Recent talk – Disagreeing with nice people

I did a short presentation at work last week and I thought it would be good to reproduce it here. The talk was titled “Disagreeing with nice people”.

Why nice people?

Disagreeing is, well, disagreeable. But disagreeing with “meanies and fools” sounds more satisfying than disagreeing with nice people who might be right.

You don’t want to make them feel bad if you prove them wrong; and you don’t want to appear to be a “meanie or fool” if you turn out to be the one who is wrong.

At the same time, real collaboration and real teamwork depend very strongly on the ability to challenge thinking, find better solutions to problems and even to understand each others thinking. All of those things involve disagreeing with nice people, making it a pretty important skill to have.

Typical disagreements

Let’s say someone wants to put code into production, but you think the code is going to result in a bug or production outage. I guess you should say something.

Similarly, if someone wants to drag the whole team into a big meeting when you are flat out and want to finish something before going home, you might again disagree.

In many situations we want to disagree, and we want to do it well. In my talk, I offered 4 approaches that should work well in most situations. In fact the first two will cover most situations but the final two are useful when you are in a complex discussion, where there may be multiple points of disagreement before you come to a common understanding.

Just be curious

The first approach is to defer the decision to disagree.

This may sound odd at first, but rather than share your view, you simply remain curious and seek to learn more about the other person’s viewpoint.

Asking clarifying questions and trying to fully understand and help them form a better view makes a big difference.

Firstly, you will often find that there is common ground, or even that they were not intending to state what you thought they were saying. This moves potentially wasteful disagreement and builds trust for next time you disagree.

More importantly though, it means that when you disagree, you are disagreeing the the right, thoughtful view rather than an ill formed or misunderstood thought that was still forming as it was uttered.

Disagreement, debate and negotiation that starts on a solid foundation is far more likely to lead to new insights or well aligned goals and actions.

Taking time to understand people and helping them formulate and share their thinking also builds trust, making you “luckier” in the future when disagreements or mistakes occur and you need to maintain a good relationship.

The only trick is that you need to genuinely want to learn what they think. If you fake being curious it backfires. So you need to be asking questions to understand their thinking, NOT asking leading questions designed to help them realise that you were right all along.

Curiousity makes you lucky, whatever happened to the curious cat, while manipulating conversations to outsmart your opponents always leads to trouble.

Disagree by using a simple formula

The second approach is to let them know you have a different opinion, before fully exploring theirs.

This is a good move when you think you (are more than likely) clear that you disagree with someone, and can be more honest that asking question for 10 minutes before saying that you disagreed from the start.

However, it needs to be done properly and so I propose using an fairly mechanical sounding approach, consisting of 5 distinct steps.

This is to help you avoid making some classic mistakes:

• Apologising and waffling to avoid having to disagree, when in fact you are in the process of disagreeing
• Laying out heaps of evidence and running through multiple “winning arguments” designed to overwhelm an opponent, rather than reach a shared understanding with a friend (or peer).

The 5 steps

Firstly, you confirm what you think they said, so that:

• If you are close to the mark, they know what you are disagreeing with; and
• If you completely misunderstood, you save quite a bit of potential angst.

For example, consider this short conversation:

“Just to confirm, you said you want to kill the barista?

Me confirming what I think you said

What? No. I said I would kill for a burrito – I am really hungry.

You – clarifying your comment and helping to avoid a potentially unfavourable outcome

Once you know what you are disagreeing with, you flag that you have a different perspective before you jump into your counter-argument. Note that you do not say they are wrong, you don’t explain how you like their thinking and you do not apologise for your view – you just say that you have a different view.

Flagging that you have a different perspective helps them anticipate the point you are about to make and to compare it to their own view.

Having flagged that you have a different view, you now state your view. But don’t talk about it a lot – literally state your view in 10 seconds or less.

“I don’t think we should kill the barista”, or “I don’t think we should deploy this story into production yet, I think we need to redo our testing in the staging environment first.”

Now they know that you have a different perspective, but not how you came to form your opinion. Sometimes they will interrupt you to explain their opinion again or explain how yours is wrong because of some ancient prophesy or something. That is a shame because it is not helping to come to a shared agreement. If that happens, listen for a bit and then repeat your view if you can.

Hopefully, having spoken for only 10 seconds, you will be allowed to talk for another minute or so, which is actually all the time you want to take.

Having stated your opinion in under 10 seconds, you now explain how you came to form that opinion. But spend no more than 45 seconds doing so.

Your aim is not to win a legal argument by convincing a jury or your peers, it is to put your view out their for consideration (and refutation). If you start making a lot of points here, your nice friend might get lost in the conversation, or they might wait for your weakest seeming point and jump on that.

So you have spoken for only 1 minute. You will probably not have persuaded them of your point, but they have enough information to understand it at a high level. So the next step is to check in with them. Say “I am curious to understand how you came to your view.” or ask “What led you to your view?”.

