CRN Interview: Juniper Networks CMO On The 'Huge Opportunity' For Partners To Specialize In IoT


For Mike Marcellin, CMO of Juniper Networks, it's still early days when it comes to channel partners working in the Internet of Things space, but for those that break through and learn how to specialize and scale their practices, there are "huge opportunities."

"Where our partners come in is any time there's something that's very nascent and you're dealing with IoT organizations that are flat budget, not a lot of know-how in a given space. To me, that screams huge opportunities for partners," Marcellin said.

CRN recently spoke with Marcellin at the IoT World conference in Santa Clara, Calif., where the CMO of the Sunnyvale-based company talked about Juniper's IoT strategy, where channel partners are finding traction in the space and how companies should handle data privacy issues, the latter of which has been a major focus with the European Union's enacting of the General Data Protection Regulation rules

"I suspect at some point there will be a GDPR-like thing that hits the U.S. If you go to China, the attitudes are very different about willingness to be surveilled and just all that," he said. "That tells me there is no right or wrong per se, but it does say that you got to do right by your customers and your users."

Sponsored post

What follows is a lightly edited conversation between CRN and Marcellin.

What is the state of Juniper's IoT business and strategy?

I'd say there's two big areas we're focused on, I think both of which actually have good applicability to our partner community, too. The first one is around security, which is obviously huge for any IoT deployment at scale and just embracing of it, whether it's smart cities, industrial IoT, connected cars -- the trust has got to be top notch for any level of [deployment]. And obviously at Juniper, we're going to play at the intersection of networking and security. We do networking, we do security, so that's a good intersection as you're having an IoT conversation.

Then the other one is, there's an emerging kind of architectural transition that's happening, and it's actually interestingly happening on two parallel paths that I think are going to come together nicely. One on the 5G side, as carriers do 5G, and then one on IoT, but they're both a similar transition, which is kind of the concept of edge computing and putting as many compute resources out as close to the users or the end points as you can. Because we have spent a lot of time over the past five years helping some of the big hyperscale cloud providers get off the ground -- we're infrastructure providers to them. And so, large centralized data centers and clouds are great, but as you start to then say you need low-latency, high-reliability connectivity for a variety of applications, then you start to say, "well, the speed of light alone is going to get in the way then," so push it out closer to the users. I do think […] the IoT and 5G convergence over the next couple of years is going to be an interesting one. Not necessarily that it's going to be the exact same solution. Just if you talk to the [telecom companies], they're talking about, "we need the killer app of 5G, it's not selling more handsets anymore," so it is maybe this connected car, maybe it's a smart city type of application. Where our partners come in is any time there's something that's very nascent and you're dealing with IoT organizations that are flat budget, not a lot of know-how in a given space. To me, that screams huge opportunities for partners.

They can come in and really help companies or cities or whoever that entity is get something off the ground because they're looking for answers. The good thing for us is we can really package up these solutions, I think, in a very consumable way, so it's not like partners [have] to figure out how to necessarily piece it together. If they want to take it on themselves, great, but we can give it to them in a pretty consumable way, and they can then wrap on the services piece. [In a study we commissioned with the IoT Institute], they said as of today, the vast majority of IoT implementations -- it's pretty early -- but the vast majority, I think it's like 70 percent are done by internal IT organizations. They're doing it themselves. Then the same question was next two years, what do you think it's going to be? And they are all saying they want it to be a managed service. They want to leverage partners to help them do that. I think part of that is because they find out it's hard, but also it is an ecosystem play, and that's another place where partners can really sing, because otherwise, if I'm a CIO or an IT organization, I've got to walk around this and figure out what I'm doing. Whereas if there are partners that really specialize in the space, they can take that on and come to me with some answers and some recommendations and help me manage the ecosystem, so I think it's a huge opportunity.

Does a single channel partner working with a customer alone have what it takes to handle the customer's breadth of needs for IoT? Is there room for collaboration among partners?

It's a good question. I think there could be some that get there or at least say hey, a lot of our partners do specialize in certain industries or whatever, so they might say, "I can do all of the primary applications for hospital or for whatever," wherever they choose to specialize, so I think it's possible. I don't think I can point to someone who has got it all figured out, but I do think it's possible. It is an interesting notion that I think I will take back to think about, because we do have an opportunity among our partner community to foster our own ecosystem, if you will, [and] to say, "look, if someone over here is really specializing in this application or this vertical, someone over here is doing the same," can we foster some learnings and sharing of that?

