By Andreas Pettersson, Founder, Leaders ADAPT
Two people tend to search for a CTO mentor. The first is a founder who cannot read code, who has just hired a technology leader and has no way to tell whether that person is good. The second is the person who was just handed the technology seat, the best engineer in the building, now called CTO or VP of engineering, and quietly unsure what the job is. Both are right to look for help. They usually look for the wrong kind.
I held the technology seat without a mentor. At Arcules, the cloud video company I built from zero to more than 150 people before Canon bought it, I was CEO, CFO and CTO at once for a stretch. I learned the CTO part the expensive way. This page is the version I wish someone had given me.
Quick answer: A CTO mentor is someone who has held the technology seat and meets a founder or new technology leader regularly to pass on how the job is done. The mentor shares judgment and mistakes, the mentee keeps every decision. Look for someone who has done the job at your size, talks about people and money as easily as code, and agrees an end date at the start.
Leaders ADAPT runs CTO mentoring and advisory through its own bench, priced as a fixed monthly fee or a fixed price working session and quoted per engagement after the first call. The research cited on this page belongs to its publishers, is linked with the date I read it, and their page wins if a figure has changed. Nobody named here pays us.
Mentor, advisor, coach: three different jobs
A CTO mentor develops the person. A CTO advisor improves a decision. A CTO coach helps the person think. The words get swapped all the time, and the swap is why many of these relationships drift.
Here is the sort.
| CTO mentor | CTO advisor | CTO coach | |
|---|---|---|---|
| The question it answers | How do I become good at this job | What should we do about this decision | What do I actually think |
| Has held the seat | Must have | Must have | Helpful, not required |
| What the hour is about | The mentee's habits, judgment and blind spots | A vendor, an architecture, a hire, an AI bet | The mentee's own reasoning |
| What the mentee leaves with | A pattern they can reuse | A view on one decision | Clarity they found themselves |
| How it ends | When the mentee no longer needs it | When the decision is made | When the goal is reached |
The CTO advisor page covers the decision side in depth, the vendor proposals, build or buy, security basics. This page is about the person. If you want the wider map of who sits where, the advisory board vs advisor vs coach page draws the lines across all the seats, not only technology.
In practice a good mentor does all three in one hour. The discipline is to say which one is happening. "I am going to tell you what I would do" is advice. "Here is a mistake I made in your exact spot" is mentoring. "What would you do if I were not here" is coaching. A mentee who cannot tell the three apart will start outsourcing decisions to the mentor, and then you have an expensive shadow CTO with no accountability.
Two people who need a CTO mentor, for different reasons
The non technical founder. You run a company between $5M and $50M. The product or the operations now depend on software, and the person leading it reports to you. You cannot judge their work. You cannot tell a padded estimate from an honest one. Every technology conversation ends with you nodding.
You do not need to learn to code. You need to learn the twelve questions a CTO would ask your CTO, and how to hear the answers. A mentor teaches you to run a technology one on one without fear, to read a roadmap for what is missing, and to know when a delay is a people problem dressed as a technical one. Most of the founders I talk to tell me two things before anything else. They hate one on ones, and delegation boomerangs back. With technology both get worse, because the founder feels unqualified to push back, so the decision drifts back to the founder by default or stalls.
The new technology leader. You were the best engineer, or the one who built the first system, and now you lead. The title changed on a Monday. The job did not come with a manual.
McKinsey's work on leadership transitions (May 2018, read October 2026) found that 27 to 46 percent of transitions are regarded as failures or disappointments two years in, and that 68 percent of those that fail founder on politics, culture and people. Not on the technical work. That matches everything I have seen in technology seats. Nobody gets fired as a first time CTO for picking the wrong database. They get fired for losing the room with the CEO, the sales leader and their own team.
The cost lands on the team too. A 2023 Harris Research and Oji survey of 2,066 US employees, reported by SHRM (October 2023, read October 2026), found that 34 percent of workers who had a bad experience with a poor manager wanted to leave the organization. Your engineers are expensive to replace. A first time leader with nobody to learn from is a retention problem waiting for a date.
What to look for in a CTO mentor
Five things, in the order I would test them.
1. They have held the seat at your size. A CTO from a 5,000 person company knows how to run a platform group and a budget committee. That is not your problem. Your problem is one leader, eight to forty engineers, a vendor or two, and a CEO who wants the date. Ask how many people reported to them when they held the job you are about to do.
2. They can tell a mistake in detail. Ask for the worst technology decision they made and what it cost. A mentor who answers with a story about someone else, or a lesson with no numbers in it, has either not made enough mistakes or has not looked at them. The mistake story is the product.
