Technical consulting and fractional CTO
Some companies need fewer wrong decisions rather than more code. For companies that cannot carry a full-time CTO but cannot afford to guess at technology decisions either, I work on a fixed monthly scope.
What is a fractional CTO?
A fractional CTO is an outside executive who owns a company’s technology decisions for a defined number of days per month. They do not write the whole system; they decide what gets built, on what technology, by whom — and what will not be built at all.
What matters in this role is not only technical depth but having sat on the other side of the table. I served as CIO and VP at an international game studio, ran my own IT firm for six years, and mentored at Startupbootcamp. Defending a budget, hiring, and explaining a technical decision in financial terms are not new work for me.
When is this the right choice?
In four situations: there is nobody senior enough to make the technology call; a vendor proposal needs evaluating and nobody can evaluate it; there is a team but no technical direction; or there is pressure to “do something about AI” and no idea where to start. If you need a full-time CTO, this is not the right choice — in that case I help with the hire instead.
| Scope | What it includes | Who it suits |
|---|---|---|
| Advisory | Two fixed calls a month, written opinions on architecture and technology decisions, vendor proposal review | Founders who will make the call themselves but want it checked |
| Active fractional CTO | Weekly cadence, architecture ownership, code review, participation in hiring, roadmap | Companies with a team but no technical direction |
| AI adoption roadmap | Process inventory, suitability screening, pilot selection, cost and data-residency framework, 12-month plan | Companies under pressure to adopt AI with no starting point |
| Technical audit (one-off) | File- and line-level review of the existing system, risk ranking, concrete remediation list | Before an acquisition, a takeover or a vendor change |
What does an AI adoption roadmap contain?
The roadmap starts by screening existing processes for AI suitability and ends with a twelve-month sequence. The output is a decision document, not a deck: which process, in which quarter, on which architecture, under which budget ceiling — and which processes will not be touched at all.
- Process inventory. Listed by volume, repetition, error tolerance and data sensitivity.
- Screening. Processes with low error tolerance and unverifiable output are dropped first. This step is what rescues most programmes.
- Data reality. Where the data actually lives, in what shape, and who can reach it. Most AI programmes stall here, not at the model.
- Architecture frame. Local or hosted, which step on which side, and what the data-residency position is.
- Sequence and budget. A quarterly plan with a ceiling and a success criterion per item.
What I will not do
I do not take work whose outcome cannot be measured. I do not start on a project whose scope is not written down, I do not take over another vendor’s work without reviewing it first, and I do not recommend a technology because it is popular. If I think something should not be built, I say so — that is the part of consulting that is actually worth paying for.
The most expensive decision in a project is not choosing the wrong technology. It is noticing eighteen months later.
Technical consulting: frequently asked questions
How many days a month do you commit?
It depends on scope. Advisory is two fixed calls a month plus written correspondence in between. Active fractional CTO is a weekly cadence with architecture ownership. The number of days is written into the contract, along with what happens if it is exceeded; a scope that quietly expands is bad for both sides.
Will you work with our existing team?
Yes, and that is the arrangement I prefer. Giving a team direction and adding a senior review layer produces more durable results than replacing anyone. Code review and architecture decisions are the concrete part of the work; the cadence is made transparent from the start so engineers read it as support rather than surveillance.
Can you review a vendor proposal?
Yes. This is the most frequently requested one-off engagement. Scope, assumptions, technology choices and pricing structure are reviewed, and I report in writing which line items are missing and which are inflated. I have no relationship with the vendor and will not develop one.
We are a small company. Is this too heavy for us?
It can be bought as a one-off technical audit or as the AI roadmap alone, with no monthly commitment. In a small company the proportional cost of a wrong technology decision is larger, which is why a single call before the decision often pays for itself.
Last updated: