I'll be the first to raise my hand here and say this isn't something I would consider doing until there are some major changes in the email inbox universe. I think it's great to be pushing at the very edges of the technology, but there are some substantial barriers to something like this working for "most inboxes today". We're certainly moving in that direction but most of the stuff that's out there now is REALLY dependent and too often that gets kind of "covered up" by all the excitement about "doing the thing" in an email.
Here's a clip from Litmus about the market support for html5 and css3 which, in my humble opinion, are still not "safe" (50%) for the mainstream channels of email:
https://www.litmus.com/blog/a-bulletproof-guide-to-using-html5-and-css3-in-email/
While support certainly isn’t universal, many of the leading email clients support HTML5 and CSS3. In fact, about 50% of the total market and 3 out of the top 5 email clients support them. Support may be even bigger for your particular audience.
... as far as I know, baseball batting averages (.333) are about the only thing that would be remarkable if it worked 50% of the time -- that's the stuff of coin flips.
If you've got a very specific audience in mind and you are sure they all use an inbox which supports html5 and css3 this might be something worth looking at -- but even then, you're dealing with a fraction of the functionality that you'd get by making a single compelling CTA and sending the user to a website. Demographics might also play a role in this depending on how the trends in email continue (or don't) to play out as user groups change their behaviors into the future -- maybe it'll make more sense for one group (us old timers who still remember email marketing) and maybe less for another (the new kids on the block).
From my POV, the more important piece to consider here is not "what happens when it works" (folks are delighted by the novel experience and clickity-click away) but rather "what happens when it doesn't". If you've ever gotten a really poorly delivered invite to an event or wanted to click a link but it didn't work you'll realize that the cost of thing not working is often more in terms of brand reputation and trust than the cost of delivering the bleeding edge experience in the inbox.
Another important consideration is that inbox scanners are getting better and better and "reading" email code and something like this (from a codebase perspective) is sure to sound the alarms to be flagged as "clearly a marketing email" (think Gmail's Promotions tab, but everywhere) because of the variety of techniques you'd need to use to achieve something like this (assuming it works in their inbox in the first place).
Overall, I really like the ideas that are laid out in the these articles, they're very forward-thinking and an email microsite sounds like it'd be a great addition to the email toolbox in the right situation. If I had to put a dollar down on it though, I'd bet this wouldn't be a thing (more than a yearly fad) anytime within the next 3-4 years -- and I'd pay a dollar to be wrong about it! 🙂
