Today's business post is about lead tracking, and the five stages most small businesses skip. The journal entry planned for today had a different title. It was going to be called "The follow-up that closed, and the eleven that did not", and it was written into the September calendar three weeks ago as a placeholder, with a note beside it saying the number had to come from the register or the title had to change.
The number came from the register this morning. Then I did not publish it. Here is why, and what I am publishing instead, because the rules are more useful than the count.
What the register is
I keep one file for every proposal and every invoice this business has sent since June. It is a spreadsheet, which the business post will tell you is the wrong tool, and it is right. Eleven columns. The three that do all the work are the follow-up column, the reply column and the status column: when the follow-up went out, what came back, and what the row is now.
That last column is a sentence, not a word. "Closed, not pursuing." "Parked to October." "On ice per the client, he will revert, do not chase." "Won and invoiced." Each one carries the date it was set and the reason. The reason is the part I would not have thought to record in June, and it is the part that makes the file readable in September.
The engine reads that file every morning to decide which follow-ups are due. So the sentence in the status column is not a note to myself. It is an instruction to a machine, and a vague one gets a vague chase sent in my name.
The number I am not publishing
Six days ago, in the three-month post, I wrote a line I had written in the one-month and two-month posts before it: no revenue, no client names, no deal values, no pipeline. A count of proposals sent and proposals signed is a pipeline number. It is a small business's conversion rate, in public, with the date on it.
This morning the engine computed it from the register into a file with the names and the amounts stripped out, and the file sat there ready to go into this post. I read it and applied my own rule from Monday. So the file exists on disk, the count in it is real, and it is not in this paragraph. If I change my mind, it is one edit and the post gets a second version with a date on the change.
What I can say without breaking that line: silence closed more of them than a no did. That is the thing about following up that nobody tells you before you start. The clear no is rare and it is a gift, because it ends the row. The common outcome is a proposal that gets opened, a follow-up that gets read, and then nothing, for weeks, until you have to decide on your own that it is over.
The rules that came out of it
The first three are written into the engine's standing rules, in my words, on the days I said them. The last two are how the register itself is built. Together they are the reason it is worth keeping.
The register records proposals and invoices, not every email that lands. An enquiry is not a deal. Someone who asked a question has not asked for a proposal, and putting them on the register produces follow-up drafts against a person who never asked for one. The register is for things I have actually sent. I set this on the fifteenth after the sweep surfaced three inbound conversations as "new deals not on the register", and none of them were deals.
A closed lead leaves the sweep entirely. When I close a row, I mean stop. Not a softer cadence, not one more nudge in a month, not a hygiene flag every Monday saying it is still blocked. A closed row must stop generating drafts and reminders on the day it is closed, or closing it did nothing. Re-surfacing a lead I have already killed reads as not listening, and the machine that does it is running in my name.
The size of a number does not protect a lead. The largest proposal on the board this year was closed for silence, like several far smaller ones. A big figure is not a reason for a sixth chase. If anything it is a reason for the opposite: the bigger the ask, the clearer the other side's silence is as an answer.
Every close-out carries its reason and its date. "Dead" on its own tells me nothing in November. "Declined, revisit November or December" is a plan. "On hold until further notice, inbound from them reopens it" is a plan. The reason is what lets a parked row come back without a chase, because the reopen condition is written on it.
Every touch is on the record before it is sent. The follow-up column is filled in when the message goes, with the time. That is what lets the next person, or the next morning's machine, know that the second touch happened and the third is due, rather than guessing from an inbox.
What the five rules cost
Nothing. They are five sentences in a file, and they were each written on the day something went slightly wrong: a draft against a dead lead, a nudge to someone I had already decided to leave alone, a row with no reason on it that I could not interpret two weeks later. The business post has the five stages of the process and the research on the two stages that lose the most. This is the founder's version: the process is only as good as the sentence in the status column, and I would rather publish the sentences than the score.
Chat soon.