The Curious Cult of the Connected Thermostat

Last month, Nest Labs managed to haul in a further $80 million of VC funding for their Internet-connected smart thermostat.  That’s good news for Nest, but makes one wonder what the investors are hoping to get back?  There is no questioning its success in the US.  Nest claim to be shipping about 40,000 thermostats every month.  That equates to around 5% of the 10 million a year US market, which has historically been dominated by Emerson, Honeywell, Johnson and Lux.  But how much of the other 95% can they win?

A basic programmable thermostat in the US costs under $20, not the $250 price tag of the Nest.  As such, Nest appeals to those who like buying technology and form rather than function – it’s no surprise that it sells as an accessory in Apple Stores in the US.  It has all of the glamour and pizzazz of Apple products, but with a worrying limitation – it is just hardware – there’s no service model.  In other words, it’s a bit like an iPhone without an App Store.

There is no doubt that it’s a lot easier to use than most conventional thermostats, which seem to be exclusively designed by engineers who failed their user experience courses and want to get their own back on society.  However, there are plenty of alternatives which are cheaper, just as easy to use and which work outside the US.  And there have been for the last few decades.  But these alternatives have historically failed to sell.  That’s changed, but this new generation of connected wireless thermostats has an Achilles’ heel – they need someone to support the web service for their life, which may be ten to twenty years, and I can’t see where that’s been factored in.  So is Nest going to feather the pockets of its VC backers, or make an omelette out of their investments?

When I installed my central heating system some thirty years ago I tried to find a smart thermostat.  At the time I was surprised to find there was a choice.  They weren’t internet connected, but they came with temperature sensors for inside and outside that provided basic learning about the thermal characteristics of the house and they had a decent User Interface.  It was mostly buttons and LEDs, but it worked and was about the same relative price then as a Nest is today.

Five years ago I replaced my boiler (furnace if you’re a US reader) and looked for a current smart thermostat.  To my surprise I couldn’t find one.  When I asked the major manufacturers about them I kept on getting the same story – they’d designed them, taken them to market and they didn’t sell.  I was one of a handful of customers who had ever bought one.  As a result of which they’d all been withdrawn.

Nobody doubts the advantage of having some form of smart heating control in the home.  The controllers or thermostats that are typically installed are set once by the installer and never changed.  Except for the few occasions when it’s really cold and they get turned up, and never turned down again.  Which means that almost every household is wasting money on their heating.

Now we have Internet connectivity, Smartphones and Apps, which should make these devices simpler to use and even more attractive.  And there’s no doubting that the Nest is that.  None of the underlying technology is new, it’s just beautifully packaged and very well executed.  Emerson, Honeywell, Drayton, Schneider and the rest of the industry have been trying and failing to sell smart and connected thermostats for the last ten years. They’ve failed not because of the technology, but mainly because of the channel to market.  These devices are bought and installed by plumbers and heating engineers, who in most cases neither understand them, nor see any advantage in paying $250 for a thermostat when they can buy a simple one for $20 and then add a 500% mark-up.  If a thermostat fails, a householder can either buy a replacement and fit it themselves, or call an electrician who usually does the same marked-up $20 thermostat trick.

In the US, Nest is being successful because it has found a different model and channel to market.  Despite their marketing, they’re not really selling energy savings – they’re selling desirability.  It is a beautifully designed product.  All that consumers need to do to give it a home is unscrew their nasty old thermostat and replace it with a Nest.  Then call their neighbours round and feel smug.

That model works in the US because most heating systems use 24V for their control wiring.  Here in Europe, most switch the live wire at 230V.  Which means that the self-replacement market disappears (or is illegal in many countries), adding an extra $150 of installer cost, which dents the desirability.  But I have a bigger concern about the whole connected thermostat model, which is who maintains and pays for the server?

