When people say that a webcomic is “interactive,” what they usually mean is that you have to click around inside a user interface in order to uncover static content that has been produced and hidden in advance behind, for example, a branching narrative path. No real changes are being made “at runtime,” as the engineers say. The thing exists, and you are reading it, one panel at a time, one word at a time, every word and every picture set in digital stone by a careful author, just as solid and unchanging as if it were a printed graphic novel. The order in which you may choose to read the panels can change, but not the panels themselves.
Compare that to, say, World of Warcraft, where every moment-by-moment incident is invented by the players (within the bounds of play set up by the designers). By interacting with the game, you are creating something, a sort of narrative, that didn’t exist before. “I run this way, this fast, doing these things, talk to this person, talk to that one, and make these seventy-two hundred microdecisions within a couple of minutes, this step that step jump stop jump, which in turn causes these other things to happen, which in turn makes me need to do this and this and this” and so on and so on. Even if it’s a heavily-constrained narrative (and, yes, it is) what you end up with is your own.
Anything less than that is only the illusion of interactivity.