Print
Email
Print Edition StoriesThursday, August 19, 2010
Ember execs share expectations of smart devices and Zigbee chips
By Kyle Alspach
UMass President Jack Wilson on Entrepreneurship
IBM aims to make wireless sensors easy to code
Sure, the electric grid and most buildings could some day become “smart,” with new technologies that monitor energy use helping to drive down use and costs. But no one can claim to have the smarts to know exactly how we’re going to get there.
If you ask the team at Boston-based Ember Corp., a maker of the chips that are behind many of the early innovations in smart grid technologies, they’ll describe the current landscape this way: it’s a lot like the “wild wild west” days of the early Internet, a time of tremendous uncertainty about what path the technology would take.
When it comes to widespread adoption of smart building and grid technologies, “We don’t know whether that will happen in two years or 10 years,” said Robert LeFort, Ember’s CEO. “That’s the biggest question.”
But that doesn’t mean the opportunity is any less promising, LeFort said. And his company has already secured its place as the leading producer of ZigBee chips – itself the leading technology for the wireless communications that make the “smart” technologies work.
“The markets we’re in, the products we’re in, are ubiquitous,” LeFort said. “Every light switch, every electrical outlet, every smart appliance could have one of our devices in it.”
In May, Ember announced that it had shipped 10 million ZigBee chips, the first company to achieve the milestone. Yet the pace is picking up fast; Ember is on track to ship a total of 10 million chips in 2010 alone, LeFort said.
Currently about 75 percent of the company’s chips are used in smart meters, with 20 percent in other smart home devices and 5 percent used for other applications. Major customers include the major meter companies (Itron, Landis+Gyr, Elster, Aclara) and the major in-home device companies (Tendril, Control4, AlertMe).
Ember’s revenue is on track to grow by more than 300 percent this year compared to 2009, with the company now seeing revenue of $9 million to $10 million per quarter, LeFort said. The company achieved profitability in the first quarter of this year, he said.
Ember, founded in 2001 as an MIT spinout, has also grown its staff from 40 to 50 over the last three months, LeFort said. Thirty-five of the employees are in Boston, which serves as the headquarters and hardware R&D, while the rest are at the company’s chip R&D facility in Cambridge, England.
Recently, Mass High Tech sat down with LeFort and Skip Ashton, senior vice president of engineering, to talk about how the much-touted “smarter” future will be achieved and how Ember will fit into it.
Mass High Tech: Who is going to take the lead with installing smart devices in the home?
Skip Ashton: In home is a very complex market. Utilities can drive the meters, but utilities don’t put anything in your house – but they could. In reality, you could go to Lowe’s or Home Depot, get a do-it-yourself installation with some of this stuff. Or cable operators could come in and provide some of this stuff. Probably two-thirds or three-fourths of thermostats are installed by contractors these days, so contractors could provide it. How those channels sort themselves out is very unclear right now. In the end, you’ve got to follow the money – who’s going to make money putting this stuff in your house. That’s just not entirely sorted out. So the utilities and regulators have a desire to get it in your house, for energy efficiency, for peak demand response, and things like that. But who’s going to have the business model that gets this in there is still unclear. That’s the fundamental crux of the problem around the in-home stuff.
MHT: What role will regulators play?
Ashton: The regulators aren’t going to decide who wins and loses in the question of how this gets in the home. They can’t. And so in reality, this is not much different than the wild wild west days when the Internet got deregulated – you had lots of people doing lots of things, and there were winners and losers in that environment. And right now, we’ve got a lot of people working on [these issues], and the winners and losers aren’t very clear.
Robert LeFort: In hindsight, it will be obvious. We do think that the people who are most interested in consumers and most understand consumers are going to be the guys who will take the lead. Is that ISPs and telcos? Could be. Is that utilities? Probably not. Meaning, the relationship between the utility and the consumer is not one of really wanting to understand how they do things in the home. It’s getting energy to the home.
Ashton: To be fair, the regulators can help that. In reality, utilities are paid for the amount of electricity they sell you. Until you sort that out a little bit and provide some amount of incentive for energy efficiency from the utilities’ standpoint, then they’re slicing their own wrist. To say, let’s get all of our consumers to use less electricity so we can return less to our shareholders – that’s not a great business model.
LeFort: The regulators have been very unclear about things like time-of-use pricing, shared information, how they’re going to do that. With de-coupling, the utilities could sell less but make more because they are operating efficiently. Right now, they are rewarded typically by assets – the guy that builds the most makes the most. That’s not promoting efficiency. So there are a variety of pieces. The bad news is that this is very complex – a little bit like the Internet or the cellular business. A lot of people have to come together to make the thing work. But this is not day one – this is probably three years into this.
