Tool

AI value calculator

How much of your AI spend, and how many of your use cases, have a measured outcome behind them? Runs in your browser. Nothing is stored or transmitted.

Five figures for the past year, spend in thousands (£k). Showing an illustrative example: spend from the Value Gap essay's worked portfolio, use-case counts hypothetical.

Currency

Spend

Total AI spend, £k a year

All identified AI spend: direct consumption, licensed seats, embedded SaaS tiers and infrastructure.

Spend connected to a measured outcome, £k a year

Spend with a baseline, a named owner and evidence of change. Spend that is only allocated to a team or use case does not count.

Use cases

Number of AI use cases

Material use cases you could name and give an owner.

Use cases with a measured outcome

Those with a baseline and a measured result, by the same test as above.

Optional

Estimated value delivered, £k a year

Leave at zero if you do not have an estimate. Use measured value, net of review and rework, where you can.

Attribution coverage

30%

Share of spend

Outcome measurement

25%

Share of use cases

AI value density

–

Add an estimate of value

Partly evidenced. £3,000,000 of £10,000,000 is connected to a measured outcome, leaving £7,000,000 unevidenced. A minority of AI spend is connected to a measured outcome. Leaders can defend part of the portfolio but not most of it.

Evidence behind the portfolio

Spend connected to a measured outcome30%

Use cases with a measured outcome5 of 20

To track these figures each quarter: AI Value KPI Set by Layer and AI Use Case Business Case.

Attribution coverage divides by all identified AI spend, not only the spend already allocated, which would flatter it. The bands describe your figure; they are not an industry benchmark, because no reliable one exists yet. For coverage by spend category, use the attribution coverage calculator.

Attribution coverage

30%

See breakdown

About this calculator

What it answers: What share of AI spend is connected to a measured outcome, what share of use cases is measured, and, if you have an estimate of value, how much value each unit of AI spend returns.
How it calculates: Attribution coverage is spend connected to a measured outcome divided by all identified AI spend. Outcome measurement coverage is use cases with a measured outcome divided by all use cases. AI value density, as this tool measures it, is estimated value a year divided by AI spend a year. The reading is chosen by coverage: below 25%, 25% to 50%, 50% to 75%, and 75% or more.
When to use it: A first, quick read of how much AI spend and how many use cases have evidence behind them.
Read next: The method is set out in The AI Value Gap, and value density in AI Value Management Layers. To see which layers of evidence are missing, take the AI value maturity self-assessment.
Take it further: The AI Economics Lab costs a whole use case, keeps the evidence behind it and prepares an investment review, saved only in your browser.