I don’t think I’m unique in having a different attitude to consumer products and household controls.  Whilst I might change a phone or a laptop every few years (but increasingly only do that if they’re broken), I expect my thermostat or light switch to last for fifteen or twenty years.  Where that’s a connected device allowing me to adjust the heating from my phone, I’m going to expect the service behind it to last for the same fifteen or twenty years.  I think that’s where the connected thermostat vendors are going to have a problem, because that means I’m going to expect their servers to be up and running for those fifteen to twenty years.  Which is something that no-one has done yet.  I’m not going to be content if the thermostat reverts to working in a local mode after a few years when Nest goes bust or gets acquired by someone else, because I’ll have paid for a connected thermostat which I expect to remain connected in perpetuity.   Particularly if it cost me $250.  Nest have already had server outages and their customers are far from pleased.  As one put it earlier this month, “I am being patient but it’s starting to wear thin! For what we paid for these devices they should be able to afford to give them the highest level of support on the back end”.

This is an issue with every new connected device that comes to market – who will keep the servers running?  It’s not so much of a problem for fast moving consumer products, which will get replaced regularly.  But it is very big problem for home infrastructure products like smart thermostats.  To see why, think about the economics.  These devices are claimed to save me around $180 a year.  As a user I’m unlikely to invest in one if it costs more than that $180 and I will feel aggrieved if I have to pay more than $25 a year for an on-going service.  Even that is probably double what most people would be prepared to pay.  I’ll expect that to include updated apps for any phone or tablet I might buy over the next ten years, plus technical support for each of them.   The obvious place to get that from is the thermostat manufacturer, but supporting a ten or twenty year commitment is a big ask, particularly if they’re a start-up.  Not least because just administering all of those subscriptions at that level is a major cost.  And I can’t think of any providers I’ve had a twenty year relationship with, except my bank, who I certainly wouldn’t trust with my heating control.  I switch my mobile provider, my broadband provider, my home insurer and my utility every few years.  So who is going to support my access to a proprietary thermostat over a decade or more?  As I acquire more connected devices that problem will multiply.  It’s a major hole in most Internet of Things business models and probably needs a global aggregator to come along if they’re going to work.  Which may be someone like Google…

A couple of players in the US have tried monthly fees, but none in large enough trials to know whether it works.  And these are very distorted by the local Energy Efficiency rebates that are available to utilities over there.  In Germany, Tado is trying the service fee route, and it will be interesting to see how large a customer base it can attract and retain.  Without service fees, these products are a hardware sale, which carries ongoing support and liabilities, but with no incremental revenue.  Nest has played all of its Apple pedigree cards to get to where it is today.  Hpwever, the market and most of the IP is still owned by the traditional players.  I suspect they’re looking carefully at Nest’s market share.  If Nest doesn’t get mired in problems of ongoing service availability as its market share grows, they’ll be lining up a new set of patent infringements to halt its growth.

It doesn’t help that other large players are reinforcing the “no service fee” model.  Here in the UK, British Gas has entered this market with a controller supplied by AlertMe.  They’re offering installation for £229, with no service fee, claiming it can save up to £140 on heating bills.  After an initial bit of enthusiasm, press coverage has been largely negative, with most coverage claiming that it’s not worth the money.  Despite that, British Gas claim it’s going well, but they have the largest boiler servicing business in the UK, which lets them do some attractive bundles to shift kit.  It should have given them the opportunity to deploy a service fee model.  But they didn’t.  And it’s difficult applying one once you’ve given it away for free.

So the connected thermostat looks set to remain a hardware only play.  Whilst companies like EcoFactor and Tado are quite sensibly trying to monetise the ongoing service, Nest in the US and British Gas in the UK are firmly setting a stake in the ground saying that these are one time hardware purchases.  And for new entrants and Nest’s new investors, that means thinking about how much the ongoing server support will cost for each thermostat over the next twenty years.  I hope that Nest are investing a fair chunk of their $80m to provide that ongoing support, otherwise the ensuing customer dissatisfaction may result in another premature halt to the hopes of smart heating controls.