Hey there 👋,
Two years ago I had a service I was genuinely good at.
Fractional CMO. I ran content and acquisition for founders who wanted the senior marketing brain without the salary. You could buy one, two or three days a week of me. At its peak it cleared $10K a month. No contract, three month minimum, a nice page with my exits listed down the side. I mentioned it in newsletters at the time.
It worked, but I disliked it.
Not because the clients were bad. Because of what I was actually selling, once you strip the page back. I was renting out my calendar. The whole thing priced my time. Buy more days, get more me. That is the model. That is the entire model.
And here is the thing I understood too late to enjoy the money: the work I was renting my hours for was already becoming work that didn't need hours. Content built. Funnels built. Outreach written and sent. Reporting. The exact stuff I charged a day rate for is the exact stuff an AI worker now does without me in the room. I was selling time to do work that was about to stop being measured in time. The moment time stops being the scarce thing, a model that prices time is dead. I just hadn't said it out loud yet.
So I then went down the AI agency route and launched WhyNot.Bot.
No retainer. No day rate. No meter running. One-off projects, fixed fee, paid once. I started by setting up custom GPTs for charities, then moved into building Claude Cowork workers for businesses and charities. I'd come in, build the thing, train the team, and leave. You paid for the result, not for my hours. After the fractional grind it felt like the answer, and for a while I didn’t mind it.
Then I hit the two ceilings.
The first: it still ran on my time. Every project was me, in the room, for days. Pricing the result instead of the day was an improvement, but the cap was the same shape. I only earn while I am working. It doesn't scale past the hours in my week, and I am not in the business of selling my week.
The second: it didn't capture the value it created. I charged the same flat fee whether the workers I built saved a charity an afternoon or saved a business a full salary. The price tracked my effort, not what the install was worth to the person buying it. I was building things that removed five-figure costs and charging four figures to do it. The value was real and I was leaving almost all of it on the table.
So I am winding the AI agency down because I can now see exactly where it caps out, and I would rather build the things that don't.
There are two of them, because the two ceilings need two different answers.
The first answer is the Redundant Packs. The work that could be packaged, I packaged. The same skills I used to install by hand, bundled so anyone can install them without me in the room. They sell while I sleep. No time, no ceiling, because I have removed myself from the transaction completely. That is the scaling problem solved by deletion.
And they work great. This weekend I’ve been using the Reddit Growth worker that’s inside the Website Operator pack to find threads to reply to to promote my poker web app.

The Install
The second answer is The Install. The work that can't be packaged, the bespoke build into a business with a real payroll to remove, I priced properly this time. Against what it takes away, not against my hours.
I go into a business, find the repetitive work that is quietly eating a salary, build the AI workers that do it, train the team to run and extend them, and I leave. One engagement. They own the workers when I walk out the door. It is the same buyer I had as a fractional CMO, an owner who needs the work done and doesn't want to build a department to get it, but the structure is inverted.
They are not renting me by the day and watching the clock. They buy a fixed outcome once and keep it forever, and the price is set by the cost it removes from the payroll, not by how long I spend. A worker that does a year of salaried work pays for itself inside the year and keeps paying after. I am no longer the cost on the invoice. I am the thing that deletes one.
Three models in two years. The first priced my time and capped me. The second priced the result but still ran on my hours and gave away the value. The third gives the work away to anyone (the packs) or charges what it is actually worth (the Install), and neither one needs me to keep showing up.
If you are selling fractional anything right now, or paying for it, look hard at what the money actually buys. If it is a senior person's hours on work that is becoming automatable, you are on the wrong side of a model that is closing. The operator who survives this is not the one with the sharpest day rate. It is the one who stops selling time.
The Install went live today at redundant.lol/install. It is for owners carrying a payroll on repeatable work, the ones about to make the next three hires they would rather not make. Five figures, a handful a year, no retainer, you own what I leave behind.
The $10K months taught me the lesson the expensive way. A service that prices your time has a ceiling, and AI just spent two years quietly lowering it.
Make yourself redundant from the doing before the market does it for you and doesn't ask first.
That is the whole reason this newsletter is called what it's called.
One more thing, and it's a strange one.
I've been quietly building watches, as first mentioned here back in February. Components, finishing techniques, the lot, in the background for a while now. There's a whole community around this, it's called modding, and I've been deep in it for a while now.
Here's the idea I keep circling back to. A mechanical watch is the most redundant object a person can own. Your phone tells the time. A mechanical movement does nothing your pocket can't, with hundreds of tiny parts, wound by your wrist, to achieve what a £5 digital does better. And that is exactly why I want to make one. A beautifully engineered, completely unnecessary thing. The joke writes itself, and the joke happens to be true.
So I'm thinking about a microbrand under the Redundant name, built in the brand's own colours. A blacked-out case, a dark dial stripped right back, no indices, and a single hit of that acid green as the only accent. A proper mechanical movement underneath. Murdered out, one sharp pop of colour, the manifesto on your wrist. Built for the exact people in this community, the investors and operators already making themselves redundant before AI does.
It's early. No commitments, nothing for sale, just a thing I can't stop thinking about. I shared a very early render by Grok in the Skool community). The next step is gloriously unglamorous: buy a box of parts from China, order a custom dial and build the prototype on my desk.
I've put a poll at the bottom. Tell me honestly.
Should Redundant make a watch?
Cheers
Richard
Stop letting busywork get in the way of selling
Researching accounts. Building lists. Writing sequences.
There's a better use of your team's time.
Apollo is the AI revenue engine that handles the busywork, so you can stay focused on selling.
Plus, everything you need is in one place:
230M+ verified contacts
AI-powered outreach
Data enrichment
Inbound lead capture
Meeting scheduler
And more
Stop doing busywork and start building pipeline, faster.
With Apollo — the AI revenue engine powering 4M+ users.