3. They talk about people and money as easily as code. Google's Project Oxygen, documented on re:Work (read October 2026), analysed over 10,000 data points from reviews and surveys and named ten behaviors of effective managers. Key technical skills is one of the ten. Being a good coach, empowering without micromanaging, communicating and making strong decisions are among the others. A CTO mentor who only wants to discuss architecture will help with one tenth of the job.
4. They will say the uncomfortable thing to the CEO. At some point the mentee's problem is the founder. Unclear priorities, a sales team promising features, a budget that changes monthly. A mentor who will only speak to the mentee, and never in the room with the founder, cannot fix half of what goes wrong.
5. They agree an end date before the first session. Mentoring without an end is a subscription. Six to twelve months, with a check at month three, is right for most first time CTOs and VPs of engineering.
Red flags, briefly. A mentor who wants to see your code before your calendar. A mentor who brings a tool or a platform they happen to resell. A mentor who tells you what to decide more often than they tell you how they would think about it.
How a CTO mentor relationship runs week to week
Here is the rhythm I use with technology leaders, and the one our bench runs.
Before the first session. The mentee writes one page. Three things they are good at, three things they are afraid of in the job, and the three decisions they expect in the next quarter. If the mentee is a founder, the three decisions are about the technology leader they manage. The founder of the company reads it too, unless the mentee has a good reason not to share it.
Weeks 1 to 4. The calendar. Not the code. We look at how the mentee spends a week. Most first time CTOs I have met still spend most of the week doing the work they were promoted for, and that is the first thing to change. By week four the calendar has a weekly meeting with the CEO that has an agenda, one on ones with every direct report, and a block for thinking that nobody may book.
Weeks 5 to 12. The hard conversations. The underperforming engineer who was a friend. The estimate the sales leader does not like. The budget conversation with finance. We rehearse one each week. The mentee runs it before the next session and reports back. This is where most of the value sits, and it is the part a mentor without executive experience cannot give.
Months 4 to 6. The room above. Presenting to the leadership team and the board. Writing a one page technology plan for the year. Saying no to a project the CEO likes. The sessions move to every two weeks, and the mentee brings the agenda.
Months 7 to 12. Taper. Monthly, then done. The mentee should now be asking the mentor's questions before the mentor does.
Every session ends with one written line: the thing the mentee will try before next time. Not three things. One. The line goes at the top of the next session's notes, and we start there.
When a single decision needs more than an hour, a platform choice or a reorganisation of the team, we pull it out of the mentoring rhythm into a fixed price two hour advisory session with a written view at the end, so the mentoring hour stays about the person.
Where AI fits in a CTO mentor relationship
AI has changed what a new technology leader has to judge, and it is now on the agenda in most sessions.
The pressure runs in two directions. The Stack Overflow 2025 Developer Survey of over 49,000 developers (July 2025, read October 2026) found that 84 percent use or plan to use AI tools, up from 76 percent, while 46 percent do not trust the accuracy of the output, up from 31 percent. Your engineers are using the tools and doubting them at the same time. The DORA 2024 report (read October 2026) found that AI improves individual productivity, flow and job satisfaction but negatively affects software delivery stability and throughput, so the basics of small changes and good testing matter more, not less.
So the mentor helps with three things. First, the policy, written on one page: which code and which customer data may go into which tool. Second, the measure: whether the team is shipping more working software, not whether it feels faster. Third, the founder conversation, because the CEO has read that AI will halve the engineering budget and the new CTO has to answer with evidence.
Where AI does not fit is the judgment about people. Who to promote, who to let go, whether a senior engineer is ready to lead. Our page on what not to delegate to AI has the longer list. If your company runs on EOS, the AI for EOS companies guide shows where the tools attach to the scorecard and the weekly meeting, which is a useful frame for a new technology leader joining that rhythm.
What a CTO mentor does not fix
A CTO mentor does not fix an empty seat. If nobody inside the company can lead the technology work between sessions, an hour a week of mentoring has nowhere to land. That is a fractional CTO for one to three days a week, and a good mentor says so in the first month.
A CTO mentor does not fix the wrong person in the seat. Sometimes the best engineer should not lead. A mentor can make that visible faster than anyone, and it is kinder to say it in month two than to discover it in month eighteen. The decision is the founder's.
A CTO mentor does not fix a leadership team that fights about priorities. If sales, operations and finance each want a different roadmap, the CTO is the one caught in the middle, and no amount of mentoring the CTO resolves it. That is leadership team development work, and I do it alongside the mentoring, or the mentee carries a fight that belongs to the whole team.
A CTO mentor does not fix a founder who will not let go. If every technology decision still comes back to the founder for approval, the mentee is learning to be an assistant, not a CTO. Mentoring the founder at the same time, even monthly, is often the cheaper fix.
When a CTO mentor relationship should end
It should end on purpose. Four signals tell you the work is done.
- The mentee brings decisions already made, and wants a check, not an answer.
