Working In Uncertainty

How to talk openly about uncertainty at work


We're faced with uncertainty all the time, and the way we handle it makes a big difference to our lives. Just managing one's diary is an exercise in advanced decision making under uncertainty. Almost every decision we take has some element of doubt.

Psychologists have found that we tend to have an overly narrow view of what might happen in the future. Does this improve when we have the advantage of other people around us to provide different views?

No. Unfortunately, when we get together with other people, particularly colleagues at work, our vision narrows even further. For a number of reasons we suppress our uncertainty. Consequently, we often fail to plan for problems that might arise, and fail to anticipate and cultivate future opportunities.

Weighty books and papers have been written about "risk management", "scenario planning", "decisions under uncertainty" and so on but this web page is for busy people. It suggests simple conversational techniques that can be applied immediately, without waiting for consultants, training, or a company-wide programme. Beginning today, you can influence your boss, your peers, and your subordinates, and start to put an end to senseless foul-ups and missed opportunities.

This web page explores why we often suppress uncertainty at work, and what we can say in everyday conversations and meetings to improve our awareness of the future and ability to achieve success.

Why we suppress uncertainty at work

Most of us face many situations at work where we are pressured to keep quiet about our uncertainty, and not talk about unexpectedly good or bad things that might happen in future. For example, a customer who wants a firm commitment on something that, realistically, is very hard to guarantee, or a request for a best estimate of some number or date that is difficult to work out exactly. Often, the pressure is very subtle. Here are some things your boss may have said to you, that illustrate the problems:

What was saidWhy it probably caused uncertainty suppression

"What do you mean you 'think' you can do it?"

"You're going to have to sound a lot more confident if we're to win this contract."

"To get this funding we need to put forward a convincing business case."

Pressure to say things in a way that contains no doubt.

"Your performance goals for this year should be SMART. Do you know what that means? Specific, Measurable, ..."

Specific goals a full year ahead? Most people don't have enough control or information to accurately predict what will happen a month from now, let alone a year ahead.

"Your last forecast shows that you're going to miss your target. That's not going to be acceptable. Do another forecast."

"We're falling behind budget."

Control by targets produces all sorts of pressures to suppress uncertainty. Because hitting the target is what matters there is no point in considering other outcomes. Conversations tend to dwell on past variances and commitments to specific future numbers are sought.

"So, that's my plan. I want you to get started with the first actions and update me with progress at the end of the month. Any questions?"

Your boss has announced his plan. How would it sound if you responded by listing reasons why it might not work?

"So you think this could lead to an improvement as large as that? Well, let's take that as our stretch target shall we."

Teaches you not to talk about how well something could turn out.

"I need a forecast for this. What's the number I should use?"

One number! It really should be a distribution showing how likely various outcomes are, but you're asked for a single number, and encouraged to be "decisive".

"Everything under control?"

Dare you say "No"?

"I don't think we should be talking about that at this stage. We need to manage expectations."

In other words, if we talk about things that might happen, won't we start people thinking it will happen or that we're promising it will happen?

"They're very busy so I don't think we should bother them with that question. We'll just make a reasonable assumption and get started."

They're not as busy as they will be once they discover we've done the wrong thing. Your boss doesn't want to look uncertain and now passes the suppression down to you.

"What I want is for you to deliver what was originally agreed, by the deadline, and within budget. I want to know if you're committed to doing that."

Focus on original goals (probably out-dated) tends to provoke uncertainty suppression.

"I'm looking for teamwork. Let's stay focused on our goals. I don't want any defeatist talk about problems."

Unfortunately, that also means not talking about potential problems.

Pressures to suppress uncertainty are much more frequent than pressures to be open about uncertainty so that it can be managed properly.

There are exceptions. Sometimes meetings are held specifically to anticipate potential problems and opportunities, creating an atmosphere in which uncertainty can come out into the open.

However, in most conversations in most meetings, suppression is the rule so if you want to avoid senseless foul-ups and missed opportunities you need to be able to talk about uncertainty in ways that people don't perceive as a career limiting stance.

