The number on the vendor's quote is the cheapest part of any AI system you will ever run. By the time it is in production, integrated, governed and maintained, the licence or token cost is typically 8 to 15 per cent of what the thing actually costs you over three years. Everything else lives below the waterline.
This is why so many AI business cases fall apart eighteen months in. PwC's January 2026 CEO survey found that 56 per cent of leaders had seen neither revenue gains nor cost savings from AI, and only 12 per cent reported both. That is not because the technology does not work. It is because the cost base was scoped like a software purchase when it should have been scoped like an operating model change.
For a COO, this is the central question. You own the processes AI is supposed to improve, the people whose work it changes, and the operational risk when it goes wrong. If anyone in the business should hold the true total cost of ownership picture, it is you. Most of the time, nobody does.
Why AI breaks the normal TCO model
When you buy a conventional SaaS product, total cost of ownership is reasonably predictable. Licence, implementation, some integration, training, support. You can scope it, fix it, and hold a vendor to it.
AI does not behave that way, for three reasons.
The cost is consumption-based and it moves. A simple linear workflow might cost a few pence per interaction. A more complex 2026-era system, one that calls tools, reasons over several steps and loops until it gets an answer, can cost a pound or more per interaction, an order of magnitude higher. As you make the system more capable, you make it more expensive to run, and that relationship is not linear. Your unit economics shift under you as usage grows and as the system gets smarter.
Most of the work is not the model. Independent analyses through late 2025 and early 2026 consistently put data preparation at 60 to 75 per cent of total effort on AI initiatives. The model is the easy bit. Getting your data clean, accessible, correctly permissioned and continuously maintained is where the real spend sits, and it is spend that recurs.
It is never finished. A traditional system is built, deployed and then maintained at low cost. An agentic AI system needs continuous evaluation, prompt tuning, monitoring and adaptation as the underlying models change and your data drifts. Most enterprise deployments carry a real monthly running cost the original business case never mentioned: tokens, vector storage, observability, security and tuning. It is an operating function, not a project.
Treat AI like a one-off purchase and you will underestimate the true cost by 40 to 60 per cent. That is not a rounding error. That is the difference between a business case that holds and one that quietly bleeds.
The seven cost layers below the waterline
When we build a TCO picture for a client, we account for seven layers. The vendor quote covers, at most, the first.
1. Licence and consumption. The visible cost. Per-seat, per-token, or per-call. Model this at realistic production volume, not pilot volume, and stress it for the case where adoption succeeds and usage triples.
2. Data readiness. Cleaning, structuring, pipelines, permissioning, ongoing data quality. This is the single largest line in most AI initiatives and the one most often left out entirely. If your data is not ready, and for most mid-market operations it is not, this dominates everything else.
3. Integration and engineering. Connecting the system to your existing stack, building the orchestration around the model, handling the cases where it fails gracefully. The model is 8 to 15 per cent of build cost; this is most of the rest.
4. Governance and assurance. Model validation, monitoring, audit trails, human-in-the-loop review, the controls a regulator or your own risk function expects. In financial services, insurance and legal this is non-negotiable and it is a standing cost, not a one-off.
5. People and change. Training, role redesign, the productivity dip while people learn the new way of working, and the management time to lead it. The technology is often the cheapest part of the change; the human side is where value is won or lost. We have written before about why AI transformation is 70 per cent people.
6. Run and maintenance. The recurring monthly cost of keeping it working: tuning, monitoring, security and re-evaluation as models are deprecated and replaced. Budget for this as a permanent operating line.
7. AI debt. The structural cost of moving fast without architecture: duplicated tools, ungoverned pilots, models nobody owns, integrations that quietly rot. It does not show up on any invoice, which is exactly why it is dangerous. We cover this in AI debt: the cost of moving fast without architecture.
A mid-complexity operations agent that a naïve estimate prices at around £150k over three years routinely lands closer to £350k once these layers are counted. The agent is the same. What changed is how thoroughly it was costed.
What this means for the COO specifically
The CFO will ask what it costs. The CTO will ask whether it works. The COO sits at the only vantage point from which the whole picture is visible, because the hidden costs are operational costs: they land in your function, on your people, in your processes.
You see the consumption curve before finance does. Usage-based pricing means the cost follows operational adoption. If a tool genuinely lands and your teams lean on it, the bill climbs. That is success creating cost, and you will see it in the workflow long before it shows up in a quarterly variance report.
You own the data readiness problem. The 60 to 75 per cent of effort that goes into data is your operational data, the records, the processes, the exceptions your teams handle every day. No vendor can fix that for you. It is operational hygiene, and it determines whether any of the rest works.
You absorb the change cost. The productivity dip, the retraining, the role redesign, the people who need a new path. All of that is operational reality in your function. Scoped properly, it is manageable. Ignored, it is the reason adoption stalls and the business case never pays back.
You carry the run cost forever. Once it is live, keeping it working is an operational function. Someone owns the tuning, the monitoring, the response when a model is deprecated. If that someone is not named and funded, the system degrades quietly until it fails loudly.
The COOs getting real return from AI in 2026 are not the ones with the cleverest models. They are the ones who scoped the true cost up front, sequenced the spend deliberately, and refused to let a pilot become a production dependency without an operating model behind it.
How to build a TCO picture that survives contact with reality
You do not need a procurement spreadsheet with forty tabs. You need disciplined answers to a small number of hard questions before you sign anything.
Model production volume, not pilot volume. Ask what this costs at full adoption, and again at three times that. If the economics only work at pilot scale, you do not have a business case, you have a demo.
Cost the data work properly. Before you commit to the model, get a real assessment of whether your data can feed it. If it cannot, that work is your first cost and your longest, and it belongs in the business case from day one. This is exactly what a structured readiness assessment is for.
Name the run owner and fund the run line. Decide now who owns the system in production and what the standing monthly cost is. An AI system with no named owner and no run budget is AI debt with a launch date.
Count the people cost as a line, not an afterthought. Training, change, role redesign and the productivity dip are real and they are yours. Put a number on them.
Sequence the spend. Start where the cost is low and the value is obvious. Build the governance you will need anyway. Prove the economics on something small before you commit to something large. Most of the 95 per cent of pilots that fail to scale failed because nobody costed the path from pilot to production.
Counting every layer is the cheaper path
It feels more expensive to count all seven layers. It is not. The expensive path is the one where you scope AI like a software licence, get six months in, and discover the real cost as a series of unbudgeted surprises, by which point you are committed, the pilot is load-bearing, and walking away is harder than overspending.
A true total cost of ownership picture, built before you sign, does three things. It kills the initiatives that were never going to pay back, before you spend on them. It properly resources the ones that will. And it gives you, the COO, the one thing AI business cases almost never come with: a number you can actually stand behind.
That is the difference between AI as a line of operational cost you control and AI as a liability you discover. The work to tell them apart is not glamorous. It is just a clear-eyed cost picture and the discipline to act on it.
If you want help building a real total cost of ownership picture for an AI initiative, or pressure-testing one you already have, get in touch. Our Breathe assessment is built to surface the costs below the waterline before you commit, and our fractional Chief AI Officer support keeps the run economics in check once you are live.