I do believe (or rather guess) that 4e has been just as successful as 3.x over its (continuing) lifetime, and in fact, including the introduction of the subscription model, is likely their most successful edition of Dungeon & Dragons ever (both in sales and in player base).
To assert that it's just as decent as 3rd in regards to sales, yeah, I might grant you that simply because it's hard to say. Although some other publishers have flat-out stated that 3rd was bigger than 4th, in their blogs. But truly, it's fuzzy, so if you think that 4th was bigger, fine.
However, there is absolutely NO WAY that 4th is the biggest ever. Back in the early 80s D&D was a
phenomenon that has never been reached since. Seriously, there were
millions of players, everywhere. Ads were on TV in prime time, it was featured in movies, there was mass-hysteria that it was tied to Satanism, etc. The prime movers from that time have gone on record with how well it was selling (without stating actual numbers) and they've been clear that it was halcyon days the likes of which we've never seen since. So no, I would want to chime in as an old-timer to at least make sure that
nobody ever thinks that what we've seen with 4th is what it was like back then. It absolutely is not. It's a flash in the pan compared to back then.
If nothing else, I've lived through it personally, and these days we're just pale shadows of then.
A recurring question, with unknown answer, is how much does it cost to keep that online model running and how much was the cost of creationg and expanding it?
We've had other developers in this thread speculate, and generally presume that it's profitable. And even though I utterly despise 4th edition, as a Web Engineer myself, I'd definitely say that 4th edition has a victory there.
I've been leading teams on building Web apps since 1994. I've built sites for Borland, Yahoo, Cadence, Actuate, and about a dozen others. With a team of 3 or 4, I can reproduce their online app. Since we know that there are 50,000+ members of the subscriber forum, and since we know that it's only a subset of the total number, we can safely assume that the monthly income has a
floor of about $300,000 USD. Me & my team would cost about $50,000 a month, leaving a quarter of a million in profit. Even if we assume that there is a hardware expense and a small team for that, we still walk away with about $200,000 in cash each month. That's the
floor or the bottommost projection. I would not at all be surprised to hear that it's bringing in closer to a million dollars each month. Something that brings in a yearly profit of over a million US dollars at the low end and possibly 10 or 20 million dollars at the more speculative end of the scale is nothing to sneeze at. Were I running Hasbro, this is something I would take note of and desire to perpetuate.
Of course, it's possible that since they are not really a technology company, they may have a team of 10 or 15 working on it, and working on it badly. If they aren't able to get the hot shots needed to crank this Web app out well, then yeah, they're making far less. They could even be just breaking even, though honestly, if that's the case, they're really mismanaged. Here in Silicon Valley, we do apps like this as a matter of course. I don't want to say that they're trivial, especially with heavily cross-pollinated data sets. However, I do want to say that similar things get built often, and typically with high robustness, as many Silicon Valley companies are all about Web transactions and building sites that do things with money. Since everyone is paranoid about money loss, these apps are hardened and really solid. That's what we
do. So if a typical small Web team here can do it and be profitable, Hasbro ought to be experiencing the same.
Again, this assumes they are not mismanaged. Jokes about TSR mismanagement aside, I'm willing to grant this one to them: they
are enjoying profit from the app. If they're not, that's really, really sad.