There is a trick here too. Don’t ask them to agree with you, or to disagree. Instead ask them how they came to their conclusion. You will find that discussing the thinking behind both your views is a far more agreeable form of disagreeing than just pushing your conclusions at each other.

Putting it together

This is what it might look like:

• So you think the code you wrote is ready to release into production because you are satisfied that it is tested well enough already
• I have a different view
• I think we still need to test it in our staging environment (<10 seconds to say what I think).
• My reasoning is that our development environment is quite a bit different to production and we have missed a few bugs recently when we only tested in a development environment. I worry that could be the case this time too (<45 seconds)
• I am interested in how you came to your view though, what led you to your conclusion?

Now we can argue, based on what we both actually think and, more importantly, we can explore our reasoning safely rather than just repeating our opinions

Picking the right fight

Most of the time, the two approaches listed above will help you have healthy disagreements rather than uncomfortable conversations.

But sometimes the difference of opinion is more nuanced and you need to clarify what you are actually disagreeing with.

Again, this might sound strange but, in my opinion at least, people often forget about what they are actually disagreeing on. The disagreement continues but there is no conclusion because there is no common starting point.

Vague, roaming disagreements make a great basis for conflict in TV shows but can undermine collaboration in real life. If you ask them then the two people caught up in a debate actually tell you that they are disagreeing about different things.

If you don’t even agree on what you disagree on, a simple difference of opinion can turn into a long circular discussion. This makes for good drama in a tv show, but is not helpful when collaborating with nice people.

If you don’t believe it happens, you can try a simple experiment. When you hear a disagreement that is going for a while (or you are in one), have both people write down in a single sentence what they are actually disagreeing on. Then have them reveal their sentence to each other and see if it is the same.

The longer the disagreement has been going on for, the more likely it is that both sentences are significantly different.

So a better approach is to pause and call out specifically where you think you are in agreement and what, specifically, you are disagreeing about (or at least what you think the disagreement is about).:

I agree that you have done all the testing you can do on your own, but I don’t think that you have access to the right test environment to cover all scenarios.

Me again – clarifying what we agree on and what I think is a point of difference

There may be many things you agree on and a few things you disagree on, but generally when we agree and disagree on different aspects of a complex argument, there are only 3 different categories of alignment:

• The facts (I don’t think you have tested your code)
• The conclusion (I agree you have tested your code but I don’t think that means we can put it into production)
• The action to take based on that conclusion (I agree that you have tested your code and that it is ready to go into production, but I think we should go out to dinner and then go home. I would rather go live on Monday).

Hopefully you can see that you could have 3 distinctly different conversations, depending on whether you want to debate the facts, the conclusion or the action. Alternatively, you could have quite a long, meandering discussion if you are both unclear what points to make and where there is already agreement.

But wait – there is more. If I disagree with the facts (or the conclusion or the actions to take as a result) then there are essentially 3 reasons why I might disagree:

• The facts are wrong
• The facts might be right but might not actually be relevant to the decision we are debating.
• The facts might be right and even relevant, but there is something more important that we need to consider (you may have done all the testing is possible, but I know there is an ancient curse that means we cannot release code to production until the curse is lifted or there will be a plague of bugs).

For major, complex discussions, picking which fight to have will save time and also give the nice person you are disagreeing with a lot more clarity about what, specifically, you want to discuss.

Booking a feedback session

So we have 3 different approaches that we can take. The is one more, which kind of builds on the others.

If things are complex enough, it might be worth setting time aside to examine the issue/agreement more carefully.

If we do that we can also agree on how we will discuss the topic and we will have time to give it our attention.

You might already do this in code reviews, product pitching sessions or retrospectives. What might be different here is that, again, I break the approach into specific steps, this time 6 instead of 5.

The steps

The first step is to agree to spend time going through the topic of discussion, agree the broad topic, book time and make sure you have the right people (and don’t have people who have no interest).

Once you get together, you start with one person sharing their idea, opinion or work.

Next, the others ask clarifying questions to make sure they understand what is being presented or proposed – no disagreement before understanding.

Once you understand sufficiently, you can disagree. You can ask probing questions (such as why a specific choice was made over another or what evidence there is to support a view) and you can make suggestions, share your reasoning and put your views forward.

This should get you close to an agreement, even if you agree that you both understand the issue at hand, but both have a different opinion that will not shift. In other words you agree to disagree.

Often though, it is also worth clarifying how you will “agree”, adding one more step to the process. You might agree that you will leave the decision to me, after you give feedback. Or we might agree that I will pitch the idea to you and that you will then either agree to, or veto, my proposal.

Finally, you start to agree and then you state that you have in fact agreed. This forces you to reach a point of clarity, rather than “sort of” resolving the discussion.

Conclusion

While some disagreements will still be uncomfortable, I believe that applying these four approaches will convert most uncomfortable conversations into engaging debates where you and the nice people you want to collaborate with all come away satisfied and comfortable to disagree again next time.