Dec 5, 2018
| 4 min read

Podcast #38

Connected Industry VC Perspectives – A Conversation with David Mount

In our conversation with David Mount, parter at the VC firm G2VP, he shares some of his experiences working with companies in the energy and industrial sector. He discusses some of the unique characteristics that lead to success, the interactions with key customers, and the need for VCs to adjust their perspective to customer success when working in the industrial sector. He also compares and contrasts how and why certain industries have been moved faster and been more successful in digitalizing their businesses, and discusses the potential unleashed by AI, machine learning and other technologies. Finally, he highlights notable startups in the space including Kelvin, Alea, Zomi, Xage and Particle

 

Recommendations 

Principles by Ray Dalio  

AlphaGo (film)  

 orange-line.png

 

We'll notify you weekly about new podcast episodes, upcoming guests, and news. You can subscribe to the podcast and if you'd like to be considered to appear on the podcast contact us.

 

View Transcript

Good day everyone, this is Ed Maguire, Insights Partner at Momenta Partners, with another episode of our Edge Podcast, and today we have as our guest David Mount who’s a partner at the VC firm, U2VP, and we’re going to dive into some of his thoughts around connected industry, his background, and his use of the market in the future.

David, it’s great to have you with us.

Thanks Ed, great to be on, and I appreciate you having me join the podcast.

Fantastic. First, I’d love to get a sense of your history, what in your experiences have really shaped your view of Connected Industry or IoT?

I’ve been working in the Connected Industry or IoT sector, for the best part of 10 years now. I got started as an investor at the venture capital firm Kleiner Perkins, when I started which was July 2008, the topic of the smart grid was en vogue, and we wanted to look around at which companies were going to bring connectivity to the grid, connectivity to metering and power, the power industry particularly. So, got to know a whole host of competitors in that space, and that ended up with an investment in Silverstream networks, which eventually went public, and eventually got sold onto Itron and remains a part of Itron solution package today.

That was at the beginning of this journey around the connection between new communications devices, combined with processing, combined with some industry specific solution that relates to this whole notion of Connected Industry that I’ve been excited about ever since, for whatever reason that extreme solidifier! It led to subsequent work with companies like OSIsoft which is a data infrastructure for time series data, and other data that’s used to connect to industry, Opower which was a residential energy efficiency programme that was sold to the utilities, and eventually became a part of Oracle, Element Analytics which is a business that I cofounded with Sameer Kalwani, around collecting and cleaning-up datasets that were going to be used in large IoT analytics projects, eventually ServiceMax which is a field service software business that was helping take the work that was being done by field service technicians on clip-boards, and put that into an iPad-type environment so that they were sure to have the right manuals, the right safety checks, and all those things as they were doing their work.

Another one is called Kespry, a drone company that’s doing aerial analytics and taking sensor data, and image data, to be able to run reports and do diagnostics in industries like insurance. A couple of others, Telogis which I didn’t work on as directly, but they did fleet telematics for large fleets around the US, it was eventually bought by Horizon, and then Kleiner Perkins was involved in the investment in Nest, the connected thermostat company that was also formative as I thought about the Internet of Things.

More recently, a team of us, four of us that were at Kleiner Perkins as a part of that Green Growth Fund for about 10 years, and the four of us have started a new fund called G2VP; it’s focused on companies that are applying emerging digital technology like connectivity, edge processing, artificial intelligence; all these themes that we’ll talk about in this podcast generally talks about, as they pertain to traditional industries; so, that includes energy, manufacturing, supply chain, logistics, transportation and agriculture, the industries that make up more than half of the economy, but don’t get as much attention from traditional venture capital.

So, that’s where we’re focused, that’s where I’ve been focused over the last 10 years, and exciting about continuing to focus on for the next several decades.

That’s really interesting, those are a lot of very familiar companies, in fact recently George Mathew at Kesprey was a guest on the podcast, so it’s really fascinating.

I’d like to go back a bit to the origins around smart grid, because you’ve mentioned Silver Spring, and there are a few other companies that were competing at that time. As you look back at the initial expectations around smart grid, and then some of the realities of the market and the challenges of implementation, Silver Spring is certainly one of the more successful players in that market, but there’s Converge, EnerNOC, and a number of private companies; what are some of the lessons that were learned through the process of the initial hype, and some of the challenges of building businesses based on the vision of a smart grid? A lot of it is demand and response there, but more broadly as well.

