Many people think of Amazon as a friendly retail giant, and of Google as a search giant.   Why would they get into the assistant and home audio device markets?   To understand, we must digress…

A Digression on Internet Business Models

Google and Amazon (and others internet businesses) are fundamentally similar.  They have profitable advertising service businesses.   Google attracts users with a search engine and sells those users’ page views to advertisers.   Amazon, attracts users with their retail offering and monetizes them through sales or via affiliate partners.

These businesses need three basic things to make money:

  • users
  • engagement/traffic from their users
  • data about users to improve targeting

All three determine how much money the company makes.   Some examples:

  • Facebook has tons of users, tons of engagement, and pretty good data
  • Google has tons of users, decent engagement, and extremely good data
  • Amazon has lots of users, poor engagement, and extremely good data
  • Twitter has decent numbers, decent engagement, and ok data

The key to judging the quality of the data is whether it leads to knowledge of purchase intent. Thus Google and Amazon have the best data when I search for a product I want to buy.   Facebook can profile me and predict things I might want to buy.   Twitter less so.

Engagement – or traffic – is key.  In the ideal case, users go to their browser, type http://www.amazon.com, and buy something.   But often there’s a traffic director between the user and the service who demands to be paid.   This is referred to as Traffic Acquisition Cost (TAC) and it’s often a huge part of the cost of doing business.   In 2015, Google paid partners $14B in TAC – that’s more than they spent on running the search engine, R&D, Marketing…their single biggest cost.

Who demands TAC from Google?   PC Makers, to be the default search engine and browser.   Apple and Mozilla, for being the default search engine in their browsers.   But Google’s search business is furiously profitable, so they can afford it.

Amazon – with very low margins – also has to pay TAC.   This cost could be make-or-break.   And who does Amazon pay?   Well, primarily Gooooooogle!   In 2011, when Amazon was much smaller, they were estimated to pay Google $55M in TAC.   Amazon is one of Google’s biggest customers and Google is one of Amazon’s biggest costs.

We can see there’s a hidden internet war going on over traffic.   Reducing TAC is a huge lever on all these businesses so they’re all trying to get between each other and their customers.   Some examples of this in practice?

  • The Google app and Chrome browser on iPhone cuts Apple out of Google searches.
  • Apple Siri directs traffic from Google to other services that will pay more for specialized traffic (e.g. movie reservations, restaurant search).
  • The Amazon App and website cut Google out of a share of Amazon’s profits.
  • Google Chrome makes Google the default search engine and cuts others out of the equation.
  • Android and ChromeOS require that Google be the default search engine, thus cutting phone makers out of Google TAC.
  • Amazon Prime creates loyal users who go straight to Amazon for product searches, cutting out Google’s share.

So in summary, these businesses need four things – users, user engagement, data, and lower TAC.  And a lot of the cool products and apps you love are just weapons in that war.

Great.  So what does this have to do with Echo and Alexa?

A user with Echo/Alexa in their kitchen helps Amazon helps with several of these factors:

  • User engagement.   I previously spent minutes per day with Amazon, now I’m spending hours.   I listen to NPR every morning and evening, enjoy Amazon Prime Music, and even occasionally (ok, rarely) order something.   What’s more, Echo and Alexa make me love Prime even more.
  • User Data.   There’s some value here, but for now it’s limited.   They’re learning my listening preferences and what kind of random questions I ask.   But over time they may get much deeper and learn more purchase intent data.
  • TAC.  The real gold.   Amazon Alexa will always point directly to Amazon shopping and media services, cutting out all possible TAC intermediaries.   What’s more, if it makes me stay with Prime, I’m going to do all of my shopping directly with Amazon.

One of the best way to feed these levers is to own the OS platform or device, as shown by Android.   Amazon tried to make a play here with the Amazon Fire OS, Tablet, and Phone, but they were too late.   But in this new formfactor they will hope to be a leader and direct users their way.

Of course Echo also is a product, garnering gross margin.   But that’s a secondary benefit, as indicated by Amazon’s willingness to encourage 3rd parties to build Alexa hardware.  Echo is just an incubator for the real asset, Alexa.

Does Google Care?

You bet.    Alexa (and Siri!) are directors of traffic that cut Google out of ad revenue.   Remember that the best searches are purchase searches – for shoes, or insurance, etc.   Alexa threatens to take the best searches away from Google and send them straight to Amazon.   They could also take other profitable queries and sell them to other providers – restaurants to Open Table, movies to Fandango – just as Siri has done on the iPhone – and earn the TAC for themselves.  If Alexa takes off, it could seriously dent Google’s business.

Echo shipped in November, 2014.  Let’s say it took 6 months to show signs of taking off, taking us to May, 2015.   Sometime around then alarm bells should have been going off in Mountain View, and Google would then rush into action to get moving on a response, which would be ready to announce in, say 12 months.   If I’m right, I’d expect a response to be announced in May, 2016, at the Google I/O Keynote…

Next – Google’s response and what’s likely to happen next…