Anyone can take your coverage. Nobody can take your judgment.

A crypto news publication, operating in a category where every outlet covered the same handful of events within hours of each other. Prices moved, a protocol shipped, a regulator said something, and fifteen sites filed on it the same afternoon.

Its reporting got lifted constantly. Other outlets took the work, the angles, sometimes the framing more or less intact, and ran it under their own bylines. Occasionally the lift was close enough that a sentence survived the trip.

The internal reaction was the obvious one: irritation, then a fair amount of energy spent tracking who had taken what. That kind of tracking feels like defending something valuable. What it actually produces is a list.

None of that effort changed the outcome, and in hindsight the reason is straightforward. The thing being taken wasn't the asset, so guarding it harder was never going to work.

If information is the product, the outcome is easy to forecast. Readership should fall every time the same information becomes available free somewhere else. In that category it became free within a day, most days, and often within an hour.

The fall didn't happen. Readership held through years of it, and not because the lifting was rare or slow. It was routine, it was quick, and anyone who went looking could find it inside an afternoon.

That result is easy to walk straight past, and it's the most important number in the whole case. When a publication's audience survives its own information being commoditised over and over, the information was never what the audience was arriving for. Something else was doing the work, and nobody in the newsroom had named it yet.

The pressure that did work came from a completely different direction. Over the same period a handful of rival newsletters grew quickly, and none of them were scraping anybody.

They weren't better reported. Several were considerably worse, thinner on sourcing and slower to the news. What they had was a position: a stated view on what the events meant, published under a name whose opinion readers had learned to want on a Tuesday morning.

So the case produces two findings that arrive from opposite directions and agree completely. Coverage was copyable, which means it was never defensible. Judgment wasn't copyable, because it comes out of one particular set of experiences and preferences, and there's nothing sitting on the page for a competitor to take.

If judgment is the asset, the next question is where it comes from, and the default answer is the wrong one.

Almost everyone inside a category reads the same things. The same newsletters, the same dozen accounts, the same three podcasts, the same two analysts everybody quotes. Inputs that match a competitor's inputs produce conclusions that match a competitor's conclusions, published slightly later.

That's the mechanism behind an entire category saying one thing in the same month, each outlet privately convinced it got there first. Reading a niche closely keeps a writer current, which is worth something real. It won't make them worth reading.

There's a second cost, and it compounds. Arriving late with a conclusion everyone else already published doesn't read as agreement. It reads as derivative, even when the writer got there honestly and on their own, which is the galling version of the problem.

The fix is inputs nobody nearby is using. Art history, architecture, old advertising, fashion, museums, other countries, other decades. Anywhere the thinking is good and none of the competition is looking.

A working example from our own tooling: an internal tracker for stretching and jiu-jitsu injuries, built as a Gundam-style cockpit HUD. Holographic muscle lighting, a scanner sweep, a status ring around the whole frame. Nothing in the fitness app category looks remotely like it, which is the point, and it's also why every fitness app looks like every other fitness app.

Carrying a structure across from another medium is where freshness actually comes from. Borrowed from inside a field, an idea is a copy and everyone can see the seams. Brought in from outside, the same idea reads as original, because locally it is.

Uncorrelated inputs only pay off if the reaction can be named. "This is good" is a reaction, not a judgment. Anyone can produce one, and it teaches nothing that transfers to the next piece of work.

So the question gets pushed until the answer is specific. What exactly is doing the work here? The composition, the pacing, the texture, the restraint, the level of detail, the thing it deliberately refuses to do? Run that on admired work and, far more usefully, on work that's irritating. The dislikes are sharper and almost everybody skips them.

The drill is worth doing in writing, because a named feature can be reused and a vague good feeling can't. Two lines on why a piece worked is a note you can apply to the next brief. A star rating is a note that helps nobody, including the person who wrote it.

A take that holds up at the level of specific features is a take nobody can lift, because it came out of somebody's attention rather than their feed. The coverage was always going to end up free. This is the part that stays.