Crucial Accountability QA

Don’t Agree On The Facts? How to Dialogue Anyway

Dear David,

As a certified trainer, I want to be prepared to answer students’ questions. In talking about separating facts from stories, the question came up: “What if people disagree about what each person said?” For example, I say to my son: “You said you would pick up milk on the way home.” He responds with: “I never said that.” Obviously, situations at work are more high stakes than picking up milk, but I’d love some advice on what to do when the facts are disputed.

Fact or Fiction

Dear Fact or Fiction,

Great question. In most disagreements, we expect our facts to be accepted—at least if we’ve stripped them of any conclusions, judgments, or stories. The idea is that facts are facts because they can be measured, witnessed, and replicated by anyone. Facts should be an area of agreement both parties can build on. That’s why our STATE skill is all about Starting with Facts.

However, as your example shows, people don’t always agree about the facts. I’ll suggest four different situations that result in factual disputes and outline strategies for dealing with each.

1. Trust Is at Risk. The high-stakes situation you describe with your son isn’t about milk. It’s about trust. If you don’t frame the problem correctly, you’ll end up talking about the wrong facts. Fortunately, there is a tool you can use to determine which aspect of a problem to address. We call the tool CPR.

  • C stands for Content. By “content” we mean the immediate incident or presenting problem. You’d have a content issue if your main concern is the missing milk. If this was a one-time occurrence, you might accept your son’s version of the facts. Maybe you only thought he said he’d pick up the milk.
  • P stands for Pattern. This is the conversation if your concern is for the “pattern” of disputed facts. You’d have a pattern issue if your son has repeatedly denied agreements you thought you had. You can imagine mishearing once, but not multiple times. The facts would include two or three instances of these discrepancies. Beyond these facts, you’d add your opinion that these instances fit together as a pattern.
  • R stands for Relationship. This is the conversation if your real concern is the “relationship.” You’d have a relationship issue if the incident or pattern has caused you to question whether you can trust your son. Your misgivings are a fact that your son needs to hear and address.

The mistake most people make is to focus on content—the facts related to a single incident—when their real concern is with the pattern or relationship. A related mistake is to allow the other person to drag a pattern or relationship conversation back to the details of each individual incident. When a conversation begins to get sidetracked by details, I use a script such as, “I don’t want to get into the details of any single incident. Instead, I want to focus on the overall pattern and how it affects our relationship.”

2. Trust Is Low. When trust is low and you don’t trust each other’s facts, you need verification—an agreement about how facts will be substantiated.

These kinds of verification procedures are very costly. They move you from “handshake deals” to “legal contracts” and from “self policing” to “external enforcement.”

Here is an example of verification in an environment of low trust: I was facilitating a negotiation exercise with an MBA class at Stanford and two of the teams were struggling with trust. Then one of the students threw his car keys on the table and declared, “If the instructor says I’ve lied, then your team gets my car.” His ploy worked. The two teams reached an agreement, he lived up to his side of the deal—and he got to keep his car.

3. It’s Debate, Not Dialogue. Dialogue is about adding facts to a Pool of Shared Meaning. Debate is about using facts to win an argument. When people begin to value winning over finding the truth, they use facts as pawns in their game. You begin to see them cherry picking, distorting, and denying the facts.

When you see debate, try to get back to dialogue. Dialogue requires two elements: A Mutual Purpose that both parties see as more important than any particular disagreement, and Mutual Respect that is felt by all sides. If you have a Mutual Purpose, use it to reframe the debate as dialogue. If you don’t have a Mutual Purpose, seek to find or invent one. And, in any case, demonstrate Respect.

If you can’t move to dialogue, then you’re stuck in a low trust environment where verification is required. Delegate fact-finding to a neutral party, use joint fact-finding, and incur all the costs that verification entails.

4. It’s Complicated. Sometimes it’s not the facts but what the facts add up to that is under dispute. Complex situations almost always entail a more sophisticated theory. The disagreement might be over the theory, not the facts.

