name is in this folder. On the demo I dropped `--sample`. On my own file the filename is wrong, and `ls` in the folder shows me what is actually there.

4. `Need an Index sheet (date, level) or --public-cagr`
   My workbook has no public path to compare against. Add `--public-cagr 0.08` to the command, or add an Index sheet with date and level columns.

If the shell reports that python3 is not found at all, tell me to install Python 3 from python.org, then stop and wait. Do not have me install anything with sudo.

## Rest of the path
Create a Claude Project named Excel Desk. Paste the house rules: never invent a cash flow, never pull a licensed feed, never put the internal book in a public demo, never report TVPI if residual is not zero, a named human signs. Run prompt 04 or 05 on Desk Output. Finish on prompt 08, the trust gate.

## The first session drill
By the end of this session I should have: the public sample scored and shown, then one internal workbook scored locally, a Desk Output sheet with PASS or BLOCK, and a six line note. Then show me the one line command so next month is a checklist.

Before you tell me it is done, re derive residual, TVPI, and PME. Do not say it is ready until those match the sheet. I sign it.
Step 2 · run it on your data

Step 1 set it up. These 8 prompts do the work.

the vault

The 8 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 8 prompts, numbered, in order · nothing left out.
You are standing up the Allocator Excel Desk for {{FIRM}}. Pin Claude Opus 5. Ask one question at a time and wait.

1. Path. Pick one: (A) DEMO on the public sample (B) RUN on an internal workbook that must stay on this machine.
2. Seat. Pick one: (A) allocator / family office / OCIO (B) hedge PM / risk (C) PE operating partner on a portco pack.
3. Job. (A) score fund cash flows vs a public path (B) score a hedge blotter for name concentration (C) score a PE Friday actuals pack.
4. Data source. If DEMO: use --sample or the browser tool, never their book. If RUN: (A) run desk.py locally, then upload only Desk Output (B) Claude Excel add-in on the live internal file they already control (C) they refuse to load the book, stop.
5. If they pick local Python, desk.py is a download, not something you install for them. Give this block first and wait until they confirm the file is on disk:

     mkdir -p ~/ExcelDesk
     cd ~/ExcelDesk
     curl -O https://consultance.ai/library/freebies/allocator-excel-desk/desk.py
     ls -l desk.py

   Expect desk.py at roughly 11 KB. Every later command runs from that folder, and a fresh terminal needs cd ~/ExcelDesk first.
   Demo first: python3 desk.py book.xlsx --sample --mode cashflows
   Then theirs: python3 desk.py {{PATH}} --mode cashflows   or   python3 desk.py {{PATH}} --mode blotter
   Success ends with the line: wrote Desk Output -> book.xlsx
   Four real errors, do not invent a fifth. "can't open file ... desk.py" means wrong folder, send them back to cd ~/ExcelDesk. "No module named 'pandas'" means run pip install openpyxl pandas, and if that returns "error: externally-managed-environment" have them run python3 -m venv .venv then source .venv/bin/activate then pip install openpyxl pandas in the same folder. "Missing workbook" means the filename is wrong or --sample was dropped. "Need an Index sheet (date, level) or --public-cagr" means add --public-cagr 0.08 and say plainly that a flat path is a stub.
6. Confirm house rules: never invent a cash flow, never pull a licensed feed, never put the internal book in a public demo, never report TVPI if the NAV roll does not tie, a named human signs.

Then stop. Do not score until they confirm path, seat, job, source, and signer {{SIGNER}}. If they chose DEMO, refuse any file that looks like a live blotter or capital account.
Work from the data source selected in prompt 01. List every sheet, the columns on each, the as-of date, and which sheet is the book of record. Flag: duplicate headers, merged cells, a NAV that does not sit on one row, a total that is a formula pointing at another workbook you cannot see, and any sheet named FINAL or v47.

Do not compute DPI, TVPI, PME, or weights yet. End with: READY TO SCORE or NOT READY, and the one hole that blocks you.
Work from the data source selected in prompt 01. If this is DEMO, they stay on --sample or the browser tool. Do not accept an internal file on the demo path.

