← Rig your own
Somebody asked

“Once a quarter I pull numbers out of a few spreadsheets and write them up into a board deck.”

Our answer

Don't automate this

Not worth automating. At 0.5 a month, a $6,000 setup takes 74 months to come back.

We treat anything past 18 months as not worth it: this kind of work usually changes before then.

high confidenceYou gave us the numbers this rests on.
Costs you today / month
$81
Would cost to run / month
$0.36
You keep / month
$81
Pays back in 74 months · setup ~$6,000
What went into that
How often
0.5 a month

Not often enough for small savings to add up.

you said "less often than that" — that's about 1 a month

Pays for itself in
74 months

$81 of time a month against $0.36 to run it.

we treat inside 12 months as worth doing and past 18 as not — those cut-offs are our judgement, not arithmetic

Can't be undone
1 action

draft deck — this one needs to ask first.

Time it takes now
150 min each

Long enough that handing it over is worth something.

you said "1 to 4 hours"; we used the middle of that, 150 minutes

Needs judgement
summarise changes

There's a step here that needs reading and interpreting, which is what AI is actually for.

What would change this
  • At 0.5 a month this doesn't pay for itself — it would take 74 months. You'd need about 3.1 a month before it's worth building.
  • That $6,000 is what it costs to get this running reliably — setup, a test set, and the first month of fixing it. Set it up yourself over a weekend instead and it's nearer $650 — payback drops to about 8 months.

$6,000 ÷ (12 months × $162 saved per run) = 3.1 a month

Take it to a builder

We don't build it. Your builder does that better than we ever could — it knows every one of its own connectors. Copy this and paste it in.

Where this goes

Open n8n, start a new workflow, and click the ✨ AI button. Paste this in.

n8n-prompt.md

Paste into n8n's AI workflow builder. Names the real nodes and says what n8n won't enforce.

Build me this workflow. I've had it designed already — the detail below is deliberate, so please follow it rather than simplifying it.

## What it should do

Turn the quarter's spreadsheets into a written board summary.

**Finished looks like:** A deck the board can read, with the numbers and what they mean.

## When it should run

Once a quarter — trigger it by hand until that's settled

Use a Manual Trigger node.

## The steps, in order

1. **read sheets** — Pull the quarter's figures.
2. **summarise changes** — Work out what moved and why it matters.
3. **draft deck** — Write the summary up.
   - This one is permanent. It cannot be undone once it runs.

## Accounts to connect

- Spreadsheets
- Notion

Use my own credentials for each. Don't hardcode anything.

## Must-haves — please don't skip these

- One step changes something outside the system: draft deck. Point it at a test account first.
- Don't retry automatically. Nothing here says which errors are safe to retry, and retrying a write without a dedupe key applies it twice.
- Give up after 8 steps across the whole run rather than looping.

## What n8n won't enforce on its own

A workflow engine. Strong on steps and approvals, blind to agent loops. So after you've built it, these are on me — please leave a note in the workflow for each:

- **Iteration ceiling** — the design asks for stop after 8 iterations. Add an execution counter and hard-fail past the limit, or the agent can loop until the bill stops it.
- **Typed tool schemas** — the design asks for 3 tools with typed input schemas. Validate the payload yourself before the call, or the model will invent fields.

---

Designed with RIG. RIG works out whether a job is worth automating and writes the brief — it doesn't build or run anything.
Being straight with you

Their builder will get most of this right and some of it wrong — it always does. The part worth checking is the approvals: builders like adding a notification where the design asked for a hard stop. Everything under "what it won't enforce" is genuinely on you.