If your cofounder CTO is falling behind the company, do not start by asking whether they should go. Start by working out which part of the job is failing. Most of the time the answer is narrower than it feels: one person is being asked to do three jobs, and they are good at one or two of them. That usually leads to coaching or to splitting the role. Replacement is the right call less often than an angry month suggests, and it is also the most expensive one.
This is a hard conversation partly because it is two conversations at once. One is about a role in a company. The other is about a person who took the same early risk you did and who probably owns as much of the business as you do.
The job you hired for no longer exists
The first version of the CTO job is mostly building. The person writes much of the code, picks the stack and makes each technical call at their desk. Around the time a company has a handful of engineers and starts selling to more demanding customers, the same title starts to mean something else. The work becomes hiring, setting direction, managing people, translating risk for a board and saying no to the CEO with evidence.
Those are different skills, and being excellent at the first does not predict the second. Nobody failed when this happens. The company simply changed under the person. The AWS startups team has written about the same title problem, including how a change away from a "Chief" title can feel like a demotion even when it reflects growth.
Diagnose before you decide
Vague frustration leads to bad decisions here, so split it into the separate jobs and score each one with evidence from the last two quarters:
- Technical judgment. Have the big calls held up? Look at the architecture decisions, the vendor choices and the estimates.
- Delivery. Does the team ship what it says it will, and can someone explain slips before they happen?
- People leadership. Are good engineers staying? Are they being given feedback? Can your CTO close a senior candidate?
- Company communication. Can they explain technical risk to you, the board or an investor without you needing to translate it?
Then separate three causes, because each one has a different answer:
- A skills gap they want to close. They know it is a problem, and they are trying. Coach.
- A role they do not want. They are strong but miserable in management. Split the job.
- A judgment or trust problem. The big calls are wrong, the team is leaving and the news is hidden from you. That is the case for replacing them.
The last category matters most. Being weak at management can be fixed or routed around. Hiding bad news from your cofounder usually cannot.
Option one: coach them
Coaching makes sense when the judgment is sound and the gap is in newer skills like managing managers. Give it a real shape: name two or three behaviours that need to change, agree what would count as progress, get them an experienced CTO to talk to and set a date to review it, such as 90 days.
Without a date this turns into a slow drift where everyone is unhappy and nobody says so. A 90-day review is a working figure rather than a law, but it is long enough to see a change in behaviour and short enough that the company can still act on what it learns.
Option two: split the job
This is often the best answer, and founders tend to reach for it too late. The technical cofounder keeps what they are best at, usually architecture and the hardest product problems, and someone else takes the people and delivery work. The usual split is a VP of engineering alongside the CTO, or at a smaller team a strong engineering lead.
Two rules keep it from failing:
- Say who decides what. Write down who owns architecture, who owns hiring and performance, who owns the roadmap commitments and who breaks a tie. If two people think they own the same decision, your engineers will quickly learn to shop for the answer they want.
- Let the founder help choose. A VP imposed over someone's head rarely lasts. If your cofounder interviews the candidates and picks one, the arrangement has a chance.
A founder who is already leaning this way may recognize the path in stepping back from the CTO role as a technical founder. If you are not sure that the team is big enough to justify a full-time VP, a fractional leader can run the people and delivery side for a while. How fractional compares with the alternatives covers where that works and where it does not.
Option three: a real change at the top
Sometimes the honest answer is that your cofounder should leave the role, and perhaps the company. Signs that you are there include several senior engineers leaving and naming the CTO as the reason, repeated large calls that turned out wrong with no learning, and a pattern of surprises you only hear about after the fact.
Before you have that conversation, check four things with your lawyer. This is general guidance and not legal advice, because the details turn on your own documents:
- Vesting and any leaver terms. Read what the founder agreement says happens to unvested and vested shares.
- Board and voting rights. Check whether you can actually make this change, and who else must agree. Cofounders often hold board seats and significant votes.
- Code and IP ownership. Confirm that the company, and not the person, holds everything they wrote, particularly anything from before incorporation. Who owns your code has the checklist.
- Access and knowledge. Plan who holds the keys to production, domains and vendor accounts, and who understands the systems only this person knows.
If a round is in progress or close, read what investors check after a technical cofounder leaves first. Investors will ask about it, and a clean exit is far easier to explain than a messy one.
What to do when you cannot judge it yourself
Non-technical CEOs have a particular problem: they can see the symptoms but cannot tell whether the technical calls were right. A missed deadline might be bad leadership, an unreasonable plan or a truly hard problem. Those look identical from the CEO's chair.
An independent review helps because it replaces opinion with evidence, and because it is fairer to your cofounder. A short, scoped look at the architecture, delivery record and team, like a technical teardown, will often show that the problem is more specific than either of you assumed. Tell your cofounder before it starts. A review done behind someone's back can end the relationship regardless of what it finds.
If you want to talk through the situation before raising it with your cofounder or board, book a call for an outside read.
FAQ
At what stage do founding CTOs usually hit this wall?
There is no fixed stage. The pressure tends to show up when the team passes the point where one person can review all the work, or when larger customers and investors begin asking harder questions. Many companies feel it around their first serious fundraise or a small handful of engineers.
Should the board be involved?
Eventually, yes, and investors on your board will want no surprises. But talk to your cofounder first unless there is a legal or safety reason not to. Going to the board first turns a role discussion into a fight.
Can a founding CTO stay on after moving out of the top technical role?
Yes, and it can work well as a principal engineer or similar senior role, provided the reporting line is clear and the person actually wants the job. It fails when the title changes while the old authority quietly continues.
Is this the same as outgrowing a fractional CTO?
No. Outgrowing fractional is about how many hours of leadership you need. This is about whether the person in a full-time founding seat fits the job as it now stands.