What to say when you are the boss

If you're the senior person in a conversation at work you have the power to do a lot of good. Just asking people to talk about uncertainty and risk will help enormously. However, even with this encouragement people will still have difficulty seeing the full range of future outcomes, so you need to push and be persistent. Here are some conversational techniques you can use to encourage people to be open about risk and uncertainty and respond appropriately.

  • "Can you be specific?": People often present evidence as if it is more conclusive than it really is. They may have had two complaints from customers about something but raise the matter by saying "Customers are complaining about ..." Obviously this over-generalisation is dangerous so you need to push for specifics. People will also give an "example" to support their conclusion when this "example" is actually the only case they can think of. However, it is wrong to dismiss evidence on the grounds that it is inconclusive. It is still evidence and if it's not conclusive maybe other evidence or reasoning can be found to resolve the uncertainty or the issue should be monitored.

  • "What areas of uncertainty are there?": A direct question and very effective. Get people to write things down so they can think about them more.

  • "What could happen?": Here is another direct question, but this time more detailed and specific than a question about areas of uncertainty, so ask for areas of uncertainty first.

  • "What could that lead to? What could cause that?": Exploring chains of cause and effect is very powerful because issues considered individually tend to seem smaller and more predictable than when their connections are considered.

  • "Can you give me an estimated range?": Instead of asking for an estimate, ask for a range. There are many ways to do this. For example, you can ask for a level the person thinks the most likely, or the point at which they are 50% sure the result will be less, then ask for the level at which they are 90% sure the result will be less, and the level at which they are 90% sure the level will be more. The ranges you get from this tend to be too narrow, so a better way is often to think of a range around the best estimate and ask how likely people think it is that the result will be above that range, and how likely it is that the result will be below that range. Many people do not like making such estimates but the alternative is to pick one number. Are they positive that is the number it will be? Of course not. So which would they rather give?

  • "Suppose things turn out differently; why might that be?": A lot of the psychological research on getting people to state probabilities for things has tried to find ways to encourage people to see the true extent of their ignorance. One good technique for doing this is to ask for an initial estimate of the likely spread then say "Let's suppose time passes and things work out better/worse than you currently think possible. If that happened, what might the reasons have been?" Once they've answered that question you ask for another estimate of the range of outcomes and, typically, it is wider.

  • "Can we think about some of the ways this could evolve over time?": The greatest benefit of getting people to think through the paths to different future scenarios is that it helps them recognise those scenarios quickly when they actually occur. It is also useful as a way of getting people to stretch their imaginations and think about plans that might work well across more than one future scenario. This technique is not so good for predicting the future.

  • "How can we find out more, or monitor this?": Finding out more is needed for most areas of uncertainty. But beware of the occasional areas where no amount of research, enquiries, and monitoring can help.

  • "Let's just run through some of the things that have been issues in the past.": Sometimes it seems as if people are determined not to see problems that might occur, even when past experience is littered with problems that did occur. "Shouldn't be a problem" they keep saying. Alarm bells should be ringing for you. This is the kind of macho confidence that leads to total chaos and frustration later on. If you have experience to bring to bear you can do so by suggesting potential problems and areas of uncertainty, and asking for each to be considered. Referring to past experience can give extra weight to your case for caution.

  • "What if things turned out better than we expect? What if they turn out worse?": This question encourages people to think about how they would react if events or results proved more or less favourable than initially expected. Sometimes, the action is easier to take, or only possible, if some preparation has been done beforehand. As a follow up, ask if there's anything that should be done in advance to prepare.

  • "How could we make that more/less likely to happen?": Once you've considered how to mitigate bad consequences of an event and exploit good consequences the next step is to think about how to make events that are, overall, favourable more likely, and others less likely.

What to say when you are not the boss

If you're not the boss in a conversation and nobody else is encouraging uncertainty it may be possible for you to take the initiative. With the right words you can often safely encourage people to be more open. If the atmosphere is very good you can ask the same questions as suggested for a boss (see above). However, normally you will need to be more diplomatic.

  • "I think that's a good idea. Can we talk about the things we need to do to make sure it works?": This is a way to open up a discussion of what might happen and how to manage those possibilities.

  • "And won't we then be positioned well if things go well?": This introduces the topic of potential future opportunities, but you may need to give assurances that you are not suggesting a new goal or that people should be told to expect a higher outcome.