What would you say is driving adoption of IoT for Juniper?

I put myself in the customer's shoes and what we hear is, there are a few things. There's almost always a financial incentive. Financial reason why you're doing what you're doing. Now the payback period is going to be quite long. I was looking at one and I know the payback period is 13 years. This was a smart city [project]. [For a] smart city, not publicly traded, maybe you can get away with a 13-year [payback], maybe if you layer on another application on top of the same infrastructure you can shrink that, but even with that, you gotta have a business use case that proves out, so I think that's a part of it. There is also, I'll call it, quality of life or quality of operations, depending on if it's city or if it's an enterprise. Some way to improve the operation for their employees, citizens or whatever it is, and so that comes through automation, that comes through much better visibility, much better delivery of whatever service you're talking about, much better safety and security. I think the third is a little bit interesting, which is just, there's so much buzz and focus on this that I think most companies and municipalities and cities are saying, "I need to take a look at this." If you're a city or even if you're a company, if you fall significantly behind key tech shifts, then you're not going to be as attractive as a place to live, place to work. You're not going to be able to get the top talent and so that is a factor as well, which is kind of hard to quantify, but I think we all see it. You go around here and you see there's so many companies here, I think part of the reason why they're here, and some of them are very advanced, but some of them are trying to figure out what all the hype is about and where do I start? So that's an opportunity for us to help them on that journey and for partners to help them on that journey.

In a way, it's kind of like when the World Wide Web first started.

Juniper was founded right in the thick of that. That was the first business problem that we solved, which was the scaling of the web. Everyone was jumping on it, but maybe they didn't know why because a competitor was doing it, and the internet was just buckling under the growth. I think we can see a similar thing on either the network side or the security side. I showed a slide in my [recent] talk about all the headlines about all the breaches. I mean, we hear them every day so there's interesting factions around, "IoT's a thing, gotta go gotta go gotta go," and then you have this other faction saying, "hold up, security is a huge concern here," especially if you hear about these breaches that came through the lighting system or in the cities you hear about things happening to critical infrastructure, which is huge, so we got to be very mindful of this. I think we've got some interesting approaches that mean you don't have to stop everything, but I think the point is you have to think differently than you used to think. You're talking about millions and billions of end points rather than, "I got three buildings and I've got 10,000 employees, I can get my arms around that," whereas it's you all these end points that number in a much larger number and much smaller footprint. You can't necessarily put heavyweight security endpoint software on a sensor because it just isn't built for that, so our approach is that you have to assume the network has been breached. If you come in with that assumption, then how do you think about security? That's your assumption.

With regards to the Cambridge Analytica data privacy controversy, do you think the tech industry is showing enough restraint when it comes to how we build these things, what kind of data do we collect? Are we putting enough safeguards in place?

I think the interesting thing about this is that I think we all are in for a societal calibration on what we believe is personal data, what should be shared, on what terms it should be shared. [With the European Union's General Data Protection Regulation coming online], we're all very focused on that. I suspect at some point there will be a GDPR-like thing that hits the U.S. If you go to China, the attitudes are very different about willingness to be surveilled and just all that. That tells me there is no right or wrong per se, but it does say that you got to do right by your customers and your users. That's the end thing. I call it a calibration because I don't think anyone could come out and say this is what it should be, because there are obviously a lot of people who are willing to share a lot for the benefits they get. The other people are like, "hold on, never knew I was sharing that," so I think it's a bit of a conversation calibration, like maybe different levels of that and different people opt into certain things. I think it is about opt-in, it's about transparency, and it's about that calibration that will look different for different companies and maybe even different countries.

Do you think this calibration is the result of things like Cambridge Analytica?

I think it's multiple things. GDPR for me was the bigger thing. That's been a few years in the making now where, thinking from Juniper's shoes , we sell network infrastructure and we help customers build cloud environments. And then there are certain countries in Europe that say, "well, your data can't traverse country boundaries." That was one element of and actually even before GDPR that was something that some European countries mandated. And so that made our customers who were building these infrastructures think differently about, "hey, I can't just put a big data center in Amsterdam to serve all of Europe," for instance. So they had to think about, "how do I do that?" And I was thinking well, while Europe is multiple countries and the U.S. is one country, geographically -- and you could argue, if you look at some of the parts of the country that haven't different opinions on things, we're very diverse as well, even within one country. I've been thinking for a long time that it is, how do we calibrate it, and it's not going to be one-size-fits all. The Cambridge Analytica thing is just another thing to get us thinking and figuring out what's going to be the right answer or set of answers.

