Did You Save Time, or Hand Off the Thinking?

The new reports are open. So are the old ones.

Imagine you prepare the shared work for a small team. Each lead sends a short weekly activity summary. Their manager asked for those summaries, and AI is helping them arrive earlier.

The manager is pleased. Less time writing reports. More time for the work.

You are looking at this week’s summaries beside last week’s, trying to establish what changed. Every status in the new reports is accurate. Each sender has done what the manager asked. You still have to make the comparison before preparing the team’s next conversation.

One item says waiting. It said waiting last week, too. The underlying notes show a difference: last week it needed an internal review; now that review is complete and it needs information from another team.

The status is correct. The preparation it calls for has changed.

Finding that change became your job without anyone assigning it. The manager’s account of the time saving ended when the summaries arrived.

The saving has another side

BetterUp Labs and Stanford Social Media Lab use workslop to describe apparently useful AI-generated material that leaves colleagues with more thinking and cleanup. Their public research page describes an online survey of U.S. desk workers conducted in September 2025. It is a report of respondents’ experiences, not a measurement of your team’s productivity.

Use that outside lens to ask a local question: what work begins after the output leaves the sender?

Our reporting example is hypothetical. It does not require an invented statistic to expose the question. The sender’s writing time and your preparation time are both part of getting the team’s work ready.

Here’s the thing. You are calling work faster because you stopped counting when it left your desk.

That does not mean the sender failed. In this example, the summaries meet the request. It means the request and the saving claim cover different amounts of work.

The manager asked for a summary. The team also needs someone to identify changes across those summaries. Praising the first part tells you little about what happened to the second.

Count the work left in the other chair.

They did what you asked

This is where a manager can make the wrong correction.

“These reports need to be more useful.”

Useful for whom, doing what? If the sender met the existing request, that instruction gives them a new criticism without a new job.

Sit with the receiver and follow one report through what happens after arrival. In our example, the receiver opens the earlier version, finds the matching item, checks the underlying notes, and works out whether the change affects the team’s preparation.

Some of that work may belong there. Some may be easier for the sender to supply from the information already in front of them. You cannot tell which by looking at how quickly the report was written.

The Professional Recipe calls its evidence-and-revision rhythm Feedback. Here, the receiver’s experience is evidence about whether the recurring assignment supplies what the next person needs.

The change follows the evidence. If comparing the two weeks reveals a useful difference the sender already knows, revise the request to carry that difference. If the receiver is making a judgment across several teams, keep that judgment with the person responsible for it.

Right? Feedback is not an invitation to return every summary for more detail. It should help you decide which detail would remove a particular piece of reconstruction.

Keep the judgment in the right chair

Return to the item marked waiting.

A useful addition from the sender could say that the internal review is complete and the item now needs information from another team. That tells the receiver what changed without pretending the sender has decided how the whole team should respond.

The receiver may still need to weigh that dependency against other work. They may decide it needs attention now, or that another issue matters more. That is their responsibility in this hypothetical arrangement.

You are not trying to eliminate the receiver’s thinking. You are trying to stop making them rediscover information that could have traveled with the report.

That distinction protects both people.

The sender does not have to invent a recommendation outside their role to make the report look complete. The receiver does not have to perform the comparison again merely to reach the point where their actual judgment can begin.

An unchanged item needs honest treatment too. If the sender has checked and nothing relevant changed, they can say so. If they do not know whether something changed, that remains an unanswered point. An empty change section cannot quietly mean both.

There is no need for a longer report by default. One useful sentence can do work that several polished paragraphs leave behind.

The improvement is not the extra sentence. Let me tighten that. The improvement is the comparison someone no longer has to reconstruct, if the new arrangement actually supplies it.

Your praise teaches people where to stop

Imagine the manager congratulating the leads for producing their summaries quickly while asking the receiver why preparation still takes so long.

Same exchange. Two different accounts of what good work looks like.

The sender’s contribution is visible when the report arrives. The receiver’s contribution happens afterward, among old reports and underlying notes. If the only thing the manager asks about is writing speed, the second part can remain invisible.

You do not need to decide that anyone is lazy to see the incentive. People can follow the assignment faithfully while the manager rewards a finish point that leaves necessary work uncounted.

Make it ordinary for the receiver to describe that work in concrete terms.

“I compared these two versions to find the change before I could prepare our discussion.”

That gives the manager something to inspect. It is more useful than declaring the reports bad or telling the receiver to be less particular.

And if you are the sender’s manager, make the change to the assignment yourself. Do not leave the receiver to negotiate a new expectation every week while the old request stays in place.

Yeah. The person absorbing the extra work should not also have to become the unofficial manager of the reporting process.

More work upstream is still work

There is another easy way to produce a flattering saving: ask senders to do more, then count only the receiver’s improvement.

Suppose each lead now spends extra time identifying changes. The receiver spends less time comparing. That might be a better arrangement. It might also create duplicate checking, or move a comparison to someone who lacks the information to do it well.

Look at what happened to both sides.

If the sender can identify the change from their normal work, carrying it forward may be straightforward. If the new requirement makes every sender rebuild the same shared comparison, the assignment may need a different adjustment.

Do not declare the saving because one person reports a better experience. Ask what that experience now requires from the other person.

You can also choose a change for a reason other than speed. Perhaps a clearer account makes the shared discussion easier to prepare even if total effort does not fall. Say that. A useful reason does not become stronger when you relabel it time saved.

Does that make sense? You are allowed to improve the exchange without claiming a number you have not established.

Your Monday Move: look at one exchange

Choose one recurring internal output and sit with its usual sender and receiver. Bring three recent ordinary examples and the records both people already have permission to use.

Ask what the receiver could use immediately, what they had to find or compare, and what judgment properly remained theirs. Ask the sender what the original request required and what supplying the missing comparison would involve.

Name one avoidable step. Agree who should own it and what the next output should carry. If responsibility is still unclear, name the person who can settle it. You do not need to redesign every report before fixing this one exchange.

Use any effort information the two people actually know. If they do not know the time, leave it unknown. Three examples can expose a repeated task; they cannot establish a company-wide productivity result.

Try the revised request on the next ordinary instance. Ask whether the receiver still had to reconstruct the same thing, and what additional work the sender did. Keep or revise the agreement based on that answer.

Then return to the manager’s original claim that the work got faster.

Whose work did that sentence include?