  • "Realistically, there are some factors that make it very difficult to make a precise, reliable forecast but let me be as helpful as I can.": This is the prelude to outlining some areas of uncertainty and giving a distribution for your estimate rather than a single-point.

  • "If we're going to do that really well it would be easier if we knew more about ...": This introduces a suggestion for finding out more, or perhaps monitoring.

  • "We should find it easier to work out the details as we learn more.": This introduces a suggestion for monitoring relevant sources of information over time.

The most difficult situation is when you are the subordinate in a conversation and your boss is pressuring you to suppress uncertainty. How can you survive?

  • Say nothing: Often the simplest approach is to avoid discussion of uncertainty your boss would find unacceptable, but make sure you think about and manage uncertainty yourself and with your subordinates, if you have any. Managing uncertainty properly is the best way to get good results, even if it doesn't have the support of your boss.

  • Don't waste opportunities to be open about uncertainty: For example, if your organisation's annual performance appraisal system allows you to update your personal objectives half way through the year, make sure you do.

  • Reflect the pressure back, respectfully: If it is difficult to give a precise estimate, or to commit to achieving some specific objective you may be able to get a little sympathy by handing part of the problem back to your boss. You might say "To help me with this estimate, I'd like to ask you for your views on some of the key parameters" or "Can you suggest any steps we could take to guard against potential problems along the way?"

  • Balance positive and negative: If your boss insists on positive thinking then make sure you point out potential opportunities as much as potential problems. Just trying to counter-balance your boss's optimism could lead to confrontation and get your boss arguing for optimistic expectations, which just reinforces dangerous optimism.

  • Bring in specifics: Uncertainty suppression thrives on abstract thought about actions, but struggles against the complications of real circumstances. If you can turn a conversation towards real circumstances then complexity and uncertainty tend to push through.

Putting this into practice

The best time to start dealing properly with risk and uncertainty is right now. We are all the boss sometimes, even if it's only of ourselves. The techniques described above are no more than simple questions, but applied whenever appropriate (which is almost every day) they can quickly make a difference. Soon more things will be going right and fewer things will be going wrong.

Start by just becoming aware of pressure to suppress uncertainty and watch how people react to it. This is not always easy because we are so accustomed to ignoring alternative futures. After a meeting, stop and think about what was uncertain and begin to jot down specifics and potential monitoring and mitigating/exploiting actions. Then think back to the meeting to see how much of that actually came out in the meeting. Not much, typically.

As soon as you feel you are becoming more attuned to uncertainty suppression and more confident in your own ability to think about uncertain futures, try out some of the questions suggested above. Don't expected to stimulate a perfect analysis; just experience how people react and find how far you can go with different people.

The first result of this is that some conversations will take longer, and you will have more to think about. Later that investment of time and thought will be repaid as circumstances evolve and you are able to cope with changes more easily.

With growing experience and confidence you will be able to take on more difficult challenges, like more senior managers, and builders.

To get you started, here are some imaginary conversations showing the difference between uncertainty suppression and good conversations in the face of everyday uncertainty.

Steve from IT Support

Imagine you are due to move to a different office. Imagine it's not a big move - just a few people - and a weekend has been chosen for the IT support people to move all the computers and get them connected and working. You have a phone call from "Steve" from IT Support who says he is making a courtesy call to let you know what's happening. He sounds very confident and according to him it's all very simple and "shouldn't be a problem". You know from past experience that nothing with computers is ever that simple, but his breezy confidence persuades you to say nothing. The conversation ends like this:

STEVE: [summarising what he has told you] So, we'll do the work over the weekend and when you come in on Monday everything will be ready for you.

YOU: OK, fine. Is there anything I need to do?

STEVE: No. Nothing at all.

YOU: Thanks. Bye then.

STEVE: Cheers.

Monday morning arrives and when you come into work there is a note stuck to your computer in the new office saying "Apologies - system not working yet. Steve." You call Steve, who assures you it's not a complicated problem and "should be sorted out within the hour". By early afternoon nothing has happened except for a couple of messages being left for Steve and a visit by someone from "Nets R Us" who checked some cabling and went away again. You and your colleagues are unable to work and there is nothing you can do.

