The Fractional Advantage

Here's what nobody puts on the slide. You're not hiring a CTO. You're betting on which of three people will make the handful of decisions that decide whether your architecture survives the next growth phase, and you're making that bet before you know what those decisions are. The cleanest resume has never operated at your stage. The cheapest candidate will learn on your cap table. The third has two other offers. And the stakes keep rising, because AI is quietly turning every business into a technology business, so a technical leader's early calls now reach into how you build, how your team works, and where the business goes.

The Education You're Paying For

Michael Watkins, in The First 90 Days, puts a number on onboarding a senior executive: roughly 6.2 months before they generate more value than they consume. That's the baseline for a leader stepping into a role they've done before. For a first-time CTO stepping into a first-time company, the curve is longer and the mistakes cost more, because early-stage architecture decisions have a long blast radius.

Now price it. Levels.fyi and the Pragmatic Engineer's annual compensation research both document that early-stage CTO packages, base plus equity at plan, routinely land in the mid-six figures all-in. It's the most expensive hire you'll make for a year, and the hardest to unwind. Most of that first year goes to recruiting, onboarding, learning your stack, and running the standups nobody else will run. Some fraction goes to the decisions that actually shape the company. If that fraction is a quarter, you're paying full executive comp for the quarter you needed and carrying the rest. You don't hire a general contractor to live in the house while you're still surveying the lot.

None of this is a failure of the individual: it's what happens when a company needs decisions informed by experience it can't yet afford and hires availability instead.

What the First 18 Months Actually Demand

Strip the title off and look at the work. In the first 18 months, the technical leader makes a small number of decisions that compound for years: the foundational architecture, the first five engineering hires, the vendor contracts that will either constrain or liberate the product, and the cultural pattern that sets how the team handles its first real incident. These aren't forty-hour-a-week problems. They're high-stakes, low-volume problems that reward pattern recognition and punish improvisation.

When Founding Partner B built ToolWatch, which grew through acquisition into AlignOps, the architecture calls made in year one determined whether the product could survive the integration years later. Founding Partner A's early decisions about marketplace infrastructure are what let a labor platform now operate across all 50 states. At Oxen.ai, Founding Partner C is making choices right now whose consequences will be visible in 2028. None of this required forty hours a week. All of it required having made the same class of call, badly, somewhere else first.

Camille Fournier makes the point cleanly in The Manager's Path: the further up the leadership stack you go, the more the job compresses into a smaller number of decisions with larger consequences. Hours stop being the unit. Judgment becomes the unit.

The Reflex Nobody Examines

There's a reflex in early-stage companies, call it the full-time reflex, where the board, the founder, and the candidate all quietly agree that a serious company must have a serious person in the chair forty hours a week. It shows up as pattern-matching from investors, as the founder's own insecurity about whether the company is "real," and as compensation requirements calibrated to a full-time market rate.

The reflex is doing work. It's resolving an anxiety about legitimacy and signaling commitment to the next round. What it isn't doing is matching the shape of the hire to the shape of the work. It's the CTO version of hiring a VP of Sales before you have a repeatable sales motion: the right seat, filled two stages too early.

And the reflex is expensive precisely because the hire is so hard to reverse. Repeated HBR analyses put senior outside-hire failure rates near 40% within 18 months, and the odds don't improve when the role is ambiguous and the company can't yet describe what success looks like. Hire too early and you're often buying a seat you'll re-hire anyway.

Reframing the Hire

Here's the question underneath the CTO decision. You're not choosing between "fractional" and "full-time." You're choosing between buying hours and buying judgment. A full-time hire is right when the work is continuous and the judgment load spreads across many small decisions a day. A fractional hire is right when the work is episodic and the judgment concentrates in a few high-stakes calls. Most early-stage companies are in the second category and talk themselves into the first, because the first feels safer to explain to a board.

An experienced fractional leader working across several early-stage companies is running a pattern-recognition loop. When the migration you're stuck on is the third variant of one they've watched play out this year, diagnosis gets faster and the expensive experiments don't need a fourth run. So forget best fractional versus best full-time. The comparison that matters is best fractional versus the full-time you can actually hire at the comp an early-stage company can sustain.

A Test You Can Run

Want something sharper than "the board asked"? Score these. A couple of clear yes answers and full-time starts to earn its cost. Mostly no, and a fractional partner plus a strong engineering manager is the better bet.

1. Scale
   Do you have fifteen-plus engineers, or will you within six months?
2. Decision bottleneck
   Are technical decisions blocked at the top more than once a week?
3. Org complexity
   Do you have multiple engineering teams with distinct objectives and stacks?
4. Board agenda
   Is technical strategy a recurring item at your board meetings?
5. Runway
   Is your runway eighteen months or longer at the post-hire burn?

Will Larson's writing at lethain.com makes the same point in fewer words: match leadership to the work, not to the org chart you hope to have.

When Full-Time Is The Right Call

We're not arguing fractional always wins. Once the engineering team crosses roughly ten to fifteen people, once daily architectural decisions need continuous oversight, or once the product itself is the technical innovation rather than a wrapper around one, the company has outgrown episodic leadership. At that point a fractional partner's highest-leverage move is to name the transition, help recruit the full-time hire, transfer the context and relationships, and step out cleanly. Lara Hogan's writing on leadership transitions treats the handoff as its own craft, and it is.

The best fractional engagement ends with the company hiring a full-time CTO on strong footing: architecture sound, team built, processes working, incident patterns established. The new hire inherits a functioning technical organization instead of a pile of decisions to unwind. Our success is measured by how well the organization runs after we leave, which is the opposite of the consulting incentive to extend the engagement.

So before you sign the offer letter for the candidate your investor introduced you to last weekend, ask the honest version: do I need someone making twenty-five decisions this year, or two thousand? Am I buying hours, or judgment?