There’s demand response and distributing under resource management, and eventually there’s ancillary services, and a whole host of different things that were promised around the smart grid, a lot of lessons learned there. It started with the beginnings of AMI or the smart grid, was about bringing communications to meters, and the value proposition there was pretty crystal-clear where in the old days 15 years ago, not that long ago, most meters were read either by a person who went to see the meter themselves, or by rolling a truck past an automated meter that could send a signal to the truck as it rolled past, in terms of what the bill was going to be, what the consumption was going to have been in the last month.

It was pretty clear that utilities could make a compelling case to install wireless communication in those meters, to be able to get reliable meter reads, consistent meter reads, and install those programs, and that’s what they did. Then there was the promise that once those meters had gone in, then suddenly we would want to install all sorts of other equipment, including distribution automation equipment that would allow substations, distribution level transformers, and other devices on powerlines to be able to communicate for optimal grid performance, optimal grid reliability, and all sorts of other things. Frankly, whilst we as investors were really excited about the prospect of some of those things, they didn’t pencil out as well as some of the other used cases, and unless there was an absolutely compelling totally bullet-proof case to be made for why the equipment would go out, and why the equipment would go out with automation, it didn’t make sense.

Additionally, there was a lot of perception or concern that additional telemetry, potentially additional devices out on the network could create reliability issues, and lead to increased interruptions and increased frequency of those interruptions on the grid, and that risk was not something that a lot of the utilities wanted to tolerate. So, it turned out to be more of a go-to market problem I think, than it was a core technology problem, there are a lot of technologists and investors and others who got excited about an optimal grid, but maybe lost a little bit of track of exactly what the compelling business proposition would be for it, and how you’d be able to get an economic return on it.

So, really in your view it was more of a business execution challenge, as well as clearly the technological and regulatory hurdles of course, not insubstantial.

Yes, it was a combination of all those things. There was probably more investment in the sector than there should have been, there was a signal from the market in 2010-2011 during the tarp phase, and there was a smart grid investment programme that helps pull a bunch of demand in, and I think that market participants may have viewed that incorrectly, viewed it as a signal that ‘Wow!’, suddenly there’s going to be $500 million of annual interest in buying this stuff, so let’s gear up and get ready for it. Then it turned out that was two years of demand that was getting pulled up by some grand incentive programs, and the marketing didn’t keep up on that growth path. So, that led to some business challenges as well.

Given your experience, were there any significant events or people that you worked with out of those earlier startups in your career that have had a big impact on shaping how you view the industry?

I think some of my favorite and most formative meetings that happen are from groups who are customers for these connected industry technologies. So, whilst at Kleiner, and now whilst at G2VP, one of the things that we focus on is building relationships with big industrial companies, and those can be big industrial OEMs like ABB, or GE, or Honeywell, or they may also be relationships with big industrial buyers like the large tower companies, the large manufacturers, the pulp and paper manufacturers, the food companies. Often they’ll send delegations that are interested in learning more about Silicon Valley technology, and I love those meetings because I always ask, ‘What’s holding you back from buying more connected equipment?’ or, ‘What are you looking for in IoT based product or service? It’s always a good reality check for maybe how my enthusiasm, or investor enthusiasm can be a little bit ahead of the reality of the buyer in the market.

I would say one of the lessons learned from the smart grid work was we realized the importance, and it underscored the importance of really knowing the customers’ perspective before you make investments, and that the perspective of a Silicon Valley venture capitalist can be very different than the perspective of the end customer for this industrial tech. So, what we really like to do is to get to know the industrial customer base. There was one meeting in particular, we had a Fortune 30 company come in, we start the meeting off and I say, ‘Okay, here’s our background, we love connectivity in sensing, and storage, and analytics and machine learning…’ he literally stopped the meeting, he said, ‘Dave, stop. We’ve been on this trip through the Silicon Valley, meeting with five or six companies that are all saying the same buzz-words, and I need someone just to show me the chessboard’.