In Crucial Conversations, we suggest separating facts—the basic verifiable evidence—from your story. Your story is your theory about what the facts mean. With complex issues—for example, “Will this policy improve learning?” or, “Has this change in taxes produced economic growth?”—the stories/theories are often presented as facts.

Sometimes, the theory deserves to be treated as a fact, because, like the theory of gravity, it is so thoroughly validated (of course, even in this example, we don’t know all there is to know about gravity).

Other times, the theory is still unsettled or disputed. Or, it’s settled and treated as fact by experts but has not gained public acceptance.

In these situations, it’s still very helpful to separate facts from stories. For example, it’s a fact that streets in Miami Beach flood far more today than they did ten years ago. Citizens there don’t need to agree on climate change or global warming to know they have a problem on their hands.

I hope these examples help you answer the questions you get.

Best of luck,

Want to master these crucial skills? Attend one of our public training workshops in a city near you. Learn more at


David Maxfield

David Maxfield is a New York Times bestselling author, keynote speaker, and leading social scientist for organizational change. For thirty years, David has delivered engaging keynotes at prestigious venues including Stanford and Georgetown Universities. David’s work has been translated into twenty-eight languages, is available in thirty-six countries, and has generated results for three hundred of the Fortune 500.
read more

3 thoughts on “Don’t Agree On The Facts? How to Dialogue Anyway”

  1. I love the Content, Pattern, Relationship rubric.
    I see several different things happening in these examples that might be helpful.

    I agree that the ‘Milk’ discussion might be about trust – but it also might be about communication style. So, I might (in my advance plan for the conversation, since on the spur of the moment I WILL screw this up) say, “When we discussed getting milk, what did you think was the next step? What can I say next time to make sure you understand I am asking you to do a task? What do you suggest I ask you to make sure we are both in agreement about the next step?” All of these are much more likely to protect our relationship and improve our communication going forward – and, if I am actually dealing with an untrustworthy person, let me know that it isn’t me, since they won’t hold up their end, and a new tactic will be required.

    Whenever I think of facts, I am drawn to the concept of models. Every assumption we draw is based on a model of reality – sometimes just on our perception, sometimes carefully measured, but always less complex than reality itself. Usually, that means a model that has enough of the whole in it that our decisions and reactions work for us on an everyday basis. For example, everyone I know says, ‘the sun rose”, even though that isn’t true – the earth spun! But, sunrise is completely useful as a model as long as I never need to leave the ground, but counterproductive worse than useless if I need to do anything in space.

    So, finding where are our models compatible would be a more accurate way of looking at this, then are our facts better.

  2. I’m struck by this example: “For example, it’s a fact that streets in Miami Beach flood far more today than they did ten years ago. Citizens there don’t need to agree on climate change or global warming to know they have a problem on their hands.”

    True! The next question is: Do they need to agree on climate change or global warming to determine what the best solutions are? I can see that treating the theory as an open question and starting from a common pool of facts could be helpful to start a group of people on a path toward dialogue about an undeniable problem. Do you think they will ultimately end up with the same story if they are successful in addressing the problem? In Crucial Conversations, if I remember it correctly, we’re trying to come to a common understanding, so we’ll probably end up with at least a similar story . . . even though there might be some differences. For example, some people might be prepared to take action to address climate change as “insurance” against the possibility that it contributes to the flooding even if they are not as convinced as their neighbors that it certainly does. To what extent can crucial conversations work with differing or semi-differing stories after dialogue?

  3. Great observation! I used the “Miami flooding” example in the hope of spurring this kind of conversation. My experience is that the undeniable facts about flooding will create enough mutual purpose to support collaboration around short-term solutions. My hope (or pipe dream) is that this short-term collaboration will build enough trust to allow a frank discussion of the longer-term problem, the possibility/probability of climate change. As people collaborate in productive ways, they create virtuous cycles of trust, collaboration, success, and trust. Not always, but often enough to take notice.

Leave a Reply