Alexa is now on iPhone. Quite the achievement from Amazon. Not from a technical standpoint – porting the coolly-voiced automaton into mobile likely didn’t involve too much work. A microphone is a microphone is a microphone. Nor, I think, will this lead to a sudden explosion of Alexa on mobile, confined as she is to the Amazon app, which is far from the Trojan Horse some pundits seem to breathlessly want it to be.
But to somehow persuade notorious control-freaks Apple not to block it in the first place? That’s an accomplishment. It also possibly gives us a glimpse of where this road is heading.
As it stands this isn’t a partnership at all, but a rather feudal arrangement where Baron iPhone has granted one of its more successful vassals an unusual amount of autonomy. There must be a reason for this, and I’m guessing Apple’s short-term strategic verdict is that, on balance, it’s worth getting people more into the habit of using voice with their hardware.
But in the long run I wonder whether we are seeing the beginnings of a sort of chain of command. Now that the Alexa technology is accessible through the operating system, it doesn’t feel like a huge leap to imagine Siri being the one telling it what to do – not the user.
There’s a couple of reasons why this might come to pass. First, experience points to most users wanting more functions in fewer places, with the notable exception of China. This is why cool single-purpose apps tend to get great PR, totally fail to attract substantial user bases, then get acquired by bigger players (see Word Lens).
Now that the Alexa technology is accessible through the operating system, it doesn’t feel like a huge leap to imagine Siri being the one telling it what to do – not the user.
Voice assistants, like Search, will probably be a zero-sum game with two or three winners at most. In this context, the hardware advantage currently puts Siri in a much stronger position than Alexa.
Second, Alexa is not AI. It really isn’t. Guided by its 10,000 programmed "skills", it’s almost totally deterministic and is laser-focused on fulfilment, just as its parent company is. Outside of those parameters, it’s practically deaf and blind.
I’m not saying this is a bad thing. On the contrary, it is exactly what users have shown they want from assistants right now, which is why Amazon is currently winning and why Google probably has to settle for the long game.
That’s a topic in of itself, but the upshot is that Alexa is currently designed only to give you exactly what you want – whereas Siri, like Google, is trying to be in the business of actually understanding what that is in the first place. This in time will allow it to turn complicated voice queries into crisp, direct commands.
In short, try envisaging Siri as the head butler to Alexa’s first footman. If there is a covert partnership between Apple and Amazon, and if it’s going anywhere specifically, I’d guess it’s in that direction. A compelling prospect and a powerful competitive play (Siri and Alexa currently both do one half of what Google does), but on the other hand it’s hard to imagine Amazon deciding to completely ignore Google in favour of permanently shacking up with Apple. Not impossible, though.
So what does all this mean for brands? As the entire history of technology shows, lesson one is that you probably can’t build your own thing. Previously, it was brands creating rudimentary operating systems/CMS/protocols. Today, it’s brands creating chatbots.
The temptation to flirt with voice AI as a function is strong, but unless it’s linked to the master platforms where the actual people are, it will likely be an expensive waste of time. Be the footman, not the butler.
Lesson two is related to this. Let me put it this way, who do you build websites for? The market? The consumer? The customer?
No, you build them for Google.
In the exact same way that brands structure their website information to respond to typed searches, they will need to structure all their information to respond to the symbiotic entity of voice search and digital assistants.
If the assistant can’t understand your information, it can’t link it to the voice request. And if it can’t do that, then it’s going to ignore you. Think of it as SEO for the brain.
Lesson three is not so much a lesson as a salutary reminder. We are all dancing on the knife-edge of disintermediation. There are many ways to respond to this. Some might decide that, hey, Amazon is going to keep eating their lunch anyway and so instead of throwing money at a lesser point of sale (viz. their own website) they can reinvest in brand instead. That’ll be a pretty good answer for some. Not all. It’s probably a good idea to know what yours is.
With that all said, there’s no need to panic. Voice activation is situational and unlikely to become a default behaviour anytime soon, let alone a mainstream activity in the first place. Your mileage may vary of course, especially if you target younger audiences, but there’s plenty of time to get your house – by which I mean, data – in order.
Just make sure it’s someone’s full-time job.
Phillip Dyte is strategy director at iProspect.