It was that quotation, ‘Show me the chessboard’, that gave me a very good perspective, led to an hour long conversation where we talked through, ‘If five or ten years from now I want to be a smart  connected company, what is that going to look like five or ten years from now, and then what do I need to do between here and there, to go get it?’ It gave me a real appreciation for that long-term thinking, long-term perspective a lot of these customers have, whereas folks in Silicon Valley can maybe have a short-term perspective around, ‘What are we doing this month?’ ‘What are we doing this quarter?’ the customers who are buying this are thinking about these buying cycles with longer perspective, and that’s really important.

So, that concept of, ‘Show me the chessboard’, is something that we now think about for customers that we’re engaging with, and the start-ups that I work with we think of that concept of, ‘Show me the chessboard’, where when you go in and talk with the customer you can’t just be talking to them about what you might have, you need to really understand their entire landscape, and help them think through not only the move that they’re making now, but the move that’s three, four, five steps ahead. That’s one.

A second one is the founder of OSIsoft, Pat Kennedy, is a terrific entrepreneur, been in the segment for 30+ years and has been providing solutions for tens of thousands of customers, and tens of thousands of installations of connected manufacturing, and connected energy data the whole time. One of his lessons was, focus on doing the work to show users their data, and once you get them their data, they will know how best to use it. Another one of the things that we think about is, trying to enable the companies that we work with, trying to enable the industry participants, rather than never showing up and pretending, or thinking you know their industry better than they do. That’s another important point.

Maybe a third from the founding days of Element Analytics a few years ago, working with Sameer Kalwani, for Sameer it was always important, he thinks like a true entrepreneur who just wanted to get in knowing that at the time at least there were a lot more platitudes than there was real work getting done. Sameer was always very focused on, ‘Well, let’s go in, and let’s just start getting some work done, and we’ll figure out where we can automate a process, where we can build something that’s scalable and replicable, as we go’. But that was another one, ‘Rather than study the IoT, why don’t we just get it and start doing work, then we’ll figure out where the bottlenecks are, where the challenges are, as we go’.

One of the interesting points that you just made was the time horizon of your customers, had I guess, a much longer duration than their perception of venture capital. But compared to for instance public company investors, the time horizon for investing is actually much longer, you guys make an investment, you have to be able to anticipate three, five, seven years out; I’d be interested to get your perspective in working at Kleiner Perkins which has made investments in all sorts of sectors, but is really well-known for its successes in information technology and internet. How does the view of investing in industrial-related technologies relate or contrast to some of the principles related to investing in traditional software and internet, and some of the areas that are considered more the VC Wheelhouse for Silicon Valley?

I’ll start by saying we’re now at G2VP which is a separate industry from Kleiner. I can share some of those thoughts about how thinking has evolved around investing in industrial tech, versus investing in IT from the days of Kleiner, but they are separate and Kleiner is now largely focused back on that IT enterprise technology investing, and our team that had been focused more on this industrial tech has split-off to do this out of a separate fund.

So, five or seven years ago I might have started that answer by saying, ‘Well, okay let me explain to you the traditional rules of venture investing’, things like use limited capital to ruthlessly mitigate risk, and find technologies that have high margins with low capital intensity so that they can scale really fast. Find exceptional teams that are able to not take no for an answer, or disregard when people say no to them, and find ways to just make their industries work. Those are kind of traditional venture capital rules. Again, five years ago I might have then tried deposit that there were different rules for investing in industrial tech, but I think that’s wrong. So, I would say the basic rules of venture applied both in traditional IT investing, as well as industrial tech investing, and to the extent that we try to bend those rules or thought that the rules for investing industrial tech were different, we got it wrong.

As we look at new investments today in companies that are serving the industrial sector, or that are serving manufacturing logistics supply chain, agriculture etc., we’re looking for those same characteristics around management teams that don’t take no for an answer, around ruthlessly mitigating risk with as limited a capital as possible, and all of those rules, because companies that are facing the industrial market still have to scale, still have to find ways to grow, still have to find ways to grow with margin, they have a different type of customer base, I think we may focus a little bit more on the go-to market. We focus more on finding industrial partners to work with our start-ups sooner, so that they can provide feedback on products, so that they can provide potential channels to market, so that they can provide potential customer relationships. But all the rules I’d say I learned from the investing days at Kleiner continue to apply in industrial investing, and again, historically to the extent that we thought that they didn’t, we generally got it wrong.