MHT: What are the milestones you’re looking for with Ember and with the smart grid in general?
Ashton: One milestone has gone by – we have 50 million meters under contract to put ZigBee in them. That was the first big milestone, getting meter deployments going that have this technology in them. The next one is the really widescale deployment in some area, some region, of in-home devices. Utilities are very much herd animals – once one state does it, others will do it. Texas is actually furthest along with this, where you’re seeing hundreds of thousands of devices in homes. This summer, we’ll see how many more they get to. California is behind on that, by comparison. We are waiting for the first time an article comes out that says hey, yesterday was a peak demand day in Houston and they saved 5 megawatts, using ZigBee-enabled home stuff - that will be a big milestone. We’re not there yet.
LeFort: A second big milestone is going to be someone that has a large channel, that starts reaching into the home and offering customers more services. And there we see people like the telcos and the ISPs who already have a variety of hardware in your home, who know how you operate and have an infrastructure to support consumers. ... As those services launch, and people start hearing that for a two-year contract, I can get this hardware that’s pretty cool and save money on the energy bill every month, that seems to be a model that people are understanding.
MHT: How soon do expect to reach these milestones?
Ashton: Probably in the next year.
LeFort: The announcements I’m talking about are going to be these relatively large channel announcements; I think that’s going to be in the early fourth quarter of this year. They’d be announcing services. For instance, a large service provider in the U.K. that’s offering a service with three devices – a magnet that clamps around your meter, a hub to go to your PC and a display. And the hub that goes to your PC would be feeding data into it for things like Google power meter, and they’ll probably have their own portal as well. And they’ll do all of that for a small service charge. ... In the U.S., it turns out that with home security, they see that as a big opportunity for them – the so-called quadruple play of voice, data, video, and security monitoring and automation as their fourth big service. So those are some of the announcements we expect. And once you have network for home security, it’s very easy to add a thermostat and add some smart plugs, so you can get all of this as well.
MHT: Do you have concerns about the ZigBee chips becoming a commodity?
LeFort: I have no doubt in my mind that a company like Texas Instruments or Freescale – TI is a $12 billion company, Freescale is a $5 billion to $6 billion company – if they put their best people and have a concerted effort and want to do a great chip, they can come out with a piece of hardware that will compete favorably with ours. No doubt. But they’re not going to change the way they do business. They’re not going to give the way they think about things. Do you think Porsche loses sleep at night that Ford is going to wake up and say, ‘we should build a sports car’? ... So I think we have enough, I’ll call it “secret sauce,” in the way we think about things, the way we do things. And we have the leading customers. So really, I think some day this technology will become a commodity, and then they will be able to catch up. But that’s at least five years away, and if we waste the next five years to not find other ways to differentiate ourselves, then there’s a risk. But we have plenty of time before that’s a real threat. And quite frankly, my message to my team has been, I want TI and Freescale to be successful. Because we cannot have a sustainable market with only one guy that can meet the requirements of their customers. My mantra is I’d much rather have a 50 percent or even a 40 percent market share of a billion unit market than an 80 percent market share of a million unit market. So we’ve actually done things with interoperability and testing where we try to help them be successful. Because it’s in our interest. We want this market to be successful – a rising tide rises all boats. But individually, they just can’t solve the problem very differently than we do it, and they will never compete with us directly in the area of our specialty.
MHT: Any discussions at Ember about going public?
LeFort: There are always discussions, sure. I think that what people have told us is you need a few million dollars of overhead to cover (the costs of an IPO). When you go through all the numbers, that basically says you need to be at roughly a $75 to $100 million-a-year revenue run rate before you can afford the infrastructure to go public. So that’s probably a fairly good metric for us. If we were running $20 million quarters, then we’d start to look at it. We’re running $9 million to $10 million quarters. We’d all like to have a nice IPO, that would be a great event, but it’s not the focal point. There’s nobody left here after 10 years that is just saying, I’m here for the big payoff, for the exit, for the IPO. There are people here who say ‘I think this is a good company, I think we have potential, and when the time is right, we absolutely should do it.’ But we’re not rushing this – there are definitely no edicts. We just want to stay focused on building the best business we can.
Comments
If you are commenting using a Facebook account, your profile information may be displayed with your comment depending on your privacy settings. By leaving the 'Post to Facebook' box selected, your comment will be published to your Facebook profile in addition to the space below.
RSS Feeds