In some ways, everything that these applications and connectivity power is not necessarily any different than we have always done, but the difference is scale. I can sit here and talk to you and tell you about my weekend and where I was and everything like that, and if I'm willing to do that, but if I post it on Facebook , now many more people can see it. Or if Google has tracked what I've done this weekend and somehow they use that information now , they can potentially use it in many different ways. My hope is that we do have the right conversation and dialog, we don't rush to legislation by legislators who maybe don't understand, and similarly, we need to hold companies accountable for transparency and having the honest relationship with their customers.

How are channel partners doing with IoT offerings through Juniper? Is there a lot of traction, or are they still trying to figure out how do they build on top of their traditional offerings?

I would characterize it as early. I think about our 20-year business with these partners. We're at the early stages of this. I mean, we do have some early customer success, but I would say that even within those it's not like I've seen one partner knock down 10 wins if you will or opportunities around that. It's been scattered around opportunities that have arisen. Partners have stepped up to address those opportunities, so I say it's definitely early. The big focus for us is, like I was saying earlier, about being clear on the use cases we go after, then being as clean as we can about packaging that. In some cases, we'll lead that conversation with the partner. In other cases, maybe they're already headed down the path and they just say, "hey Juniper, what do you got?" And we slipstream into what they're already doing, which is great. I'd characterize it as pretty early, but the interest is growing because the opportunities are growing, and they're seeing that, so they're turning to us, and I'm sure others, to ask, "what can we do to help?"

Can you give an example of how an IoT deployment can come with both financial incentives and quality-of-life improvements?

One of the interesting ones is Ireland's Marine Institute. It's basically this organization in Ireland that has sensors underwater and they're monitoring marine life, their natural resources and there's a little element of coastline monitoring as well, a safety and security element of it. So that's a really interesting one where that's quite costly to do if you're sending divers down, so there's a completely obvious business case to do it in that way, although it's an upfront investment. [But it's also about] quality of life for the environment, but also for the citizens as well. We're involved in smart cities and other things too, but this one I thought was a very unique use case. I also like the fact that we've got Juniper stuff deployed at the bottom of the ocean. I can't think of one [deployment] where you don't some element of both [financial and quality-of-life incentives]. The financial element is kind of the table stakes to get the deployment and funding approved, but then ultimately, it's delivering that as well, some type of quality of life. It could be for a company trying to increase efficiency of their employees and eventually bottom line.

Has the channel program evolved in how you're talking to partners?

I would say probably not for IoT specifically, but where it has evolved are a couple key dimensions. One around the cloud, and cloud plays a big role in IoT deployments as well, thinking about where your application and monitoring sits and how you communicate with an endpoint back into that. So we've done a lot around that infrastructure because that's a huge opportunity for us. The other area is automation, so helping our partners understand how to automate infrastructures becomes huge in an IoT environment where you're dealing with such a huge volume of endpoints and how you automate the monitoring and managing of that. So we've done that. [For security, we're] talking about using the entire network as a detection and enforcement mechanism, which is particularly critical for IoT.

What is the IoT-related company you're most excited about, right now?

I think IBM is doing some really cool things. They've had this vision for a long time. I really appreciate someone when they're a leader and they're stepping out. I won't say they made IoT the thing, but they were one of the shapers of how we think about this now. So I've been really impressed with what they've done and some of the applications that they've done. And frankly the upfront investment, a little bit of if we build it, they will come. type approach that they've taken, which is a little bit out over their skis and a little bit of taking a risk, but I think it's the right one. Because of that, they're recognized as a leader and certainly there are occasions we have to partner with them, and we value those significantly.

Where are the next big opportunities in IoT?

Where I think the opportunity is: I think we're going to get from that kind of deep single deployment to broad-scale deployments. And that's both scaling a given application in greater number and density, if you will, and that's multiple applications. I think we're just scratching the surface about, "OK ,if I truly have 'everything connected' as a city or whatever, what can that empower for the users, be they employees or citizens or whatever." I think there are probably some things we haven't thought through back to the quality of life thing. Now there is still that calibration of what's OK, so I think that's going to be the conversation we need to have. I think at least the technology will enable us to do some very interesting things to get more efficient, to improve quality of life, to improve safety and security in ways that I think people will be very interested in.