If this is RUN and they can run Python, they need the script on disk before any scoring command. Give only:

mkdir -p ~/ExcelDesk
cd ~/ExcelDesk
curl -O https://consultance.ai/library/freebies/allocator-excel-desk/desk.py
pip install openpyxl pandas
python3 desk.py {{PATH}} --mode {{MODE}}

If pip returns "error: externally-managed-environment", have them run python3 -m venv .venv then source .venv/bin/activate then pip install openpyxl pandas from ~/ExcelDesk, and run the scoring line again.

They may upload Desk Output. They should not upload the blotter or the capital account unless they explicitly choose the add-in on a file they already control.

xlwings is allowed only when they say a VBA macro must run inside installed Excel. OpenBB is allowed only when they paste public index levels they already have the right to use. PyPortfolioOpt is not part of this run.

Confirm the Desk Output sheet exists before you continue.
Work from the data source selected in prompt 01. Read Desk Output. Restate calls, dists, NAV, residual, DPI, TVPI, IRR, PME, verdict, reason.

Trap: TVPI above 1.0 with residual not zero is a plugged NAV, not a return. Trap: TVPI above 1.0 with PME at or below 1.0 means the public path did the work.

Write a six line IC note. Last line is PASS or BLOCK copied from the desk, not a softer word. If verdict is BLOCK, do not recommend a recommit.
Work from the data source selected in prompt 01. Mode blotter. Restate names, gross MV, top ticker, top weight, HHI, verdict.

Policy cap is {{MAX_NAME}} (default 15 percent). If the desk BLOCKed, name the ticker and the weight. Do not average it away with "the book is diversified."

Write the risk note a PM at a multi strat would have to sign before adding risk. End BLOCK or PASS.
Work from the data source selected in prompt 01. Using only Desk Output plus the Cashflows sheet, answer: unfunded if {{COMMITMENT}} is stated, DPI, years since first call, and whether the multiple in the GP email matches TVPI on the desk.

If the GP email TVPI and the desk TVPI differ, hold the email. If residual is not zero, hold TVPI. If PME is at or below 1.0, say so in the first sentence.

This is not a Burgiss extract. It is the Excel the allocator already received.
Work from the data source selected in prompt 01. The file is a portco KPI pack, not a fund tape. Map actual vs budget for {{PERIOD}}. Rank the three lines that moved cash. Hold any line that does not tie to the trial balance or the pack's own total.

Do not rebuild an LBO. Do not invent next year. Write the note the operating partner sends before Monday's call.
Re-derive residual as NAV minus (calls minus dists). Re-derive TVPI as (dists plus NAV) / calls. Re-derive PME from the Index sheet or the public CAGR stub they used. If any of those three do not match Desk Output, BLOCK and name the gap. Do not argue that a FAIL does not matter. A named human {{SIGNER}} unblocks it.
Built on open source
pandas + openpyxl (default). Optional: OpenBB, xlwings, PyPortfolioOpt ↗

The code is public and free. The setup instruction above installs and wires it for you. You never need to open this link.

Got the prompts. Want them wired into your actual stack? We map that on a free AI audit.

Book the free audit
🔥 optional · live interactive tool

Open the public demo ↗

Browser-based. No signup. Drop in your numbers and see the trade in real time. Opens in a new tab so the prompts stay where you left them.

Launch the live tool ↗

Rent it forever, or own it once.

For allocators, hedge PMs, and PE operating partners: demo the desk on a public sample you can show in a meeting, then run the same gate on the internal workbook that stays on your machine

Path A · free

You just did it

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

We wire it into your business

The desk scores a file you load by hand. Consultance stands it up on the workbooks your office already runs: the extract from Burgiss or the PMS dump mapped once, a scheduled job that writes Desk Output before the IC pack, review gates where your policy requires them, and the whole thing handed over so your team owns it. Reply "wire it" for a 30-minute slot.

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.
  • • Wiring the open-source piece to real systems? Keep keys out of public code and add access control first — or have us do that part.
the fine print

Credit the original author

Prompt set and desk.py authored by consultance.ai. openpyxl and pandas are used under their own licenses. OpenBB, xlwings, and PyPortfolioOpt are optional sleeves under their own licenses and are not required for the first run. This is not a performance audit, not a Burgiss extract, and not investment advice. A named human signs every output. Your workbook stays on your machine.

