Blog Post

Honeywell killed off its learning thermostat 20 years ago

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!
Honeywell's thermostat with Opower software

Honeywell (s HON), one of the world’s largest thermostat makers, tells me that twenty years ago it tested out thermostats that can learn the home owner’s behavior and adapt the heating and cooling accordingly, but ultimately decided that consumers didn’t take to them, and would rather control their thermostat themselves.

I asked Honeywell’s President of its Environmental and Combustion Controls division,┬áBeth Wozniak, in an interview if Honeywell was interested in making learning thermostats, because there’s been so much discussion about the startup Nest, which has created what it calls the world’s first learning thermostat.

“We found that consumers prefer to control the thermostat, rather than being controlled by the thermostat,” said Wozniak. Instead of learning thermostats, Honeywell is focused on adding intelligence to digital and connected thermostats through simple UI, mobile apps, and partnerships like its one with Opower.

Opower will be providing the analytics and data to help Honeywell use home and building thermostats for demand response programs, where utilities can ask home owners to turn down their heating and cooling slightly during peak times of day. The Opower thermostats are being piloted with utilities right now, including at PG&E. The Opower software will also be used to create new ways for the home owner to save money on their energy bill, and Wozniak says by the end of the year the partnership will launch other products too.

For Honeywell, connected thermostats are still a small part of the company’s overall thermostat sales. While Wozniak declined to say what percent or what volume of Honeywell’s thermostat sales are connected thermostats, she said it’s the very early days of the connected thermostat market. Honeywell sells a whole host of other connected home products such as humidifiers and security systems, and a “total connected home system.”

Who knows if Nest and its learning thermostat will one day make a dent in the thermostat market, but Wozniak acknowledges that the startup has brought some much-needed attention to consumer thermostats in general. “Cell phones and tablets have set a whole new bar for how things can be connected.”

14 Responses to “Honeywell killed off its learning thermostat 20 years ago”

  1. Rich Hilt

    No matter how smart the meter, you’re still heating or cooling the whole house. Until you change over the heating and cooling systems to take care of individual rooms. Most hydronic systems at least have the fundamentals for this. But the cost of forced air heating systems are so much cheaper to install and maintain.

    Just one more reason why Sillycon Valley’s extending “Moore’s Law” to CleanTech hasn’t produced what some of us had hoped for.

    Thanks Juliet Eilperin at Wired for holding the mirror up.

  2. Teusg rettemoc

    Interestingly, the CEO of Palm (makers of the Palm pilot, treo and other handheld devices since the 90s) in 2006 said about the iPhone: “We’ve learned and struggled for a few years here figuring out how to make a decent phone… PC guys are not going to just figure this out. They’re not going to just walk in.”

  3. I’ve been getting a kick out of a thermostat being developed in Silicon Valley. There are very few Minnesotans or others who live in “real winter” areas who are going to be willing to spend even a few January morning waiting for their freaking thermostat to learn anything. The first thing they do when they get a programmable thermostat is set that baby up to turn on the furnace before they get out of bed.

    • I’m in Florida, in the winter, and I pre-programmed all three of my Nests. I’ve had quite a few programmable thermostats over the years (having three separate AC systems really makes you think about controlling usage) and the Nest is far and away the best. Honeywell makes cookie-cutter bare-minimum junk. This isn’t a patent suit, it’s restraint of trade.

  4. There is the key – much needed attention!

    The average consumer has too much on their mind to spend time working with their thermostat, or any electricity device for that matter.

    Not sure how this will change – but like many, we are working on the underlying issue – our energy attitude!

  5. Kary Krismer

    Honeywell realized 20 years ago that it’s not a good idea to have a thermostat automatically go up just because someone turned it up yesterday or last week. It’s an energy wasting device, IMHO.

    • Honeywell didn’t have the automatic “Away” feature that sets the t-stat to minimum power usage when it detects nobody is home. Honeywell didn’t have smartphone integration to let me adjust the t-stat while I’m stuck in traffic, or out grabbing a late dinner or movie. Honeywell didn’t have a web-based UI for manually tweaking the learned schedule if I didn’t like what it was doing.

      Get the picture?

  6. MichaelBrianBentley

    I’m not sure the tech required to implement Nest existed twenty years ago. What processor and how much memory was included in the Honeywell smart thermostats? What philosophy did Honeywell use to implement the software?

    The motivation for Nest, in part, is to reduce the heating bill as well as be more in tune with what the occupants want. Honeywell users complained that their smart thermostat was bossing them around; did everyone crank the setpoint past 72, and the thermostat just not let them do it?

    I bet that the Nest is smarter about paying attention to what people want.

  7. The key thing to consider is that it has to be easy and attractive enough for people to use. Just like there were tablets before the iPad that didn’t catch on, learning thermostats will have existed before the Nest. But if one company releases a gamechaning product that introduces a radical new way of communicating with these devices a new market appears.

    That’s why I don’t really put much importance in a company saying: “We tried, but people didn’t like the concept”.

      • Things were different in that technology wasn’t at the point it is now. We can build more user-friendly devices with better interfaces and better integration with other devices such as our phones.

        If you can adjust your thermostat with a device you have on you almost every minute of the day, you’ve already got a big advantage to anything that was available 20 years ago.

        Also just learning about human behaviour and having better algorithms in place (which may require stronger processors) can make a huge difference.