Wrivio
Get Wrivio
6 min readBy Wrivio Team

How to Give a Time Estimate in Writing Without Overpromising

Someone asks “how long will this take?” in a chat thread, and you have about thirty seconds to answer. The honest answer is “I am not sure yet.” The tempting answer is “Should be done Thursday.” Thursday then becomes a promise, a calendar entry on someone else’s plan, and the number you are measured against.

The problem is rarely the estimate itself. It is that a bare date in writing carries no assumptions, no confidence, and no expiry. The reader treats it as a commitment, and you meant it as a guess.

Research on the planning fallacy has long found that people tend to underestimate how long their own tasks will take, even when they know similar tasks ran late before. The classic paper is Buehler, Griffin and Ross, “Exploring the planning fallacy” (1994). So your first instinct is probably optimistic, and the message you write is the place to correct for that.

Give A Range, Not A Single Date

A single date implies certainty you do not have. A range tells the reader how much to trust it, and it protects you without sounding evasive.

Before:

Should be done Thursday.

After:

I expect this to land between Wednesday and Friday. Thursday is the most likely day.

Keep the range honest. “Between Monday and next month” is not an estimate, it is a shrug. If your range is that wide, say what you need to find out to narrow it, which is the next section.

The most likely day is worth naming because planners will pick one date to schedule around. Give them the one you actually believe, and let the range carry the risk.

State The Assumptions The Estimate Depends On

An estimate is a conditional statement. Write the condition down, so that when it breaks, the conversation is about the condition and not about your reliability.

Before:

Two days.

After:

Two days of focused work, assuming the data export arrives by Tuesday and the design needs no more than one round of changes.

Two or three assumptions is plenty. List only the ones that could realistically fail, such as an input you are waiting on, a review you need, or another project that could pull you away. Do not pad the list to look careful.

Name What Would Change The Number

Readers are more forgiving of a wide estimate when they can see how it will narrow. Tell them what unknown you are resolving and when.

I cannot narrow this until I have seen the vendor’s API limits. I will have that by end of day Wednesday and will send a firmer date then.

This turns “I do not know” into a plan. It also gives you a built-in reason to send the update, which matters because the worst version of an estimate is the one that quietly stops being true while nobody says anything.

Date-Stamp The Estimate And Promise An Update

An estimate has a shelf life. If the thread is still being read in two weeks, the reader should know when the number was made.

Estimate as of Monday 3 November: Wednesday to Friday. I will update this thread by Wednesday noon either way.

“Either way” is the important part. It tells the reader that silence will not be the signal that something slipped. If it does slip, the right move is a short, early note, and how to write a project risk update covers exactly that. If it has already slipped, see how to explain a missed deadline professionally.

Separate Effort From Elapsed Time

Two days of work is not two days on the calendar. Meetings, reviews, and waiting on other people stretch it. Readers usually hear the calendar version, so say which one you mean.

Before:

About three days of work.

After:

About three days of work, which with my current meetings means finished by the following Wednesday.

If a stakeholder is trying to pull a date earlier, you will be on the other side of this conversation, and how to ask a team to move a deadline up shows the framing that keeps it fair.

A Wrivio Context for estimates could say:

Rewrite this as a short, professional time estimate. Give a range with the most likely date, list at most three assumptions, and say when the estimate will be updated. Keep every name, date, figure, and commitment exactly as written. Do not turn a range into a single date, and do not add reassurance I did not write.

Press Ctrl+Shift+Space, paste the draft, and check the diff. Watch for any place the rewrite quietly removed a hedge or sharpened a range into a promise; that is the one change you do not want.

Common Questions

Should I pad my estimate?

Adding a flat buffer in your head and reporting one number hides the reasoning. It is usually better to report the range and the assumptions, so the reader can see where the risk is instead of guessing how much slack you added.

What if my manager wants one date?

Give the date you believe is most likely and keep the range next to it. A line such as “Thursday is the plan; Friday if the review runs long” gives them something to schedule against without removing the information that protects you both.

How do I estimate work I have never done before?

Say so, and estimate the first step instead of the whole. Offer a firm date for finding out more, then commit to a revised estimate at that point, which is more honest than a precise-looking number built on nothing.

When should I update an estimate I already gave?

As soon as you know it is wrong, and on the date you promised if nothing has changed. An early note that the range has moved is far less costly than a missed date discovered by the person waiting for it.

Download Wrivio for Windows to tighten an estimate in seconds while keeping the hedges and dates that make it honest.