3 ms·
Is this always true? Or can we just not come up with compelling counter-examples, because the brands that won in their category are more compelling than the br
by trjordan 11y ago
Is this always true?
Or can we just not come up with compelling counter-examples, because the brands that won in their category are more compelling than the brands that split a category?
- shalmanese 11y agoThe IM market in the 2000s, a market where you would expect the most network effects, remained fairly evenly split between ICQ, MSN and AIM for a long time. This always struck me as one of the most interesting counter examples since it seemed so utterly defiant of the model.
- rtkwe 11y agoI think part of it is that with IM clients you're mostly talking to a closer group of friends first then random strangers next so whatever your group chose you would use but there was no strong incentive to join with what everyone was using because you'd just be chatting with unidentified strangers if you moved out of your friend group into chat rooms. Also the barrier of entry is really low and there's minimal effort in using two or more IM clients running, you could have both running and easily jump into whichever had the group you wanted to chat with. And unlike social media, like Myspace vs Facebook (when they were still competing), participating in both wouldn't mean uploading everything twice.