“Fractional” technical leadership. A CTO for when you don't need a full-time one yet.
Expensive technical decisions —architecture, vendors, who gets hired— get made anyway, and almost no one reviews them. This gives you a counterpart who decides, takes responsibility for what they approve, and leaves the why behind every decision written down on your company's side.
Illustrative dialogue · 1:18 · plays muted
It isn't a project with a start and an end: it's an ongoing engagement on a fixed cadence — a committee every two weeks, continuous review of what gets built, a weekly 1:1 with your tech lead, the point of contact with vendors, and a decision board that's yours from day one.
What a Fractional CTO is
Judgment, decisions and technical governance, at the level of commitment your company needs today. It's not extra hands: if what you're missing is development capacity, this won't solve it, and you'll hear that in the first meeting.
Decide
Review
Document
telling me
the truth?
The problem
The issue isn't the missing CTO: companies run for years without one. It's that expensive decisions get made anyway, and no one reviews them.
“They told me this takes five minutes”
Said by someone who never put it in production. No one measured how long something comparable actually took, and the estimate is an opinion.
“We were offered agents that work on their own all night”
Whoever sells it was selling a different revolution yesterday. Without a technical counterpart, you buy the pitch, not the result.
“Our senior decides everything, and it works”
It works until they quit. The company's judgment is one person's judgment, and it leaves with them.
Decisions run on gut feeling, and it gets expensive
With nothing to compare against, the cost of being wrong only shows up once it's already been paid.
The third one is the most expensive, and the one no one treats as urgent until that person quits.
Who it's for
You don't need a big team, or a team at all. You need to be about to make technical decisions that are costly to undo, with no one to check them against.
A startup getting started, or about to
There's no tech team yet, or it's just getting started. The decisions of the first months are the most expensive to undo later.
A non-technical manager leading a technical team
You're being pitched a migration, a vendor, “AI in everything”. You need a second view to separate what's useful from the fear of being left behind.
Head of engineering or technical manager
You have the team, but no peer at your level to debate with, without politics getting in the way.
A CTO who wants a second opinion
Not to replace you: so big decisions go through someone who doesn't answer to your org chart.
Mid-level to senior, deciding without a safety net
You ended up making architecture calls because you know the most, not because anyone prepared you for it.
A company in transition
The CTO left, or is about to. Decisions have to be held steady while you figure out what's next.
If what you're looking for is for you and not for your company —a promotion, resolving a conflict, sorting out your career— that's mentorship, and it's bought differently.
What it's not
This table matters more than the one above: it's the one that prevents the misunderstanding that breaks the deal in month three.
| It isn't | It is | |
|---|---|---|
| Scope | Hands. If you lack development capacity, this won't solve it | Judgment, decisions and technical governance |
| Accountability | A name that signs off and never shows up | A counterpart who takes responsibility for what they approve |
| Commitment | Round-the-clock availability or on-call duty | An agreed, stated range of commitment |
| Team | A replacement for your in-house tech lead | Direction for the team you already have, and support for whoever leads it |
A Fractional CTO without clear boundaries ends up coding by month three, which is exactly what you can buy cheaper elsewhere.
The cadence
Five pieces, each on its own schedule. What it produces isn't activity: it's decisions made, with the why written down.
Tech committee
Every two weeks, fixedDecisions made, with the why written down.
Review of what gets built
ContinuousDesigns and changes reviewed: the author is not the approver.
1:1 with your in-house lead
WeeklySo judgment stays in-house and doesn't depend on the vendor.
Vendors and audits
As neededThe technical counterpart you don't have today.
Decision board
Live, and yoursWhat was decided, when, and why.
What it decides and what it only recommends
It's agreed at the kickoff meeting, with your in-house tech lead in the room. Never behind their back.
The technical calls, and owns them
- Architecture and technical standards
- What gets approved for production
- Who approves what: change governance
Anything that isn't a technical decision
- Hiring and letting people go
- Budget
- Product priorities
How it starts
A two-to-three-day diagnostic. Something small and verifiable before committing to an ongoing engagement.
Decision inventory
The ones made without review, and the pending ones no one made.
Technical risk map
What breaks first, and what depends on a single person.
The three decisions for the next 60 days
With what each one costs.
If the work continues, the diagnostic becomes the basis of the ongoing engagement. If it doesn't, it's yours anyway.
The case · a 4-year-old fintech
Faced with a “this takes five minutes,” the previous comparable integration was measured exactly as it had happened:
The conversation changed tone. What's being sold isn't speed: it's knowing whether what you're being told is true, measured against your own company's history. When someone tells you it's easy with AI, this is what to ask them for: the number.
Anonymized reference case: no client name, no absolute amounts, no third-party data. Volume figures are kept exactly as measured.
The exit, in writing from day one
A service sold on how it ends is easier to buy. The real fear is getting locked in, and that gets answered before starting.
The board is already yours
The decisions and their why live on your company's side from day one, not at the end.
Notice on both sides
Stated in the agreement, with a handover clause that says what gets delivered and when.
The account gets switched off and everything keeps running
Named, revocable access. When it ends, nothing breaks because the provider left.
When something comes up that no committee can answer
Three signals. Until they show up, nothing else is offered: the why behind a decision isn't sold, it's needed.
“I don't remember why it ended up like this”
Said for the second time in the same committee.
Nothing to compare against
The estimate can't be made because the project's history isn't anywhere.
The why lives in a single head
And that person is on their way out. Every week before they leave is worth double.
That's what Company Brain is for
Your company's memory, searchable, with the source cited in every answer.
Companies and projects I've been part of
Over 25 years in fintech, banking, retail, media, streaming and government: the companies I worked at, and the projects and clients that went through my teams.
Names belong to their owners; the circles are initials, not logos. In blue, the companies I worked at; around them, projects and clients.
Write it down.
Stand by it.
“The company's judgment is one person's judgment. And it leaves with them.”
Let's talk. First step.
Two or three days, and an inventory of your decisions on the table. A short conversation is enough to know whether you need it.