Skip to main content

Requests With Please Run Late 38 Percent of the Time, What to Write Instead

Task requests containing please were completed late 38 percent of the time against 31 percent for neutral, direct asks. How to write requests that get done.

Requests With Please Run Late 38 Percent of the Time, What to Write Instead

Task requests containing the word please were completed late 38 percent of the time, against 31 percent for requests written in a neutral but direct fashion, according to a two year analysis of project management conversations reported by HR Dive. Requests containing ASAP or exclamation marks ran late at a similar rate to the polite ones. At Metaintro, we track roughly 50 million job postings , and communication is one of the most requested skills in them. The finding here is not that politeness is bad. It is that softening language often replaces the specifics that actually make a request actionable.

What Did the Analysis Actually Measure?

It looked at how work gets asked for and whether it then arrives on time. The data analysis, published on 6 August by Chaser, covered two years of project management activity conducted through Slack. The comparison that produced the headline was between requests carrying politeness or urgency markers and requests phrased in a neutral, direct way, with the polite version late 38 percent of the time and the neutral version late 31 percent of the time.

That is a seven point difference rather than a catastrophe, which is worth stating plainly before anyone rewrites their entire communication style. It is a meaningful edge rather than a transformation. The larger effects in the same analysis came from structure rather than wording, and those are the ones worth acting on first.

The study also found that deliverables were most often finished on time when tasks were completed voluntarily and when a single worker had ownership of each task. Both of those findings say the same thing in different ways. Clarity about who is doing something beats any phrasing of the ask.

Why Would Politeness Make a Request Later?

Because softening usually travels with vagueness. The words that make a request feel gentle are frequently the same words that remove the specifics, so a message that reads as courteous can leave out who exactly is doing it, by when and to what standard. The reader is left to infer all three, and inference is where deadlines go to die.

Urgency markers fail for a related but opposite reason. When ASAP or exclamation marks appear on requests that are not genuinely urgent, they stop carrying information. A team that receives three urgent requests a day learns quickly that urgency is a tone rather than a fact, and the marker gets discounted. The analysis found these carried a similar late rate to the polite requests, which is consistent with both failing to communicate anything precise.

None of this argues for writing rudely. A neutral and direct request is not a curt one, and the research compared neutral phrasing against softened and urgent phrasing rather than against courtesy itself. The practical translation is to keep the warmth in the relationship and take the ambiguity out of the request. Our guide to professional email etiquette rules covers the tone side, and why most meetings fail and how to fix them covers the same failure in a different format.

What Actually Gets Work Delivered On Time?

Single ownership, by a wide margin. Tasks with one owner were almost 70 percent more likely to be completed on time compared with tasks owned by two people. That is the largest effect in the analysis and it is the one most teams get wrong, usually with good intentions, by assigning a pair to something so that nobody is left unsupported.

Volunteering was the other strong signal. Tasks people volunteered to do were more than 70 percent likely to be completed on time. This has an obvious implication for how work gets distributed. Asking a team who wants to take something, before assigning it, costs one message and buys a meaningful improvement in delivery. It also surfaces capacity problems early, because silence in response to an open ask is information.

For managers who cannot always assign to one person or rely on volunteers, the analysis offered two structural fixes. Share the ask in a group message rather than one to one, because tasks shared in a channel were 44 percent more likely to be completed on time. And attach a concrete deadline, because deliverables without due dates were two and a half times as likely never to be completed at all. Our guide to fixing team communication without adding meetings covers how to do this without turning every request into a process.

How Should You Write a Request Instead?

Four elements, in one message. Name the person rather than the group, state the specific deliverable rather than the topic, give a date and a time rather than a relative word, and say what happens next so the person knows why the date exists. That last element is the one most people skip and it is what converts a deadline from arbitrary to reasonable.

Compare two versions of the same ask. The first says, when you get a chance please could you take a look at the client numbers. The second says, could you send me the corrected Q3 client numbers by Thursday at midday, so I can put them in the board pack that goes out Friday morning. The second is not less polite. It is simply complete, and completeness is what removes the delay.

Two habits make this easier. Write the deadline before you write the request, because it forces you to decide whether one genuinely exists. And put the ask in the first line rather than the last, since a message that opens with context and buries the request at the bottom relies on the reader getting to the end. Our guide to time management techniques covers the personal half of this and why multitasking damages productivity explains why fragmented requests cost more than the time they take to read.

Does This Change How You Should Ask Your Manager for Things?

It changes it in your favour, because the same rules work upward. Requests to a manager fail for exactly the reason described above, namely that they are softened into vagueness. Asking whether there might be any scope to discuss development at some point is a request with no owner, no deliverable and no date, and it will behave accordingly.

The structured version names all three. Ask for a thirty minute conversation in the next two weeks about what would need to be true for a promotion case, and say you will bring a written summary of your work. That is a request a busy manager can act on immediately, and the specificity signals seriousness rather than pressure. Our guides to getting career feedback from reluctant managers and to building a case your manager will approve both work on this principle.

The same applies to job hunting. A follow up that asks whether there is any update on the process is easy to defer. One that says you are checking whether a decision is expected this week, and that you have another process moving, is a request with a shape. Our guides to sending follow up emails after an interview and to what a resume email subject line should say apply the same logic to a hiring process.

Is This a Skill Employers Actually Hire For?

Increasingly yes, and it shows up under names that do not obviously describe it. Written communication, stakeholder management, project coordination and delivery ownership all describe some part of the ability to make work happen through other people without authority over them. In distributed teams it is close to the whole job for anyone coordinating across functions.

