A spreadsheet. That's it, and that's deliberate.

Blueprint §3.2's verdict on this one is light and optional, and the reason is worth stating up front: making the tracker itself "smart" would make it less trustworthy, not more. You need to be able to look at this and know it's correct. A row that something else edited is a row you have to check.

1The columns

2The pipeline

contacted → replied → scoping → quoted → negotiating → booked → delivered → paid
                ↓         ↓        ↓          ↓
                       no / silence  →  re-pitch [DATE]

"No" and "silence" are not the same state and shouldn't share a row colour. A no with a date attached — "try us in the new year" — is a warm lead in the future. Silence is unknown. Section 3's follow-up script converts silence into one or the other, which is the point of sending it.

"Paid" is a separate state from "delivered." Obvious, and the reason most freelancers discover they're owed money three months late is that their tracker stopped at delivered.

3The two views that make it useful

1. What's due this week. Sort by next action date. That's your Monday morning, and it's the whole reason the column exists.

2. What's owed. Filter to delivered-but-not-paid. Check it fortnightly.

4The habit

• Log every pitch as you send it. Not at the end of the week — you won't.
• Every message either way updates last touch and next action date.
• A no gets a date. Whatever they said, turn it into a month, and put it in next action date. That row is now a re-pitch waiting to happen.
• Review it once a week. Ten minutes.

5What it tells you after a few rounds

This is the part that pays for the discipline. With thirty rows you can answer questions you cannot answer from memory:

• Which source actually converts — cold, warm, or inbound?
• How many contacts does a booking take?
• Which category replies and which doesn't?
• How long between first contact and money?

Section 3 says most pitches get no reply and that this isn't a verdict on you. The tracker is how you find out whether it's the pitch or the volume, and those have completely different fixes.

6Starting it

Fifteen minutes, one spreadsheet, eleven columns, one row per brand from your Module 0B list. Don't buy software for this.

⎯⎯ AI note — small, optional ⎯⎯

The only sensible touch, per blueprint §3.2.

Drafting the follow-up when one comes due. You open the tracker, see three rows with today's date in next action date, and need three short messages. That is a real use: paste the row's notes and the script from 03-outreach-scripts.md, get a draft, read it, send it.

What not to do, and this is the actual content of this note:

• Don't let anything write to the tracker. You need to be able to trust what's in it at a glance. A row you didn't type is a row you have to verify, which costs more than typing it.
• Don't ask it to score or prioritise your leads. It has no information you don't have, and a confident-looking ranking of your own pipeline is worse than no ranking — you'll defer to it.
• Don't automate the sending. Section 3: one follow-up is professional, a sequence is a campaign, and people can tell.

The tracker is a record. Its whole value is being accurate and boring.