In this note
The team says the new workflow saves 40 hours. Finance still pays the same salaries. Some people clear a backlog, someone stops working late, and a few hours have no assigned use. Multiplying all 40 by an hourly rate would conceal those different results.
I’d ask the owner to account for where the time went before calling it a financial benefit. Released capacity is useful information. Its value depends on what the organization does with it and what it spent to create it.
Measure the time after the new work is included
In this hypothetical 28-day exercise, the same onboarding tasks used to take 80 hours and now take 30. The new process also needs six hours of review and four hours of support. That leaves 40 hours free, not 50. Include the new work even if another team does it.
- Task-level reduction
- 80 hours − 30 hours50 hours
- New review and support
- 6 hours + 4 hours10 hours
- Net capacity released
- 50 hours − 10 hours40 hours
In your own comparison, keep the case IDs, tasks, kinds of work included, period, and checking method. Look for corrections left unfinished or moved into an unmeasured mailbox. Count hands-on work separately from waiting; fewer staff hours won't necessarily mean faster onboarding for the customer.
Note whether you timed the work, inferred it from system events, or asked someone to estimate. Each has limits. For the calculation here, assume the workflow owner accepts the invented time study. A live claim would need its own sampling evidence and uncertainty estimate.
Give each released hour one use
Of the 40 free hours, ten replace work that would otherwise need paid overtime. Twenty go into ten additional onboarding cases at two hours each. The other ten have no recorded use by the cutoff. Those three uses add up to 40; no hour can appear twice.
| Use of released capacity | Result to report |
|---|---|
| 10 hours remain unused | Available capacity; no monetary benefit assigned |
| 20 hours support additional work | 10 additional independently accepted activations; monetary value remains unknown |
| 10 hours replace overtime | A separate payroll comparison supports $450 lower paid overtime within the scenario |
The extra cases must represent work customers wanted and successfully used. In the example, ten qualifying requests from the backlog reach first valid use, with current access and support in place. The user and a separate checker confirm the result. Sending invitations or recording logins wouldn't establish that.
Track those ten cases separately from the ones used to estimate time saved. Check whether they would have finished this period anyway. More demand or another team's help could explain the increase. We haven't put a monetary value on the extra completed onboardings.
Check the expense that changed
The invented payroll figures show 14 paid overtime hours in the comparison period and four now, at $45 an hour. That is $450 less overtime within the scenario. Assume the owner has checked that this reduction came from the ten hours assigned to replacing overtime. Regular salaries stay the same.
- Prior overtime expense
- 14 hours × $45$630
- Current overtime expense
- 4 hours × $45$180
- Expense reduction
- $630 − $180$450
For a real claim, check that link rather than assuming it. Workload, staffing, overtime policy, holidays, and other process changes can affect the bill. Compare similar periods or use a stronger comparison if they differ. A quiet month could reduce overtime without any help from the automation.
Don't also multiply those ten hours by a salary rate and add that as another saving. The $450 already counts their financial effect. The twenty hours spent on extra onboarding belong to the extra-output result, not reduced expense. If you later price that output, include its own costs and avoid counting the same benefit twice.
Leave hiring avoidance in the forecast
The manager hopes to avoid a planned hire too. Check expected demand, the approved plan to hire without this change, and capacity the service can sustain. Keep that claim in a forecast with an owner and review date, outside this period's financial total. One good month or an open vacancy doesn't prove an avoided salary. When the future period arrives, count each hour's use once.
The ten unused hours can stay reported as unused. The owner may assign useful work later or reduce the service's scope. For now, show available capacity and the cost of creating it, without inventing a financial benefit.
Price the added service beside the supported benefit
We can now ask a narrower financial question: does the supported overtime reduction cover the resources used by the added service? We cannot yet answer whether the whole change is worth doing, because the extra onboardings have no agreed monetary value and possible harm and loss remain unknown.
The added service uses $700 of resources this period: $400 for the six review and four support hours, $100 for tools, $150 as this period's share of delivery cost, and $50 for operation and recovery excluding labor. All amounts are illustrative USD. The $400 assigns a cost to staff time used by the service; it doesn't mean payroll rose by that amount.
This is an added-service comparison, not a bill for the whole onboarding process. Of the original 80 task hours, 30 remain in the changed process and stay in the work record. The twenty released hours used for extra onboarding belong with that extra output. Include their additional costs if you later calculate its monetary benefit.
The ten review and support hours appear in both the time record and the cost record. They reduce released capacity from 50 hours to 40, and use $400 of staff resources. There is no second monetary subtraction here: the 40 hours have not been priced as a financial benefit. If you price net hours in another comparison, reconcile the labor treatment so you don't subtract it twice. Assume these ten hours include all added staff work; the $50 row adds no person-hours. Include further maintenance or recovery effort once if a real service needs it.
- Allocated cost of the added service
- $400 + $100 + $150 + $50$700
- Supported expense reduction less that cost
- $450 − $700−$250
The supported $450 overtime reduction is $250 below the $700 cost assigned to the added service. That answers the narrow question: the overtime reduction alone doesn't cover those resources. It does not establish the complete net value of the change. Keep extra-onboarding value and possible harm and loss unknown, rather than inserting zero, and track actual payment timing and the full one-time build cost separately.
Decide what to check before expanding
Before expanding, ask the owner to check the extra demand and its value, whether lower overtime persists, and what to do with unused hours. Keep exceptions, conflicting records, and the original period's report. If the benefits can't support the cost of operating safely, limit the service or test the missing evidence before expanding it.