Google wasted a whole year coming up with an answer to the iPad, and when they did, it felt rushed to market and half-baked. It was buggy and feature incomplete. Not to mention that it seemed (and it still does) that they weren't even considering 3rd party applications for tablets. It was pretty obvious from the beginning that they thought the phone apps are perfectly fine for tablets, so they never really bothered to fix that.
While the iPad is increasing its headstart in apps by hundreds of thousands of apps, Google now has another problem as well: Windows 8. It baffles me that Google didn't seem to take this threat seriously as well, and instead risk becoming the 3rd platform for tablets, even though by the time Windows 8 tablets will launch, they would have almost a 2 year head start, and yet so far it feels like those years didn't even exist. That's how poorly Android tablets are performing in the market right now.
Google is quickly running out of time to become an established (at least) top 2 player in the tablet market. So here's the 5 things I think Google needs to fix ASAP (within months) before it's too late:
#1 Redesigned UI
The phone UI of Android has never worked on tablets. Honeycomb improved things a bit, but was very buggy and slow. ICS improved on that, too, but I think there's a large opportunity for Google here to do a whole lot more with their tablet UI, and I think the direction is for them to somehow take better advantage of their advanced widgets, but without making the screens look too cluttered. They should be much more functional than both Microsoft and Apple's tablet UI's, and yet still very easy for everyone to use.
I'm not a believer in having the exact same UI for every type of device, every screen size and every form factor. Each product category should have its own UI, fully optimized for it, rather than a mediocre one that works across all devices. That doesn't mean they can't look similar or have the same design language, but they should really empower the user to be very productive with that form factor, which I think neither Apple nor Microsoft accomplish this very well (well at least Apple has 2 OS's for their devices).
#2 Apps, Apps, Apps
I can't stress this enough. I remember when Apple originally announced the iPad, 2 months before its launch. Basically they made the SDK available from that day, and gave developers 2 months to optimize their iPhone apps for the iPad. This is why Apple had 2,000 apps at launch, and even though many though it's just a bigger iPhone at first, they quickly started to change their minds when they saw that tablet apps can offer significantly higher utility than phone apps.
Google needs to stop pretending that phones apps are just as good. That's the first step. If they can't even admit that to themselves and everyone else, then they will not do anything about trying to bring tablet apps to Android. And this is very important because tablet apps need their own ecosystems, and if Google ignores this they risk losing their tablet market - forever.
#3 Desktop functionality
If Google is really serious about Android, they will try to attack Windows in the desktop space as well. Larry Page said recently in an interview that tablets will get more desktop functionality in the future. I'm curious to see what that will be, and if we're going to see it in the upcoming Nexus tablet. My guess is that we won't because this is just a 7" tablet, so it won't make much sense to show too much desktop functionality for it, but we might see more of that with Android 5.0.
One thing that might stand in the way of this is their other OS, ChromeOS, which is kind of competing for the desktop as well, at least in their minds. But I think ChromeOS will end up mainly as an OS dedicated to enterprise customers who have moved to the cloud, and they want cheap hardware and maintenance for their thousands and thousands of employees.
The reason I think Android needs more desktop functionality is because when people want to be productive, they will want to work in the "desktop mode", with a physical keyboard, maybe a mouse, and so on. This is why while I think that a 7" $200 tablet is a great move from Google, it really should've come earlier than Kindle Fire, maybe at last I/O. Now they should've spent more time focusing on these Transformer-like hybrids to fend off the threat from Windows 8 to promises to be a good OS on the desktop, too, at least on the more expensive and higher battery draining Intel chips.
#4 Incentives for developers
If the iPad never existed, iOS developers would've probably switched to develop for Android first by now, thanks to the sheer number of Android devices that grow at a rate of like 3x more compared to iPhones, but "only" about 2x compared to the whole iOS ecosystem. But what the iPad did, is refocus developers efforts on iOS. Just when they were about to start caring more about Android, the iPad appeared, and since Google failed to gained momentum with Android tablets for all the reasons stated above, most developers just decided to focus their efforts on iPhone and iPad apps.
Google really needs to do something about this if they want to finally become the #1 platform for developers. They need to do all the technical work to improve Android across devices, to make sure that developers don't need dozens of devices to test their apps on. They need to provide them with financial incentives to make tablet apps designed with the Holo theme, and offer them better visibility on the Play Store, either with a specific badge or a different category in the market. It should be obvious for a user as well, which is an app that doesn't have a tablet version, and which is. This should also force the developers to come up with a tablet version before their users start yelling at them.
#5 More unified updates
One major thing they could do to diminish "fragmentation" is to stop pushing out 2 or more versions of Android every year. Manufacturers are slow enough with one upgrade even for their flagship devices, let alone 2 upgrades. ICS was also a pretty big overhaul, so it took manufacturers a lot of time to upgrade to it (6 months compared to 2 months for the Gingerbread upgrade for some devices).
I think it would really help if Google would just push one major version of Android every year, which means by the time the new major one comes out, even those manufacturers that needed 9 months to upgrade certain models, will still be on the latest version. This would of course help not just tablets, but phones as well.
Google is running out of time to remain at least a top 2 OS developer in the tablet market, and they risk losing even a bit of the smartphone market if Microsoft is successful in tablets. That's certainly not guaranteed yet, but while let them get so close? To fend off this threat, they really need to show some impressive this year, either with Android 4.1 or 5.0, and they need to do something drastic about the tablet app situation.