The AI ROI Equation: Why Multiplication Beats Addition
Updated
Knowledge on this page was mainly distilled from The Tokenmaxxing Equation: AI ROI Was Never About the Tokens.
AI ROI is not a sum. It is a product of four terms, and because it multiplies, every term can veto the others.
The Equation
Value created (V) equals human capability (H) times AI capability and task fit (A) times the economic leverage of the problem (L) times your ability to capture the result (C): V = H × A × L × C. Call the total cost K (your hours plus the model bill). Then: AI ROI = (V − K) / K.
Why Multiplication Matters
We habitually count work as if it adds up: more tokens, more output, more tickets, one running total you grow by feeding it. Almost nothing that creates value behaves that way. In a multiplicative system, there is no safe place to be weak. Push any single term to zero and the entire product collapses, regardless of how large the remaining terms are.
A brilliant engineer (high H) with a capable model (high A) who captures value cleanly (high C) still produces nothing if the problem does not matter (L near zero). Cheap tokens cannot rescue a zero. They can only make the zero arrive faster at a lower unit price.
Q&A
What are the four terms of the AI ROI equation?
Human capability (H), AI capability and task fit (A), problem leverage (L), and value capture (C). These four multiply to produce total value. Because multiplication governs the relationship, strength in three terms cannot compensate for weakness in the fourth. A zero anywhere zeroes the whole product.
Why is human capability usually the biggest swing factor?
A capable person picks a problem worth solving, provides context that makes the model's answer good, catches wrong answers quickly, and turns rough output into something that survives in the real world. The METR 2025 study found experienced developers were actually 19% slower with AI tools while believing they were 20% faster. The model did not change between the two groups; the human judgment did.
How does problem leverage affect the equation?
Leverage (L) determines whether the value at stake is large enough to matter. Pointing your best person and best model at a trivial problem spends premium attention to produce something elegant and worthless. High leverage means the solved problem unlocks significant economic value, making even costly token spend rational.
What does value capture mean in this framework?
Value capture (C) is your ability to turn valuable output into economic return. You can produce genuinely good work and still see no return if you cannot ship it, sell it, get users to adopt it, or prevent a competitor from taking the upside while you absorb the cost. The value existed; it just never landed in your account.
How should I use this equation practically?
You do not plug in real numbers. You use it to diagnose which of the four terms you are actually short on. If your output is good but nothing ships, capture is your constraint. If output ships but barely matters, leverage is your constraint. The equation is a diagnostic, not a calculator.