Christensen's finding that established firms are usually beaten not by better products but by worse ones that are cheaper and good enough, aimed at customers the incumbent does not want.
The starting observation is a paradox: the firms that got disrupted were often extremely well managed. They listened to their best customers, invested in higher margins, and improved their products. Every one of those is correct management, and together they are the trap.
Sustaining innovation improves a product for existing customers along the dimensions they already value. Incumbents nearly always win these.
Disruptive innovation starts worse on the dimensions the mainstream cares about, but is cheaper, simpler or more convenient. It takes root either at the low end — customers the incumbent is happy to lose — or in a new market of people who were not buying at all.
The incumbent rationally ignores it. The new thing is worse, the margins are lower, and their best customers do not want it. Then it improves — as products do — until it is good enough for the mainstream, and by then it is too cheap to fight.
The word is badly overused. Something is not disruptive because it is new or successful. It is disruptive if it started worse and cheaper, and moved up.
Test the definition before using the word: did it start inferior on mainstream criteria, and cheaper? If not, it is competition, not disruption.