The Leverage Ratio: one senior operator versus a staffed team

The Leverage Ratio is a simple measure: how much one senior person, using AI, can now deliver compared with a conventional staffed team. For most of my career it was low, which is exactly why serious work needed a team. It has risen sharply for a lot of AI implementation work, and that single shift is the reason I run my practice the way I do.

It is worth naming because most people are still pricing and staffing projects as if the ratio were where it was in 2015. They are not comparing like with like, and it costs them.

What the ratio measures

Not typing speed, and not raw output for its own sake. It measures how much of the work a project actually needs, from first scope to shipped system, one experienced person can now produce on their own by directing AI, rather than handing it down to specialists. The judgement stays human. The production that used to require a row of other people is the part that has changed.

Why it was low for twenty years

For most of my twenty years in digital, a senior person could think, decide, and design at a high level, but could not physically produce enough alone. You needed specialists: the build, the content, the data work, the research. Specialists meant a hierarchy, which meant handing work down and hoping it came back close to what you intended. That handing down has a cost of its own, the quiet quality loss I call the Handoff Tax. A low Leverage Ratio is the reason agencies are structured the way they are. It was not a racket. It was arithmetic.

What changed

AI collapsed the cost of the production layers. The first draft of the code, the first pass of the content, the data wrangling, the research scaffolding: the work that used to occupy a team for weeks is now something one senior person can direct in a fraction of the time, then bring their own judgement to. The scarce ingredient is still that judgement, knowing what to build and why this one first. But the team you used to need around it has thinned out. So the ratio rose, and for a good deal of AI implementation work it rose a lot.

A high Leverage Ratio does not make the senior person optional. It makes the team around them optional.

How to read it for your own project

The ratio is not the same for every job, and the honest use of the idea is knowing where it is high and where it is not.

• High ratio. Work that is mostly judgement plus production AI can shoulder under supervision: reworking a process around AI, building a bespoke internal tool, a first customer-facing feature. One senior operator beats a team here, on quality and on cost.
• Lower ratio. Work that genuinely needs many hands in parallel: round-the-clock operations, large-scale manual data collection, a programme spanning many teams at once. A team still wins, and I will tell you so.

Why it matters when you hire

When the ratio is high and you hire a junior-staffed agency anyway, you pay twice: once for the coordination overhead of a team you did not need, and again in the Handoff Tax as the senior who sold the work hands it to others. One senior operator, AI-leveraged, removes both. That is the whole case behind AI-Leveraged Delivery, and behind my argument for one senior person over a team of fifteen.

So the useful question before you commission AI work is not "how big a team can they put on it". It is "how high is the Leverage Ratio for this job, and am I paying for people I do not need". Ask it, and a lot of expensive proposals start to look like what they are.