Want this running in your business, not just your laptop? We build it and hand you the keys.

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.

the newsletter

AI news worth opening.

The AI tools, launches, and shifts that actually matter, in plain English. New library drops the moment they land.

100% freeNo paywall, everUnsubscribe anytime

More like this

Other builds worth a weekend

All repos →
Finance and data

Free Portfolio Quant Research Desk

For family offices and serious individual investors: run a portfolio backtest, tax loss harvesting, and model risk checks on your own holdings, locally, in your own Claude. Replaces the $250k quant seat you would otherwise hire.

Setup guide →
Finance and data

Private Equity Deal Sourcing Playbook

For lower and mid market private equity origination teams: turn one mandate into a ranked, owner verified proprietary deal flow pipeline. Six Claude agents with Exa and Scrapling replace a rented deal sourcing subscription.

Setup guide →
Finance and data

Free Jira Alternative for Deal Teams

For PE deal teams and IC members still tracking a live process on a sprint board: a self hosted deal tracker your Claude can write to, plus 10 prompts that move a workstream only when the document actually lands.

Setup guide →
Get the free kitBook a call

manual now, installed when you're ready

What you have

The free version above. It runs this read on a file you paste in, in your own Claude. Yours to keep.

What we install

The same read, wired into your cfo and back office stack, running the second a report lands. Nothing to upload, no new login.

See the installed version

Forward this to whoever owns the workflow.

The person drowning in this every week is the one who'll actually want it.

Forward by email
in one line

What is Allocator Excel Desk?

Allocator Excel Desk is a finance and data build in the consultance.ai AI Build Library. For allocators, hedge PMs, and PE operating partners: demo the desk on a public sample you can show in a meeting, then run the same gate on the internal workbook that stays on your machine. A local Python desk (openpyxl + pandas) writes DPI, TVPI, IRR, Kaplan Schoar PME, and a PASS or BLOCK back into the file. Hedge mode flags name concentration. The browser tool is the public demo only. Not a Burgiss or Aladdin replacement. It fits Family office CIOs, OCIOs, hedge PMs, and PE operating partners who will not paste the book into a vendor demo, and whose real file still lives in Excel after the PMS or data vendor extract. Setup difficulty is Medium, with 4 plain-English steps.

What does Allocator Excel Desk do?

For allocators, hedge PMs, and PE operating partners: demo the desk on a public sample you can show in a meeting, then run the same gate on the internal workbook that stays on your machine. A local Python desk (openpyxl + pandas) writes DPI, TVPI, IRR, Kaplan Schoar PME, and a PASS or BLOCK back into the file. Hedge mode flags name concentration. The browser tool is the public demo only. Not a Burgiss or Aladdin replacement.

Who is Allocator Excel Desk for?

It fits Family office CIOs, OCIOs, hedge PMs, and PE operating partners who will not paste the book into a vendor demo, and whose real file still lives in Excel after the PMS or data vendor extract.

How hard is Allocator Excel Desk to set up?

Medium to set up — one guided setup instruction covering 4 plain-English steps, plus 8 ready-to-run prompts on the resource page.

How would consultance.ai build this out?

The desk scores a file you load by hand. Consultance stands it up on the workbooks your office already runs: the extract from Burgiss or the PMS dump mapped once, a scheduled job that writes Desk Output before the IC pack, review gates where your policy requires them, and the whole thing handed over so your team owns it. Reply "wire it" for a 30-minute slot.

What are the licensing terms?

Prompt set and desk.py authored by consultance.ai. openpyxl and pandas are used under their own licenses. OpenBB, xlwings, and PyPortfolioOpt are optional sleeves under their own licenses and are not required for the first run. This is not a performance audit, not a Burgiss extract, and not investment advice. A named human signs every output. Your workbook stays on your machine.

Want this built into your workflow?

Allocator Excel Desk is the starting point. On a free AI audit we map where it fits your stack and what consultance.ai would build around it.

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 your free AI audit