Now let's see one way to deal with Steve's breezy confidence during that first phone call:

STEVE: [summarising what he has told you] So, we'll do the work over the weekend and when you come in on Monday everything will be ready for you.

YOU: OK, fine. Is there anything I need to do?

STEVE: No. Nothing at all.

YOU: We can't work without the computers. What should we do if it's not done by Monday morning?

STEVE: It will be.

YOU: But just supposing. What if everyone in your team falls ill, or someone trips and drops a computer, or some piece of kit fails, or the power sockets don't work? There are things that you can't control aren't there.

STEVE: Er. Yes. Well, OK, I think we've got some workstations on the 3rd floor that are networked but not in use at the moment. You might be able to use them.

YOU: OK, can you check that they are available and working please? I also want to be sure that backups will work.

STEVE: I suppose so, yes all right.

YOU: I know it's unlikely that there will be a problem, and I'm sure you've got it well planned, but imagine we did get to Monday morning and our computers weren't working, why might that be? For example, have there ever been problems in the past?

STEVE: No. Well, nothing that should be a problem now.

YOU: [silent]

STEVE: We did have an issue with some wiring last time. That's not down to me. It's Nets R Us. They didn't have the right connectors.

YOU: Don't they check out a job before they do it?

STEVE: Not normally. It's really quite simple stuff.

YOU: Is there anything else that can potentially be unusual or create unexpected problems in a job like this?

STEVE: [silent]

YOU: Can you get them to do some checks before the weekend please, or do it yourselves, to reduce the likelihood of any nasty discoveries on the Sunday?

STEVE: [sigh] OK.

YOU: Thanks. Bye then.

STEVE: Cheers.

Transformation project proposal

Imagine you are a senior person in a large company and you are in a meeting to discuss a proposal to introduce a new way of working. Consultants have been involved so the proposal is called "Transforming Performance" and has a "vision", a "road map", and over a hundred PowerPoint slides.

A bullish presentation is given promising all sorts of improvements. Imagine that, while you think it's generally a good idea you want to support, you think the benefits are being exaggerated to try to win support.

First, here's the wrong way to talk this through:

PRESENTER: [winding up] ... So, in summary, this change fits the strategy and will generate savings of 13.1m a year. Downside risks are minimal and there's a great opportunity to take it further and create even more value.

YOU: Thank you very much for that interesting proposal, and well done to everyone who has worked on it. One thing though. I'll be frank with you. I'm not convinced the benefits will be quite as high as you say. Still, I'm inclined to go ahead anyway. But, what I want from you are some very clear measures and criteria for success established at the outset, so we can measure whether the project has been a success or not. OK?

PRESENTER: Absolutely. Benefits management. Yes. That's part of the proposal.

YOU: Well, perhaps you can call my secretary and get a slot in my diary next week to agree some precise targets.

What could possibly be wrong with that? Time passes, and the project goes ahead. After 6 months quite a number of the original assumptions have proved incorrect. The plan to introduce changes to all units in one go is in trouble because some units have genuine problems with the changes. Consequently, everything is stalled. The benefits envisaged look increasingly unrealistic although some new advantages have been found. It seemed like a good idea at the time, and it even seemed that the project team had considered the 'risks' carefully. Sadly, it wasn't enough. All the commitments made have been broken and there's nothing anyone can do.

Here's how it could all have been different:

PRESENTER: [winding up] . . . So, in summary, this change fits the strategy and will generate savings of 13.1m a year. Downside risks are minimal and there's a great opportunity to take it further and create even more value.

YOU: Thank you very much for that interesting proposal, and well done to everyone who has worked on it. Can we just go back to your analysis of the risks and opportunities. You mentioned it but didn't say much.

PRESENTER: [presenter talks about it a little and it seems rather superficial and biased] ... so that's why we feel that the upside is good and the downside risks are pretty minimal.

YOU: [picking one key point] When you assessed the risk of problems fitting the model to all units as being "Low", what was the basis of that assessment?

PRESENTER: err . . . experience - well it didn't seem likely to be an issue.

