The finding
Arclight has a problem. Two obvious fixes present themselves — improve the worst funnel stage, and add capacity. Here is what each is worth.
| What you fix | Deals closed | Annual margin | Gain |
|---|---|---|---|
| Nothing | 20 | $1,440k | — |
| The funnel alone | 20.6 | $1,485k | +$45k |
| Capacity alone | 20 | $1,440k | $0 |
| Both | 30 | $2,160k | +$720k |
The parts sum to $45k. The whole is $720k. Sixteen times.
That is not diminishing returns and it is not a rounding difference. The levers multiply. And it explains something that happens constantly in real businesses: a company diagnoses correctly, fixes the constraint, measures +$45k against a +$720k forecast, and concludes the diagnosis was wrong.
It wasn't wrong. It was incomplete.
