NEW YORK, NY
That data, from Content Marketing Institute's 2026 B2B research, almost floored me.
The actual experts are nowhere near the keyboard, or the conversation, or the process at all, meaning that, in the vast majority of organizations publishing "thought leadership," the person with the expertise has been fully removed from the thing being published in their name.
There's a version of this conversation that turns into a referendum on ghostwriting itself: is it authentic, is it allowed, should executives be embarrassed. I don't think that's a useful question, and I don't think it's even an interesting one. Executives have always had help in the form of speechwriters, communications teams, and editors — nobody expects a CEO to have typed every word of a shareholder letter alone at a keyboard at midnight. Getting help isn't the problem, but the absence of their point of view in the work really is.
You generally can't tell the difference at first glance. Same byline, same headshot, same LinkedIn format. But the substance of the work makes it glaringly obvious. In the first case, someone sat with the executive, pulled out the actual opinion — the thing they said in a client meeting that made the room go quiet, the belief they keep repeating at dinner because it genuinely bothers them — and then did the work of shaping that into something publishable. In the second case, nobody pulled anything out of anyone, because nobody asked. A generic industry take got written, a name got attached, and it went out.
At minimum, the expert needs to be in the loop: reviewing, correcting, pushing back on anything that doesn't sound like something they'd actually say or believe in. Better than that, they're driving the point of view from the start, even if someone else is doing the drafting, the polish, the formatting, the posting schedule. The craft — the actual expertise a good thought leadership partner brings — was never in generating opinions on someone else's behalf. It's in drawing the real opinion out of somebody who has it but hasn't had the time, the framework, or the outside eye to see what's actually distinctive in what they think. Then packaging that into something that reads clearly.
Then distributing it somewhere it'll actually be seen. Then knowing when to publish it — because the same idea landing during a live industry debate lands completely differently than the same idea six months later when the conversation's moved on.
That's four distinct skills — excavation, packaging, distribution, timing — and an executive can reasonably hand off three of them. The one that can't be handed off is the first one, because if nobody draws out the actual point of view, there's nothing for the other three to work with. You get content that's fluent and says nothing anyone will remember, not because the writing was bad, but because there was no idea underneath it to begin with.
The CMI stat is actually measuring a proximity problem — how far the actual expert has drifted from the thing being published under their name. Some of that 96% of executives who are publishing thought leadership are deeply present in their own process: they're reviewing every draft, pushing back on lines that don't sound like them, treating the writer as a craftsperson translating their thinking rather than a vendor generating content for them. Others have handed the whole thing off and stopped paying attention, and it shows, eventually, to anyone paying attention on the other end.
So here's the question I'd actually want an executive to ask themselves, instead of "should I use a ghostwriter": when's the last time you said something in your content that you hadn't already told someone in an actual conversation? If the answer is never, the issue was never who's typing. It's that nobody's asked you what you think in long enough that you've forgotten you're allowed to have an opinion before the writer does.
The executives who'll matter most over the next decade will be the ones who stayed close enough to their own thinking that whoever helped them say it well had something real to work with.