It is also one of the few skills that is easy to demonstrate in an interview. Being able to describe how you unblocked a stalled piece of work, and to name the specific change you made to how it was assigned or scheduled, is far more persuasive than claiming to be a strong communicator. The research gives you the vocabulary for it, since single ownership, volunteering, public asks and real deadlines are all concrete mechanisms.

The market context helps here. Communication consistently appears among the capabilities employers say they cannot find, and our coverage of where human communication still beats AI drafted messages explains why the judgement half is the part being paid for. A tool can produce a courteous paragraph in seconds. Deciding who owns the task and what date is realistic is not a drafting problem.

What About AI in All of This?

The analysis is careful to note that productivity problems now extend past human habits like politeness or forgotten deadlines. Managers and employees have both expressed concern about losing problem solving ability and developing dependence on AI tools, and some workers report spending more time managing AI than producing work, a phenomenon the report calls botsitting.

That matters for request writing specifically, because an AI assistant introduces a new ownership ambiguity. If a task is delegated to a tool, the question of who owns the output, who checks it and who is accountable for the errors becomes exactly the two owner problem the data identifies as the largest source of lateness. Naming a human owner for AI assisted work is the same fix as naming a single owner for anything else.

The report's practical suggestion for teams facing resistance or reluctance was learning support focused on coaching and time management. That is a reasonable ask to bring to your own manager if delivery in your team is slipping, and it is more constructive than a complaint. Our coverage of three practices for running team meetings in an AI era and of how AI is rewiring workplace dynamics covers where these habits are heading.

How Do You Fix a Team That Is Already Missing Deadlines?

Diagnose before you prescribe, because the three causes in this data need different fixes. If work is late but everyone is busy, the problem is usually ownership rather than capacity, and the test is simple. Pick five late items and ask who owns each one. If any answer contains the word we, you have found the issue, since tasks with a single owner were almost 70 percent more likely to land on time than tasks with two.

If work is late and nobody can say when it was due, the problem is deadlines. Deliverables without due dates were two and a half times as likely never to be completed at all, and in most teams that shows up as a quiet backlog of things everyone assumes someone is handling. A single pass through the backlog attaching real dates, or explicitly killing the item, usually recovers more capacity than any tooling change.

If work is late and the same two people are carrying it, the problem is distribution. Volunteering worked in this data, with volunteered tasks more than 70 percent likely to be completed on time, but volunteering only works when the whole team sees the ask. That is part of why the channel effect exists, with public requests 44 percent more likely to be completed on time than private ones. A request sent privately to your most reliable person quietly guarantees that your most reliable person is overloaded.

None of these fixes require authority, which is why this is a genuinely portable skill. An individual contributor who consistently names owners, sets dates and asks in the open changes the delivery rate of a team without ever being given the mandate to do so, and that is exactly the evidence a promotion case runs on.

What Does This Mean for Your Career Right Now?

Four changes, each of which takes less than a minute per message. First, name one owner. The single largest effect in this analysis was that tasks with one owner were almost 70 percent more likely to land on time than tasks with two, so the habit of copying in a second person for safety is costing delivery rather than protecting it.

Second, always attach a real date. Deliverables without due dates were two and a half times as likely never to be completed, which makes a missing deadline the most expensive omission available. If you do not know the date, say what it depends on and when you will confirm.

Third, ask in the open. Requests shared in a channel were 44 percent more likely to be completed on time than the same ask sent privately. Visibility creates accountability without anyone having to enforce it, and it also lets someone else flag a conflict you did not know about.

Fourth, ask for volunteers before assigning where the work allows it, since volunteered tasks were more than 70 percent likely to be delivered on time. Taken together these are the cheapest performance improvements available to anyone who works through other people, and they are entirely within your control regardless of your job title. Keep the please if you like it. Just make sure the message would still work without it.

Related Articles

People Also Asked

Q: Does saying please really make work take longer?

A: Requests containing the word please were completed late 38 percent of the time, compared with 31 percent for requests written in a neutral but direct fashion. Requests using ASAP or exclamation marks ran late at a similar rate to the polite ones. The likely reason is that softening and urgency markers often replace the specifics, such as the owner, the deliverable and the date, rather than that courtesy itself causes delay.

Q: What is the single best way to get a task done on time?

A: Give it one owner. Tasks with a single owner were almost 70 percent more likely to be completed on time than tasks owned by two people. After that, ask for volunteers where you can, since volunteered tasks were more than 70 percent likely to be finished on time, and always attach a due date, because deliverables without one were two and a half times as likely never to be completed.

Q: Should I ask for things privately or in a group channel?

A: In the channel, based on this data. Tasks shared in a group message were 44 percent more likely to be completed on time than the same request sent one to one. Visibility creates accountability and lets colleagues raise conflicts early. Sensitive or personal requests are the obvious exception and should stay private.

Ready to level up? The way you write a request is a skill employers pay for, and it belongs in how you describe your own work. Metaintro tracks roughly 50 million job postings so you can find roles that value coordination and delivery rather than only technical output, and you can create a free account to get matched with them.

Share this article

For job seekers

The right role, found for you.

Built for people who’d rather be good at their job than good at job hunting.

  • MatchCompare live roles against your current evidence.
  • PositionTurn proof projects into role-specific applications.
  • ImproveUse market feedback to keep the skill plan current.
Return to navigation