Another question is the existence of technology hurdles or challenges, that you look for potential investments to be able to overcome when you look at industrial technologies. Are there any unique aspects or capabilities of start-ups that are focused on industrial problems that you look for, that you think are really critical for differentiation?

In industrial tech teams one of the important things is that they reallyhave a clear view of the problem that they’re solving, so they need to know exactly what the customer’s problem is, and ideally be able to speak with the voice of a customer as to why what they’re working on is going to save time, money, effort, and make the buyer a hero. One of those traditional venture capital adages is that only desperate people buy from startups, that’s one that’s been baked into my mind! So that’s, ‘Only desperate people buy from start-ups’, and that’s not only desperate customers buying from start-ups, that’s people. As someone who’s selling into the industrial world, you have to find your customer, your champion, and that champion generally has the opportunity to buy from a startup, or to buy from IBM, or to buy from Microsoft, or to buy from a company that they have had a trusted vendor relationship for a long time.

So, as a start-up you need to be able to go in and offer them something that those incumbents can’t, and generally they’ve tried and failed on something before. The best companies in the industrial market have a very-very clear view of that problem that they’re solving, because if they don’t, even if they have an exceptional technology team, they can flounder around for a long time, maybe too long before they get to that product marketplace. I guess it’s less of a technology hurdle or enabler there, it’s more of a market and customer awareness point, but that’s something we found to be very-very important in the industrial tech companies that we work with.

I’d love to get your take on what we talk about as a disconnector, or let’s say overly ambitious, or enthusiastic expectations around industrial IoT, starting around 2013-2014 when we had some of the large companies that were throwing out some very aggressive forecasts of the number of connected devices, economic value-add, and of course the reality has been much more deliberate than some of the expectations for an inflection. What is your perception of how the dynamics in the industry have played out relative to some of the more aggressive initial expectations?

I think on balance the rod is happening, I would say the transportation industry has exceeded expectations, and the manufacturing and classic process industry have probably lagged expectations, and on balance maybe the market is moving, but it moved in unexpected places. 2013-2014 I would have expected that by now there would have been much more mature offerings coming out of a handful of start-ups that were bigger than they are now, as well as more mature industrial offerings coming out of the big cloud vendors. They are working on them, but as I see it, they’re still in the phase where they’re looking for champion use cases more than they are just simply selling these programs on to industrials.

The manufacturing side and process industry side I think has lagged, because in all the hype of 2013 and 2014, that era, a bunch of vendors viewed the IoT as kind of the next great frontier for selling compute, selling consulting services, and selling $100 million projects, instead of selling ERP, and frankly I think that turned the potential buyers off a bit. There’s already a challenge in this market around who is your buyer, and the way the distinction is made, and you’ve talked about this, and others certainly have around IT versus OT; is it the IT team that is used to buying all the databases, and all the hardware? Or, is it the operations team that’s used to running the plant that’s going to be responsible for buying this technology? That’s a serious distinction, it creates some infighting.

When you start getting organizations thinking and talking about their new ‘platform’, it makes the market nervous, or I think it makes potential buyers nervous, because they weren’t sure exactly how they were going to be able to wade into IoT solutions and start getting quick wins. So, I think the market never really took off, because the perceived cost of adopting IoT was really high. There was a lot of enthusiasm from some of the largest technology vendors in the world around helping big companies deploy these solutions, and they all seemed like they were going to be big clunky solutions, so it didn’t really happen.

Additionally, I think there’s a specific challenge around data quality and the idea again three or four years ago, was just take all of the IoT data that you have, dump it into a data lake, and then we’ll figure out what we’re going to do with it later. That architecture for an IoT project didn’t really work, there’s too much data from different systems, it’s really hard to line up, it’s really hard to then go from that into some sort of analytic and immediately deliver business value. There probably wasn’t enough time spent upfront on what is the business value that we’re trying to deliver here, and more time spent on architectures and getting data moving before the real reasons for the projects were set out. So, that’s on the manufacturing side.

Now, on the transportation side I would say that market has actually moved, and that may have been because of regulation; the ELD mandate came into place for the electronic logging devices, where large trucks were now required to have telemetry in them, or at least to have a data logger in them that tracked the amount of hours that a driver was driving, making sure they were taking the appropriate breaks, not driving whilst tired and other things. That created a whole new market where there’s a new set of data available, now we can do things like optimizing routes, optimizing teams on various lanes, doing all kinds of work around fuel consumption, all sorts of things that telematics companies are able to do.

