Deployment Labs
Module 1 · Lesson 2 of 5 · 14 min

The goal, and where freed time goes

A goal you can measure, and a destination for every hour AI gives back.

An aim has two halves. The goal says what you want more of. The destination says where the time AI gives back will go. Most rollouts have neither, so the time leaks away and nobody can say what changed. You'll write both in this lesson, and they go straight onto your aim page.

Time saved is not value captured

AI does give people time back. The problem is what happens to it next.

42%

of regular frontline AI users save a full working day a week.

BCG, 2026

66%

of them get little or no guidance on what to do with that time.

BCG, 2026

>50%

don't redirect the time into more valuable work.

BCG, 2026

Saved time doesn't vanish. It turns into more email, more meetings, more drafts and more rework, because nobody decided otherwise. The hours are real. They just never reach anything that shows up in the business.

Play with the numbers below. Put in roughly how many people use AI and how much time they save, then see what naming a destination is worth over the first 90 days.

Try itThe freed-time ledger
05001,0001,5002,000Day 0Wk 2Wk 4Wk 6Wk 8Wk 10Wk 121,5601,170624
Hours saved Put to work, destination named Put to work, no destination

By day 90: about 1,170 hours put to work with a destination named, against 624 without one. The difference, 546 hours, is what naming a destination is worth.

Illustrative. The 40% share with no destination is an assumption, not a measured figure: BCG found most people who save time get little or no guidance on using it. Your pilot measures the real numbers.

Write a goal you can check on day 90

A goal has four parts. Miss any one of them and you won't be able to tell, on day 90, whether it worked.

  • A measure: something you can count or time.
  • Where you are now: today's number, even if it's rough.
  • A target: where you want to be.
  • A date: usually the end of this loop or the next.

“Use AI to improve customer service” isn't a goal. “By 31 March, time to first reply goes from 9 hours to 2” is. Pick an outcome your leadership team actually cares about this quarter. You don't need to know your bottleneck yet: module 4 finds what limits the goal you set here.

BuildWrite your goal
Start from an example:
Your goal

By a dateDate, what you measureMeasure goes from today's numberNow to the targetTarget.

MeasureWhere you are nowTargetDateTarget differs from todayDate is in the future
Efficiency, growth, or both

A goal can be about doing the same with less, or doing more with the same: faster quotes, more proposals, a quicker close, more customers served. Both count. Inventing a whole new business model is outside this course.

Name where freed time goes, before it arrives

Freed time can go to four places. Your aim names one or more of them in advance, so you can count what was put to work and not just what was saved.

  • The bottleneck: more hands or attention on the step that limits the goal.
  • Lower cost: less overtime, fewer contractors, slower hiring.
  • Better quality: fewer errors, more checking, better work for customers.
  • Growth: serving more customers, selling more, trying new offers.

For a first loop, we recommend the bottleneck. It's the only destination that moves the goal directly, and it's the one your day-90 measure will see most clearly.

BuildName where freed time goes

Pick one or more. Freed time only counts once it has gone to one of these.

The idea

Freed time only counts once it has gone somewhere you named in advance.

CheckQuick check

Your team saved 300 hours last month using AI. What value did it create?

Your goal and destination are saved to your aim page. Next: the four promises, which make it safe for people to tell you what's really happening.

Finished this lesson?

Mark it complete. Your answers are already saved in this browser.

← Previous
Why most AI rollouts return nothing
Next lesson →
The four promises