From hours saved to growth: AI return scorecard
For anyone reporting AI results to leadership who today has only hours saved to show.
When to use it
- Your AI report says "we saved 300 hours" and leadership asks what happened to those hours.
- You have several live use cases and none of them has a business number next to it.
- You were asked for the return on AI and the only thing measured is efficiency.
The sheet
One row per use case. Hours saved stay; what gets added is a growth indicator, an owner and the platform it runs on.
| Use case | Hours saved per month | Growth indicator | Value today | Target and date | Owner | Approved platform? |
|---|---|---|---|---|---|---|
| [process] | ____ | [sales, new clients, clients served, response time] | ____ | ____ | [name] | Yes / No |
Illustrative example: "Quote drafting · 40 hours · client response time · 3 days · 1 day by 31 March · Marta, sales · Yes".
Where do the hours saved in this row go? ______________________________
How to fill it in 10 minutes
- List the use cases already running, not the planned ones. One per row.
- Copy the hours saved you already report. Do not recalculate them; the goal is to add, not replace.
- Pick a single growth indicator per row: sales, new clients, clients served or response time. If you cannot find one, that row is efficiency only and it is worth saying so.
- Put a named owner from the area that uses the tool.
- Mark whether the case runs on the company's approved platform or on a tool someone tried on their own.
- The most common mistake: turning hours into money by multiplying by an hourly cost and calling it return. A saved hour nobody reinvests shows up in no report; the growth column exists to know where it went.
What the study says
Fact: Wharton and GBK published their annual report on AI adoption in companies on 13 October 2026. Among companies that measure at least one growth indicator, such as sales or new clients, 35% report a strongly positive return. Among those that measure only hours saved, 13%.
Fact: among companies that use a single approved platform, half report a strongly positive return. Among those still trying everything, 9%.
My read: a saved hour nobody reinvests shows up in no number. Measuring growth forces a decision about where that hour goes. And standardising on one platform does not cause the return by itself; it is a relationship, not a cause. What it does do is put the use case's owner, data and support in the same place.
Vic Sol · Decision Desk · vicsol.org/resources/ · Free to use with attribution. This sheet structures a conversation; it is not legal, financial or audit advice.