Monday, April 21, 2014

Will Access Costs Need to be an Order of Magnitude Lower in Global South?

“Heads I win, tails you lose” is an aphorism about “unfair” coin tosses that applies to Google and Facebook. To a significant extent, both firms can correlate revenue growth with the growth of “people who use the Internet.”

Internet service providers have no such luxury. China Mobile grows revenues primarily when people use its branded services, not when people get Internet access from any other ISP.  

That correlation of revenue in the case of Google and Facebook, and the causation of revenue for any other ISPs, explains why Google’s business behavior is so different from that of any other ISP.

Though partly an ISP itself, Google actually can grow revenue whenever additional Internet users can be created. With growth largely saturated in the developed world, Google (and Facebook) have turned their attention to how to add billions more users in the developing world.

Correlation is not causation, it always is important to remember. It is not actually the case that the incremental users “cause” Google’s revenue to grow.

Rather, in an indirect way, the growth of Google users creates a more compelling business case for advertisers to use Google venues. So revenue “follows” user growth, even if the user growth does not directly drive the revenue creation.

How different a process that is, compared to the revenue model for AT&T or Orange, where each new customer actually represents direct revenue.

That, simply enough, explains Google’s serious interest in seeing “everybody” get connected to the Internet.

Of course, as you might suppose, eventually, when “everybody” does get connected, revenue growth--to the extent it historically has been correlated with growth of the Internet user base--app providers such as Google will face new revenue growth issues.

At that point, at least for its Internet app businesses, Google’s interest will shift from user growth to average revenue per user growth, as often happens in most other businesses reaching maturity.

But to sustain growth beyond the present era, both Facebook and Google will have to retool their business models to chase “something else” that drives rapid and sustained growth, after growth from the current model slows and then largely ends.

It is that longer term “what will we do next” question that underpins Google’s experiments with driverless cars, robotics and now drones, even if drones might also play a role in the more-immediate issue of getting billions of new Internet users connected, as rapidly as possible.

Also, as with most other businesses, not all customers are equal. In Google’s case, the value of incremental users likewise is unequal.

“In the most recent quarter 53 percent of Google’s revenues came from the U.S. and U.K.,” says Horace Dediu. “Five years ago the proportion was 61 percent.”
Screen Shot 2014-04-18 at 8.28.15 AM
source: Horace Dediu

That matters for ARPU since U.S. and U.K. is at least $86 per user, while users in the “rest of the world” represent about $12 per user.

Google’s “rest of world” category includes many developed countries such as all of Europe and Japan, where getting additional users will be quite incremental.  

Without any doubt, the fact that most of the remaining users will be added in the emerging markets is key.

The problem for Google is that it has an order of magnitude less income per user. In part, that is why Google and Facebook spend so much time trying to figure out whether there is a cheaper, faster way to bring Internet to billions of people who do not have access to the Internet.

If one assumes revenue upside is an order of magnitude lower than at present, then access costs, all other things being equal, have to be an order of magnitude lower as well.

Aside from providing a big challenge for new providers, those assumptions also will affect revenue potential for fixed network and mobile service providers as well.

As has been true for much of the telecom business over the last couple of decades, getting costs aligned to revenue remains a key challenge for access providers.



No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...