This happens all the time. A subject matter expert at a professional services firm tells a content director, or a writer hired to help her, that she has written a thought leadership article about whatever. All it will need, she says confidently, is a light edit.

Then you read it, and that’s simply not true. It will take an enormous amount of work to make it publishable.

The expert is smart; she is respected in her field. How could she be so wrong about her article?

In 1999, two Cornell University psychologists, David Dunning and Justin Kruger, sought to answer this question. They performed experiments, and wrote a scholarly article, “Unskilled and Unaware of It,” in which they reported that “ignorance more frequently begets confidence than does knowledge.” Our expert was confident she had written a good article because she did not know what a good one is. She was confident in her writing because she did not know what good writing is, and “the failure to recognize that one has performed poorly leaves one to assume that one has performed well.”

Ignorance, Dunning and Kruger found, far from making one humble, leads people to overestimate their performance. (Conversely, the more an individual knows about something, the more likely she will underestimate the esoteric nature of her knowledge, assuming everyone knows as much as she does. This is why it often is so hard to understand experts, and also why they need professional help to communicate their ideas to people outside their special subject matter areas.)

But back to ignorance. It cannot be total. There needs to be a “minimal threshold of knowledge” for a person to overestimate his or her abilities. For instance, if you don’t know how to play chess, you’d have to be nuts to think you could beat a grandmaster, and Dunning and Kruger were not trying to explain why crazy people think crazily. They were attempting to figure out why smart people think they know things they don’t and believe they can do things they can’t. When it comes to writing, of course, just about everybody reaches that “minimal threshold,” and experts at professional services firms easily surpass it. They’ve all gone to college; they’ve all written reports and papers; they’ve gotten good grades, and their professors have praised them.

This has caused many of them to develop unrealistic confidence in their writing skills. (Writing for a professor is one thing; writing for thousands of people who don’t have to read you is quite another.) Confidence, Dunning and Kruger found, often is inversely proportional to performance. In many cases, the more confident a person is about something, the more likely he will be bad at it.

So what can writers, editors, content directors, and CMOs do when confronted with an expert exhibiting the Dunning-Kruger Effect? How can they work with this person to improve a piece of thought leadership?

The first thing one should understand is that one is dealing with someone experiencing a cognitive failure, and it is best to be gentle with people with cognitive difficulties. Being confrontational, telling them their ideas are poor, or their presentation inadequate, will not work. In fact, according to Dunning and Kruger, negative feedback precludes learning. In the coils of the Dunning-Kruger Effect, people are trapped by knowing what they know (even though what they know is erroneous), and who the heck are you to tell them differently?

Instead of being blunt, it probably is best to suggest that although their article is excellent, it could be improved. (After all, what cannot be improved?) And to help the writer improve it, the expert might first explain more simply what she means and, perhaps, provide an example (or two, or three) from her experience to help you understand. After that, all it will need will be a light edit.

In this transaction, the writer or editor does not try to break down the barrier erected by the expert’s confidence; one recognizes the barrier and works with it.

Will this always work? Probably not. In fact, the worse the article, the surer the expert will be that it is excellent, and if you have a problem with it, that’s your problem, not hers.

What does one do then?

I have no idea. This Dunning-Kruger Effect is powerful stuff. So powerful, in fact, that, like Dunning and Kruger themselves, I worry that I could be displaying overconfidence in my ability to offer advice, and could be mistaken about how to handle experts exhibiting the Dunning-Kruger Effect.

I don’t think so but, of course, how can I be sure?