- The founder has stopped asking the mentor what the CTO is doing.
- The mentee has had each of the hard conversations at least once, alone.
- The sessions have been about the same two topics for a month.
If three are true, plan the last session. End with a written note from the mentor: what changed, what to watch, and the one habit to keep. Then stop. A mentee who wants to keep someone outside the company to think with can move to a lighter advisory rhythm. That is a new agreement, not a mentorship that never ended.
The wrong reason to end is that it got hard. Months three to five are usually where the mentee meets the first real conflict with the founder or the team, and where the value is highest. If a mentee wants to quit then, that is the conversation to have, not the reason to stop.
Who is on the bench for this
Leaders ADAPT is the CEO's bench. I build the leader, and when a founder or a new technology leader needs a CTO mind for an hour a week, I bring one from the bench and stay in the room.
Sergey Dmitriev is the fractional CTO and AI systems partner on the bench. Sergey builds and runs the AI and data work Leaders ADAPT delivers, including digital twin and company knowledge systems, and trains leadership teams on AI tools with me. In a mentoring engagement, Sergey is the technologist the new CTO talks to about the team, the systems and the AI questions, and the person who helps a non technical founder learn which questions to ask.
Thor Legvold is the leadership psychologist on the bench. When the question is whether the person in the seat can grow into it, or why the founder and the CTO keep missing each other, Thor runs the assessment and the fit conversation.
I take the founder side myself. I have sat in the CEO, CFO and CTO chairs, sometimes in the same week, and most CTO mentoring problems turn out to have a CEO half. You can see the shapes we work in on the Fractional Executives and Advisory page, and the fractional executive services hub explains how the bench fits together. I take the first call, match the partner and stay accountable for the outcome.
CTO mentor FAQ
What is a CTO mentor?
A CTO mentor is someone who has held the technology seat and now meets a founder or a new technology leader regularly to pass on how the job is done. The mentor shares judgment, patterns and mistakes from their own career. They do not take decisions, manage the team or own the roadmap. The mentee keeps every decision and gets better at making them.
What is the difference between a CTO mentor, a CTO advisor and a CTO coach?
A mentor develops the person by sharing experience from the seat. An advisor gives a view on a specific decision, such as a vendor or an architecture. A coach asks questions so the person finds their own answer, and does not need to have held the seat. Good relationships mix all three, but the mentor should say which one is happening in a given hour.
Does a non technical founder need a CTO mentor?
Often yes, but for a different reason than a new CTO does. A non technical founder needs to learn enough to hire, judge and hold a technology leader to account, and to tell a good estimate from a bad one. The mentor teaches the founder the questions a CTO would ask. The founder does not need to learn to code.
How often should you meet a CTO mentor?
Weekly for the first three months, then every two weeks once the mentee runs the basics without help. Sixty minutes is enough if the mentee brings an agenda. A short written note after each session, with the one thing to try before the next meeting, matters more than the length of the meeting itself.
What should I look for in a CTO mentor?
Someone who has held the seat at a company your size, not only at a large one. Someone who can explain a mistake they made, in detail. Someone who talks about people and money as easily as architecture. Someone who will tell your CEO something uncomfortable. And someone who agrees an end date with you at the start.
When should a CTO mentor relationship end?
When the mentee makes the decisions the mentor used to help with, and makes them well, without asking first. A planned end at six to twelve months works for most first time CTOs. If the company needs more hands on technology leadership than an hour a week, the right move is a fractional CTO, not a longer mentorship.
Can a CTO mentor help a VP of engineering become a CTO?
Yes, it is one of the most common reasons to hire one. The step from VP of engineering to CTO is mostly about the work outside engineering, talking to the board, owning the technology budget, and making trade offs with sales and finance. A mentor who has made that step can show where the time goes and what to stop doing.
Write the end date on day one
A CTO mentor is the cheapest way to shorten the year a founder or a new technology leader spends learning the job alone. It works when someone capable is in the seat, the mentor has held a seat like it, and both sides know when it ends.
Tomorrow morning, write one page. If you are the founder, three things you cannot judge about your technology leader. If you are the new leader, three decisions you expect this quarter and the one conversation you are avoiding. That page is the first session's agenda.
What the first conversation looks like
It is a 30 minute call with me. The founder and the technology leader are both welcome, together or separately. You bring the one page. I tell you whether this is mentoring, a single advisory decision, a fractional CTO seat or a leadership team problem that has a technology face. If the answer is none of those, I say so.
If it is mentoring, Sergey joins the second conversation, we agree the first three months, the one line rhythm and the end date, and the engagement is quoted as a fixed monthly fee after that call. I stay in the room with the founder for the length of it.
Your step for tomorrow morning: take the free assessments so I can see how the leadership team works before we meet, then book a fit conversation and bring the page.