I think the telematics market has moved very quickly, Horizon made a couple of acquisitions in that space and then built it out, also Sun Star has done really well in that segment. So, there is movement happening, I’d just say that the manufacturing process industry universe hadn’t moved as fast as many would have predicted three or four years ago.

Are there some common threads across either the companies or the industries, that seem to be moving more quickly, whether it be technology, culture, or leadership?

I think generally the companies that are moving on this stuff fast enough, have embraced the concept that every company is a software company, that Satin Nadel is very good about continuing to mention this concept that if you are making a hardware device, think of yourself not only as a hardware manufacturer, but also hardware in monitoring, or hardware eventually in connected services. Or, if you are thinking of yourself as a power company, or an oil and gas company, that data on your production is a valuable resource, and thinking about that from the top down has been important.

I think of a couple of companies that are doing that well, one of which is on the OEM side is Flowserve, a public business that makes pumps and valves, and 5+ years ago decided that they were going to make smarter connected pumps and valves, so that they could provide not only the hardware, but also the awareness of what was going through their systems, to be more of a value-added vendor. They experimented with PTC, and they do work with PTC, they’ve done experiment and work with HP, they publish that work, they share it, it’s become part of their identity as a business, and that’s something that I think has led to some of their success.

BP in the oil patch is doing some of the same work, and I think they’re doing experimentation, they’re committed to figuring out how to create business value from the data coming off of their assets, and they’re beginning to talk more about that over time as well. So, I think the oil and gas industry does seem to be leaning forward on using data, then beginning to talk about it. I think manufacturers are beginning to do the same, some of the industrial OEMs are viewing data as a potential asset, or something that’s going to help them generate more revenue, and those are the types of companies that are leading.

Are there any technology forces or enablers, that you think may have a lot greater impact on potential adoption and business value creation than may be appreciated, or any technology you’re looking at that you feel have a lot of promise?

The table stakes technologies now that I think we would have talked about three or four years ago, are around the cheap communications, the edge processing, and cheap compute that have happened, and those are table stakes, and that’s probably what I was most excited about a few years ago. Today, what I think is going to lead to acceleration of adoption, or surprisingly powerful, or surprisingly valuable use cases, are some of the work that’s happening in open source machine learning, as well as automation and some vision applications. So, the gains that have been made in open source machine learning, and in some of the vision libraries and in some of the automation work that’s happening, its moved really fast.

Jokingly we talk with computer vision experts, or automation experts, and you’ll say, ‘Hey, we haven’t caught up in three months. How are things going, what’s new?’ and they’ll sort of say, ‘Everything’s new, it’s been three months’. So, the open source libraries for doing things like failure detection, or if it’s not failure detection it’s looking at images and being able to recognize things in those images or being able to do processing on a phone as opposed to a supercomputer, or an Invidia graphics processor. The advances in those capabilities are moving very-very fast, so every 3-6 months you get surprising results from people who are talking, who are using some of those most advanced compute paradigms, libraries, and that’s really fun.

The second one I think is controls in automation, so one thing that we haven’t really talked about here is, what’s the solution? I feel like this discussion has been more about the challenges of the last few years, more than looking forward to how and why greater adoption is going to happen down the road. I think one of the keys to unlocking that future value is again, knowing exactly what your use case is, knowing exactly how you’re going to create value, and then doing it, and I believe that’s going to come with controls. So, the typical IoT installation, or at least the typical IoT concept five years ago was, let’s go put sensors out into the field, or let’s collect data coming off of sensors that we already have, let’s organize them properly, then connect them into some analytics, and figure out when a device might fail, or figure out how we might be able to optimize this system. That’s great, but typically that also then adds data to operators who may already be overwhelmed by the amount of data that they’re looking at.

I think the next step once all that work is done is to say, ‘We’re going to take what we know about what’s going on in these assets, and we are going to begin to optimize the performance of these assets using automation or, using the machine learning and control algorithms that are being developed with the help of that data. Once that happens, and there are a few companies that are doing that now, I think that is when the real value of the IoT is going to be unlocked.

