The word has been worn down to mean "new and important", which makes it useless — under that reading every successful launch is disruptive and the term has no content. Christensen's original claim was narrower, checkable, and considerably more interesting.

The mechanism

A disruptive product starts out worse on whatever the established industry competes on. Lower quality, less capability, fewer features.

It is better on some other dimension the incumbents do not price — cheaper, simpler, more convenient, available to people who were not customers at all.

Because it is worse on the main dimension, the incumbent's best customers do not want it. It gets adopted at the bottom of the market or by people who were previously priced out entirely.

Then it improves. And because the dimension it was already better on does not degrade, there comes a point where it is good enough on the main dimension and better on the other one — at which point the incumbent's customers move, quickly, and the incumbent has no answer because catching up would mean abandoning the business it has.

The part that makes it a real argument

The incumbent's failure is not stupidity, and Christensen was emphatic about this because it is the whole point.

Every step of the incumbent's reasoning is correct. The new product has lower margins, so pursuing it dilutes profitability. Its customers are the least valuable ones. Its capabilities do not meet the requirements the best customers state. Listening to your best customers, protecting margin, and investing where returns are highest are not mistakes — they are what competent management consists of, and they are precisely what produces the failure.

That is why it is called a dilemma. There is no version where the incumbent is simply paying attention and therefore survives.

Why the distinction is worth keeping

A superior product entering at the top and taking the market is competition, not disruption, and the two have opposite implications.

Against competition, the incumbent's answer is to improve — same game, harder. Against disruption, improving is accelerating the failure, because moving upmarket to protect margin is exactly the move that cedes the bottom and shortens the runway.

Collapsing the two into one word destroys that distinction, which is the only thing the theory was for.

Using it without overusing it

Three questions, and a genuine case answers yes to all three.

Is the new thing worse on the dimension that currently decides purchases? If it is better, this is not disruption.

Is it serving people who were not buying, or buying the cheapest option? If it is taking premium customers, this is not disruption.

Is the incumbent's reason for ignoring it a good one by its own numbers? If ignoring it is obviously foolish, this is not disruption; it is negligence.

The theory has taken real criticism — the case selection in the original work has been questioned, and the predictive record is weaker than the explanatory one. That is worth knowing and does not much damage the useful part, which is the observation that a firm can be destroyed by doing everything its own management discipline tells it to do. That structure recurs, and it is not visible without the concept.