Who Should Own AI Strategy in a Mid-Market Company?
Not the CTO. Not the CIO. Not a committee. The honest answer to a question most leadership teams get wrong.
"Who owns AI at your company?" is a question with four common answers, three of which are wrong. This post walks through each and lands on the ownership pattern we see actually work in mid-market companies with $10M-$100M in revenue.
The four common answers
1. "The CTO"
Instinctive, and usually wrong for the strategic layer. CTOs at mid-market companies are focused on the product roadmap and engineering delivery. AI strategy inside a mid-market business is mostly not a product decision — it's an operations decision. Putting the CTO in charge tends to over-index on custom builds and under-index on process automation of the sales, ops, and finance functions where the ROI actually lives.
The CTO owns AI inside the product. That's a real, distinct scope.
2. "The CIO"
Closer, but still incomplete. The CIO owns the systems AI plugs into and often owns the security review. But the CIO rarely has the P&L authority to make prioritisation calls between operations, sales, and finance. Ownership without authority is bureaucracy.
3. "A steering committee"
The default answer, and the slowest one. Steering committees produce alignment and consume months. They're useful as a governance forum. They're terrible as an owner.
4. "A Chief AI Officer"
Sometimes right, usually premature. If you have $500M+ in revenue and are running multiple AI programmes across divisions, a dedicated leader makes sense. If you're a $30M business trying to figure out where to start, hiring a CAIO first is the wrong sequence. Do the audit, ship the first three projects, then decide if you need a full-time exec.
What works
In our experience the pattern that actually ships looks like this:
- Executive sponsor: the CEO or COO. Not a delegated owner. The person who can kill a bad project without a meeting.
- Programme lead: a senior operations leader with credibility across departments. Often the COO or a VP of Ops. Owns sequencing, prioritisation, and change management.
- Technical partner: internal or external. The person who translates process opportunities into technical specifications and vendor selection.
- Governance forum: a monthly review with finance, legal, and IT security. Reviews, doesn't own.
Why this pattern wins
Three reasons. First, executive sponsorship removes the political overhead of cross-functional decisions. Second, an operations leader in the programme seat prevents the drift toward "we should build our own." Third, the governance forum handles risk without slowing down execution.
What to avoid
- Delegating ownership to a "special projects" hire with no P&L authority.
- Letting the technical partner drive prioritisation. Technology should serve the programme, not shape it.
- Standing up a Chief AI Officer role before you know what they'd do on day one.
The honest bottom line
AI in a mid-market company is 70% operations, 20% technology, and 10% governance. Ownership should reflect that mix. The CEO sponsors, the COO or ops leader runs the programme, and the CTO/CIO plays a role that fits their scope. Everything else is a bureaucratic reflex.
FAQ
Frequently asked questions
Do we need a Chief AI Officer?+
Not until you're past the first three projects and running multiple programmes concurrently. Then maybe. Not before.
Should the CTO run this?+
For AI inside the product, yes. For AI across the business, no — that's an operations leader's job.
How much time does the programme lead spend on this?+
Roughly 30% for the first six months, dropping to 15-20% once patterns are in place.
Building something similar?
Let's talk in 30 minutes.