Maybe one of the points that I didn’t underscore earlier as to why some of these projects have had challenges, I think that the teams that are working on deploying this stuff are really strapped, they don’t have a lot of time to be experimenting on new tech. I think if vendors are able to go to these customers and say, ‘Tell me your problem, I will solve it, and then I can deploy the solution and continuously make sure that those problems don’t show up’, I think that will be a big win, and that’s what I think is going to come in the future of delivering value, around IoT based solutions.

You mentioned a lot of open source machine learning, I guess things like TensorFlow and a number of other tools that are out there. As you look at potential applications, what are you excited about, what do you think the potential can be when we start to see full application, full realization of some of the potential? As you mentioned, I know it’s evolving really quickly, but as you think about where we are today with some of the solutions and companies that you’re working with, and where we could go, what gets you most excited?

I’ll give a couple of examples, one is a company called Alea Edge, it does work in water. They’re working with Google on some of this stuff, they have a camera that they can set up to look at readings from various types of meters, they’ve been able to train algorithms that can look at those images to determine what is going on, on a water meter that currently does not have connectivity, or that is buried, or has never been put onto a water distribution network before. They can tell whether those meters are working or not and help those water utilities that are planning better investing allocation, and better revenue collection. That is the type of business that is now cost-effective for the vendor, and valuable for the utility in ways that it never has been before. That came in a couple of months ago, that is one of those where that solution makes a ton of sense, and it’s doing stuff that was not possible two years ago based on where technology was.

Another is a business called Kelvin which is a startup business that is doing some work in distributed equipment monitoring and some controls, where they are helping coordinate systems of gathering assets or other assets in oil and gas, to be able to get tens of percentage points of efficiency improvements, whilst reducing emissions and other unwanted work stoppages by using some of these coordinated controls, some of these capabilities like TensorFlow has for other things.

So, both of those are examples of businesses where there’s a clear value proposition, there’s new technology that’s being used, and a problem that’s being solved in a way that wouldn’t have been possible three or five years ago, and that’s the type of stuff that we’re pretty excited about.

Any thoughts on the application of blockchain?

We think about that, and typically as we’re thinking about blockchain the question is, ‘Why does it have to be blockchain to make the solution work? Why couldn’t it work in a database, a secure database, an encrypted database?’ So, that’s the general filter that we use to think about blockchain. I do think there are a couple of potential use cases where blockchain may be very helpful, or in fact required, related to potentially supply chain, and chain of custody around supply chain as goods are transferred between parties. Power settlements in the power market in particular to do with renewable energy, or renewable energy credits as those become more and more important, as led by some technology companies that wants to buy lots and lots of renewable electricity; and potentially things related to the ServiceMax ecosystem where you have service records on assets, and maybe you want to transfer service records between owners of those industrial assets, that’s a place where blockchain could be relevant. But we have not yet seen companies that are commercial or doing a lot of commercial work yet. We’re keeping our eyes out but haven’t seen that yet.

What are you optimistic about, how do you see the markets that you focus on evolving over the next five to ten years?

I’m optimistic about a lot of things! My general view is to be optimistic, and I can kind of be a true believer in these markets, which is dangerous in some ways, but it is true, it’s just the reality. Over the next five or ten years I see a great proliferation of a lot of these underlying technologies, I think we are going to see OEMs that build equipment worth more than $5,000 a unit will incorporate IOT solutions into their solutions by default. That will mean that there are subscription types of offerings, or service types of offerings, or consumables monitoring types of offerings that go into every piece of equipment that goes out. It may or may not be turned on all the time, but the option for a customer to buy those products as a subscription, or to have monitoring as subscription, will become table stakes, I believe.

I also believe that there will be more data moving from existing brownfield systems today, up into the cloud, for more sophisticated analysis. As more data goes into the cloud I think there are going to be some great opportunities for combining datasets between systems, so for these bigger projects you want to be able to take data from the sensors, and then combine it with data from the maintenance records, combine it with data from the ERP transaction log system, combine it with a whole host of other systems, and I think more of those combinations will happen and there will be value that accrues from doing that.

I also think there will be this ongoing platform battle that will continue. I think there are some big companies that are committed to this market that I don’t expect will go away, because I think the market will continue to grow and mature; that’s companies like Microsoft, Google, Amazon, and IBM, and well as companies like Accenture, and some of the other system integrators that are going to be playing a big part of this market, and helping create real industrial enterprise grade offerings in these markets.

