Persuasion is, on the whole, a more difficult task than providing information.

Sure, there are exceptions. Persuading someone to carry an umbrella when the forecast says rain would usually be easier than, for example, explaining how to perform a trustworthy analysis of variance.

For the most part, though, persuasion presents the greater challenge. To inform, you have to know the subject yourself and explain it well. To persuade, you have to know the subject yourself, explain it well, and find ways to handle the other person’s evidence and logic, and find a way to soften the not very subtle implication built into most persuasion situations: I’m right and you’re wrong.

The persuadee’s ego is the persuader’s biggest barrier.

Persuasion is usually about change — about convincing someone that tomorrow shouldn’t look like yesterday, that they need to change their direction or behavior, or perhaps join an effort instead of being an opponent or bystander.

A straightforward formulation can help: Problem, solution, plan.

As William Bridges explains in his outstanding book Managing Transitions: Making the Most of Change (2003), the key step is to sell the problem. Once the other person agrees that a problem is real, the rest is … well, not necessarily easy, but easier.

You have plenty of tools at your disposal to sell the problem, some unsavory (propaganda), some more palatable. Sad to say, evidence and logic are two of the weakest.

This isn’t because people are intrinsically stupid, or because they’re willfully ignorant. It’s because those who claim people only use 10 percent of their brains are wrong. People use it all: On the average, 10 percent to think and 90 percent to rationalize.

(If you’re wondering how I know this, the answer is that as a result of careful observation, study and analysis of a statistically valid sample, I made it up. Also if you’re wondering, I define rationalize as “the evidence and logic people use to disagree with me after I’ve presented my case.”)

To slide past the egos and rationalizations of the people you’re trying to persuade, questions are your best friend.

“Here’s the evidence I’m looking at. The conclusion is inescapable,” will simply offend someone who has reached a different conclusion. Compare it to, “Here’s the evidence I’m looking at. What conclusion would you draw from it?”

You’ve just taken their ego out of the equation, because you aren’t offering a point of view different from their own. You’re open to whatever interpretation they reasonably come up with.

Not really, but it sounds that way.

If they somehow manage to reach a different conclusion than the one you privately reached, you’re free to ask them to clarify their logic, so long as you do so in an open-ended, unargumentative way: “I don’t see how you got from here to there … can you help me understand?” is fine (it’s another question); “Your logic is all wrong,” doesn’t work very well.

Questions are effective in bringing people to your solution, too, assuming your solution is a good one: “Any ideas about what we should do?” requires the other person to engage in mutual problem-solving with you, where “Here’s the right answer,” does not.

In many cases, once you understand the problem and the environment for which you’re solving it, the number of discrete solutions is pretty limited. Ask, and the other person will reach the same solution you did.

Sometimes that isn’t the case. When it isn’t … maybe their solution will be better than yours. Maybe not, or maybe it’s just too late because the decision has already been made. If it has, there’s no harm in (at last!) giving an answer instead of asking a question: “I’m part of a team that’s been looking at this. We looked at that alternative, and quite a few others. Here’s what we came up with. Do you think it will work?”

If they accept the problem as you’ve defined it and your solution really is workable, you can probably get them to buy into it.

Which brings us to the subject of today’s column (what — you thought it was how to persuade?).

More and more data centers are running low on power, and just running in another circuit isn’t always feasible. If you were faced with this situation in your company, how would you deal with someone who proposed replacing something that’s already in production and works with a different technology that calls for adding more servers to your already-stressed data center?

Me too. As you know, many industry experts still recommend using Citrix and similar virtualization technologies to move processing from the desktop to the data center. Does that seem like the best solution to you?