That's still mostly true, but looking at it some more, it's more of a Burger King problem.

When Burger King started out, presumably, it made a lot of money and expanded quite a bit. But what they found out (along with other fast food businesses) is that customers are fickle. Unless you mix it up now and then your customer base will slowly melt away to go to Five Guys Burgers, or some other new fast food place that everybody thinks is cool this week.

So what they and other restaurants started doing was launching these short-lived, heavily advertised changes to the menus. Miniature burgers. Mexican burgers. Burgers on a stick. Whatever. They kept a bunch of ideas on the back burner, rigorously A/B tested and used trial markets, then launched something new every now and then. There was always something new and different enough to keep folks from drifting off to the shack down the street.

Startups are now forming a somewhat similar function for these huge tech giants. A startup has some catchy idea, gains millions of addictively faithful users, then sell out to McGoogle (as an example)

Works great for everybody. Google gets new talent with a proven record of coming up with ideas that capture eyeballs, they get an instant boost in eyeballs for their services, and, better still, later on these features can be rolled into the code base. Maybe. Maybe they just buy a startup up to keep the other guy from having it. Works the same way.

I'd also note that this is at a much higher level than the "lifts" restaurants get. It's entirely possible that the next eyeball catcher has no kind of long-term value at all; but is still worth tens of billion in lift to some acquirer over a period of a decade.

So articles like this get it wrong. There's absolutely real innovation and value, but at least to me it doesn't seem like the same kind of innovation and value you get when you cure the common cold. It's more like making a cool new kind of french fry.