What keeps you up at night, on the flipside are there some concerns or obstacles that you see, moving forward?

I want this market to move faster and that’s what keeps me up at night, I’m always clapping my hands thinking, ‘Okay, let’s move, let’s move’, and trying to keep patient about that. The more time I spend with customers for these types of products, the more I understand why they’re taking a metered approach to it. But at the same time, what I really want to see is some companies do well by adopting, and adopting more aggressively, that means companies who have revenue gains from implementing IoT, companies that are showing real benefit to not only the bottom line, but the top line from adopting this stuff. So, that’s something that I want to see more of.

The other thing that keeps me up at night I guess, is this notion of there are few vendors who came on strong looking for $100 million projects, and I think got some players in the market thinking that beginning the IoT journey meant that you were beginning to sign up for a $100 million project. I think that did the industry a disservice, and I hope we can move past that and we can start to see a playbook that is more like walk before we run, get some quick wins, test and iterate the classic venture style movement towards these IoT based solutions, rather than more the way that ERP has bought and sold movement into IoT solutions.

That’s great. This has been super-interesting, I just wanted to ask since you’re looking at a number of interesting companies, you mentioned a couple earlier and I was wondering if there are any technologies or notable startups that you’re keeping your eye on.

Yes, lots of those. I did mention a couple, Element Analytics is doing the work on taking those big jumbled datasets and making sense of them, turning them into asset models and combining them with other datasets, it’s a business that I’m very fond of as a founder. And then another company is called Kelvin, I mentioned this doing some of those controls, Elayer that’s doing work in water meters and vision. In security which we didn’t talk about a couple of companies there that are doing great work, one is called the Zomi, another is called Xage that are both trying to make sure that as more and more of these IoT devices proliferate that security vulnerabilities do not, and finding ways to make redundancy and devices turn into a safety feature, as opposed to a security vulnerability, which I think is really important, it comes up with a lot of potential customers; we’re not investing in any of those, but they seem to be doing great work.

Another one is called Particle. Particle is building a kit and developer framework to help more and more devices hook into the IoT, they’re similar to the way I think that Pridex is thinking about or was thinking about building their developer community. Particle has built a very clean simple way to get data, or to turn an analogue device into a smart connected device, and they seem to be getting a lot of traction with their developer communities, so that’s another one that we’re watching. There are lots more, but those are a few that come to mind.

That’s super-interesting, and auspicious as you mention Xage, because I believe at the time we’re having our conversation, Momenta just announced an investment in Xage, so that’s nice to hear that independently.

And I had not seen that, I had not seen that!

It literally hit the tape earlier today. Finally, David, the last question I also like to ask is a recommendation of a book or resource for our listeners that you might be able to share.

A book that I’ve read, really liked and have reflected on a lot this year is Principles by Ray Dalio. As we build out our firm and think through working as teams, trying to grow and evolve, Principles is a terrific book, and Ray Dalio has done a bunch of podcasting and supplementary materials that have gone with that, and appreciated.

A movie that’s worth watching that’s maybe a little bit more relevant here is AlphaGo, the book about the Google DeepMind team that took on creating a program that could be the human grand masters in the game Go. It is a terrific story, the movie is on Netflix, it’s a great documentary movie and it also touches on a lot of these themes around advances in machine learning, advances in parallel processing, and the potential implications those have on automation.

Those are great recommendations. I’ll have to check out AlphaGo, I’ve been reading Principles, dipping into it all year. Ray Dalio has got such relevant wisdom, that’s a terrific recommendation. I think you’re the first podcast guest to recommend it, he’s got an incredible story.

One of those adages that I really like and use here is, it’s the simple adage to, ‘Embrace reality and deal with it’. So, as we think about startups in companies, and opportunities, you’ve got to always try to be clear about what is actually happening in the market, and then respond to what is actually happening as opposed to what you think might happen, what you wish would happen, what you’re nervous about what might happen, those types of things; just trying to be as clear as possible about where are we right now, and what is the best most opportunistic way to move forward from here.

Absolutely. It’s been a great conversation.

Again, this is Ed Maguire the Insights Partner here at Momenta with another Edge Podcast, with our guest David Mount, partner at G2VP. David, thanks so much for taking the time to speak with us.

Thanks so much for having me.

[End]