YOU: What I would really like is a realistic exploration and description of the uncertainty remaining in this idea. I like the idea, and on balance I think we should probably do something like it, but I want to make sure we do it efficiently and sensibly, knowing what we are getting ourselves into. Uncertainty at this stage is inevitable but I want to feel confident that you and the team have thought about it and the plan reflects that. For example, what about a more phased approach? What would we do if some units could not implement the change? I also want a realistic assessment of the likely benefits. I don't mind you not being certain, but I don't want to be asked to make decisions on the basis of best guesses. For example, I want a probability distribution for the estimate, not just a best guess.


A bad plan

One of the most awkward situations is being asked to carry out a plan your boss has decided on when that plan has not been thought through very well, or at least it looks that way. You don't want to be involved in something that goes wrong, but you don't want to argue with your boss, especially if it turns out he/she was right.

First, let's see what happens when uncertainties stay under cover:

BOSS: I've been thinking about your role in this department and I think it's time you had more responsibility. I want you to take on a more proactive role in product marketing. You should be liaising directly with delivery units. I also want you to take over overall supervision of marketing support.

YOU: I see. That's quite a change. Don't we already have three people doing that? I mean, it sounds like I'm being asked to do the work of three people.

BOSS: Well ... this is a step up. Don't you want more responsibility?

YOU: Sure, but this looks like an unreasonable amount of work. Why don't you think about it a bit more and we'll talk again?

BOSS: Hmmm.

This was a tricky one. You could have kept quiet and risked being lumbered with an impossible job, but the shock and fear gave you unusual bravery as you dived for safety. But was that the right thing to do? Would saying nothing but "Yes Sir" have been any better?

Here's another version of that conversation in which you bring the uncertainty into the open:

BOSS: I've been thinking about your role in this department and I think it's time you had more responsibility. I want you to take on a more proactive role in product marketing. You should be liaising directly with delivery units. I also want you to take over overall supervision of marketing support.

YOU: I see. That sounds like quite a change, and of course we've never had someone in that role before. Can you explain in a bit more detail how you think that might work?

BOSS: [gives a bit more but still rather vague] ... so I think this is a step up for you. Do you want the responsibility?

YOU: I'd be very happy to take more responsibility and make a bigger contribution. Still, I think we need to be aware of some areas of uncertainty around this. It's the first time we've had someone in this role, and in some ways it almost looks like a role that replaces three people we have now. We probably should give it a bit more thought and also be prepared to make adjustments as we go forward. Do you agree?

BOSS: [apparently determined to ignore uncertainty] ... absolutely. I thought we could start by talking about your personal objectives for the role and agree some SMART objectives.

YOU: [trying to open up the conversation towards circumstances] Why don't you start by telling me why you think this is a good time to create the role you mentioned?

BOSS: I think there are three reasons. Firstly, we're having to do more marketing and less face-to-face selling in this business so marketing is more important. Secondly, you've really developed quickly and I'm keen to get you involved at a higher level and with more important projects. The third main reason is that I don't think it's working well to have you insulated from the delivery units by our product managers.

YOU: [reflecting some uncertainty back] How much scope for improvement do you think there is?

BOSS: A lot, especially with marketing events.

YOU: Are you ready to express that in measurable terms? The improvement I mean.

BOSS: Twice as good.

YOU: Exactly twice as good? In what sense?

BOSS: Well perhaps you'd like to give that some thought and suggest something?

YOU: Alright, so you think a big improvement is possible. I'll have a look at what we've been doing and get back to you. Then we can go through some ideas and try to judge their likely effects.

This is not the end of the conversation but at least you have secured a small breakthrough. Your boss accepts that not everything is known and it is not time to make any precise commitments or forecasts. You have taken enough control to go away and start managing uncertainty around the new role in your own way and when you report back you will be able to bring out the uncertain influence of circumstances.


There are limits to what can be done in everyday conversations to bring uncertainty into the open and deal with it. Some people are determined to ignore reality, and some management systems are simply dysfunctional. Still, there is a lot you can do as the above examples show. Even under pressure you can often improve your position with some diplomacy and the right questions.

Made in England


Words © 2003 Matthew Leitch. First published 14 March 2003 (updated 17 June 2011).