Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Google Privacy

Google's Business Plan For Nest: Selling Your Data To Utility Companies 167

jfruh (300774) writes "Google spent $3.2 billion on Nest. How is it going to make its money back selling high-end electronic thermostats at $250 a pop? Well, keep in mind that Google is a company that makes its money off information, not hardware. In fact, Nest is developing a healthy revenue stream in which it sells aggregated user information to utility companies, to help them more efficiently plan their electricity-generation scheduling. The subscriptions net Google somewhere in the range of $40 per user per year."
This discussion has been archived. No new comments can be posted.

Google's Business Plan For Nest: Selling Your Data To Utility Companies

Comments Filter:
  • by Jason Pollock ( 45537 ) on Friday April 25, 2014 @11:56AM (#46841529) Homepage

    The single biggest thing that power companies worry about is demand. Long term, short term. They worry about it second by second, millisecond by millisecond even. If someone has a better short term model they can make money.

    This is why power companies worry about the weather so much. It has very little to do with the sun or the wind. Who cares if you're a coal/natural gas plant?!?

    Except you do. The generators all have consents that say that the can only take a certain amount of water from the river for cooling, and they can only raise the temperature of the river a certain amount. That means the temperature of the water (and air) are very important. There is a direct correlation between weather and the amount of power they can generate.

    Add to that demand prediction. Sure, they've got a model, and smart meters tell them when you are likely to use power (based on previous patterns). However, Nest's data will tell them when you are going to use power. 100%. Even better, Nest is able to _delay_ that power use, or shift it to when it is cheaper. It will even result in a more stable grid, since that data feed will allow the generator to know when there's about to be a brownout.

    New Zealand already does this with "ripple control" on water heaters. Suppliers turn water heaters OFF at the meter when power prices get to high.

    This is not about snooping on what you do (the power companies already know), it's about the grid itself.

  • by unimacs ( 597299 ) on Friday April 25, 2014 @12:12PM (#46841673)
    The utility company doesn't know when you go to bed at night, when you get up in the morning, when you leave for the day, and when you get back. The NEST does.

    Further the NEST knows how long it takes to warm or cool your home to a given setting as it relates to weather conditions. The NEST knows how quickly your home loses heat in the winter and how fast it heats up in the summer.
  • Re:demand response (Score:4, Informative)

    by unimacs ( 597299 ) on Friday April 25, 2014 @12:16PM (#46841717)
    That's already an option with our electric utility. In exchange for a reduced summer time rate, we let them cycle our A/C. The don't do it through the thermostat. They have a box directly connected to the compressor. There are limits as to how much time it can be off. There's been times when it's been activated at our house but it's usually on again before there's been very much of an impact on the the temp of the home.
  • by swillden ( 191260 ) <shawn-ds@willden.org> on Friday April 25, 2014 @01:12PM (#46842391) Journal

    The most important part of Nest's utility-based revenue model (which predates the Google acquisition, BTW) isn't the data that flows from Nest to the utilities, it's the command signals that flow the other direction, enabling the power company to adjust your thermostat in order to reduce the demand from peak AC utilization, in a way that isn't likely to bother you.

    The basic idea is that the utility company can tell Nest that peak usage is between, say, 4 and 6 PM. Nest can then tell the thermostats to turn the AC on from 3 to 4 PM, to pre-cool your house and keep your AC from running during the peak hours. The details depend on your desired temperature ranges, what your thermostat learns about the thermal characteristics of your house, predicted external temperatures, etc. Nest can also try turning your thermostat temperature up a bit to see if maybe you are willing to put up with a little higher temperatures... and the device can learn whether or not that's the case based on whether you manually bump it back down.

    Utilities are willing to pay quite a lot of money to Nest for the ability to better spread and manage their loads. That is the main reason they pay Nest, not for the usage data... and according to some articles I've read, Nest generates more profit from the utilities than it does from selling devices. Consumers generally save money by reduced overall consumption as well.

    http://techcrunch.com/2014/04/18/nest-uses-its-data-to-turn-electric-utilities-into-cash-cows/

    I can see Google integrating the thermostats with other information to make this even more effective. For example, suppose information about your location (from your phone) were used by your thermostat to determine when you head home from work. If you're working late and no one else is home, there's no reason to kick on the AC until you actually leave the office. They could add in information from your calendar as well, to allow thermostat to potentially predict that you won't be going home at 4 PM because you have a 5 PM meeting, even though you normally do go home at 4.

    I expect integration with Google Now as well; imagine a Google Now card that tells you your home's current temperature or one that warns you when your energy usage is higher than normal, so your monthly utility bill will be higher. Perhaps you could be notified that your utility company will give you a couple of bucks if you're willing to turn your thermostat up a few degrees today, and given the opportunity to say "yes" or "no" on the spot.

    (Disclaimer: I work for Google but the above speculation about what Google may do is just my personal speculation about what's possible.)

All seems condemned in the long run to approximate a state akin to Gaussian noise. -- James Martin

Working...