F5’s new CEO shares his early impressions, his vision for the app delivery powerhouse
- 28 April, 2017 06:43
F5’s new President and CEO, Francois Locoh-Donou, took over the reins of the application delivery networking company April 3, replacing longtime leader John McAdam who retired in 2015 but stepped back into the position when his replacement resigned in December of that year for “personal conduct matters.”
Locoh-Donou was mostly recently COO of Ciena, a telecom equipment supplier. Network World Editor in Chief John Dix talked with Locoh-Donou yesterday about his plans and ambitions for F5 after Locoh-Donou hosted his first F5 quarterly earnings call (Q2 revenue up 7% year-over-year to $518 million).
Given your history in the technology business, are there any experiences in particular that will guide you in this new role?
A couple of things. There are similarities between Ciena and F5. Both companies emerged in the late ‘90s as high-flying technology companies that became publically traded and survived some of the downturns in the technology space. And they are similar size in terms of employees and revenue. So some of the experiences I had at Ciena will be valuable here.
One is, if you look at F5, we’ve been very successful and we’re going to continue to be successful, but the environment and the way customers consume our technologies is evolving, and we want to be leading our customers into that change and that implies taking some risks, making some bold moves, doing the right development organically.
I experienced significant change in environment and consumption models at Ciena, and lead the organization through those changes. I think that will be valuable.
Ciena also had a lot of build versus buy decisions to make as we grew the company over time, and my expectation is F5, like every technology company of this size, will find itself having to make similar decisions. The discipline with which you do that is very important, and I think my experience of both failures and successes will be valuable here.
And the third one is a huge part of the strength of F5 today -- the talent and experience and the passion in the folks here. At Ciena I learned how to nurture and grow that passion and make sure we don’t do things that erode it. The lessons I’ve learned around the things not to do, as well as the things to do, I think will come in handy.
Regarding the evolution you mentioned, is part of that the transition from your core hardware business to software products?
Hardware to software is one dimension, but if you think about it, the core asset of F5 is actually software. It’s TMOS [the company’s Traffic Management Operating System]. It’s an incredible proxy technology that allows us to make applications run faster, run better, run more securely.
It so happens that the predominant mode of consumption of that technology has been hardware, but what we’re seeing is a significant increase in other modes of consumption. It’s not that it’s moving from hardware to software. It’s moving from hardware-only to hardware and software in a number of dimensions.
Think about service providers, for example, which is a very fast growing segment for F5 [22% of sales, compared to 65% enterprise and 13% government].
Most of the service providers want to buy hardware, and one of the reasons we’re so successful in that space is because of the scalability of our hardware for high performance applications. But we’re also seeing customers want to consume our technology in the private cloud and some of that is hardware, some of it is virtual editions or purely software.
Within software consumption, some customers want to consume our technology in the private cloud, some in the public cloud, some want to buy perpetual licenses, some want to buy subscriptions and some want to be billed by the hour and just pay as they go. Some want to use software to support multiple applications, some want to use our software just for a single application.
The number of ways in which this core technology gets consumed is just exploding. Hardware to software is a part of that, but it’s a much more varied and complex set of use cases.
One of the pleasant surprises I found when I joined is how much the company has already done and the presence we have in private and public clouds, how much of our revenue already is pure software. In the short term, I think a chief portion of my job is to maintain execution and focus, and in some cases accelerate some things we’re already doing to lead our customers through these changes.
I know some software-only ADC players like Avi Networks are having some success. How would a young company like that be able to get a foothold in this mature business?
The way startups typically get a foothold is they focus on a single use case and try to do that very well. My sense of Avi is they are a good fit for a very small number of use cases. The hardware ADC market potentially declines over time, but there is significant growth coming for the virtual ADC market segment and I think a fast-growing market like that can attract new players.
Am I worried about that for F5? No. Because while a new player will get a foothold, our software revenue in the virtual ADC market dwarfs anything they’re doing today, and we’re going to continue to grow in that market
However, in a way I think that kind of focused competition is healthy for us because it points to areas where we’ve got to be moving faster to address these use cases, and we’ve got to be executing flawlessly. I think it’s healthy and it’s a reminder for us that these new use cases are emerging and we need to be awake at the wheel.
You mentioned pay as you go as one emerging consumption model, will revenue for that show up in the product line item on your balance sheet or the service line? How do you account for that?
There’s a short-term and long-term answer. I will tell you that, in the long term I don’t know the answer to your question because we’ve got to do some modeling. With by-the-drink consumption, customers pay for the usage of the license, and maintenance and upgrades and all of that is bundled into a single price because they don’t care about maintenance.
They just say, “I want to use this thing when I need it and you’re going to make sure it works and I don’t need to worry about which version I’m using.” In the end, it’s a single thing they’re paying for so, longer term, I don’t know the answer to your question. Short term it is in our product revenue right now. It’s not meaningful enough to move yet.
Traditionally F5 is revenue is about 50%/50%, services and products. Services is largely maintenance, but a small portion is professional services, like consulting engineers going on-site to help customers configure and deploy equipment. All the licensing for software is in the product revenue.
With you at the helm, what will employees and customers notice as being different about F5?
I have a lot of admiration for John McAdam and the company he has built. His personality is actually one of the things that attracted me to the company. I say that because hopefully there are some things employees will notice as being the same.
One of those things is John’s natural humility, which I think permeates through the organization. F5 is not an organization that has a lot of tolerance for egos and politics and personal agendas and that, I will make sure, remains the same.
I’m hesitant to talk about what people will notice that’s different because of me, because what’s actually going on is this fascinating period of change in the IT industry that we’ve already talked about. Consumption models are different and, because of that, F5 is already doing some things differently and we’re going to do more things differently.
The other thing is, F5 as a company is entering a new phase. We have largely been, I would say, a single product-family kind of company, and we have aspirations to go beyond that and we have multiple propositions in play --- what we’ve done, for example, in security.
Those changes require us to be very good at certain disciplines, or if we’re not very good at them today, to become very good at them. We’re going to be very disciplined about how we invest and where we invest, so that’s something employees will see. We’re going to be very deliberate about how we prioritize our investments.
I want to turn to Network Function Virtualization, which seems to be gaining steam. What does the shift to virtualized network functions mean to F5?
It’s a catalyst for us. In the service provider space, mostly what we’re doing today is on a hardware consumption basis. It’s a lot of ADC and/or security functionality and they need the hardware acceleration for scalability. One of the reasons we’re growing so fast with service providers is because others can’t match the performance at scale we have.
And NFV in the enterprise space is also a catalyst because, typically when enterprises want to virtualize more of their data center infrastructure they go build a private cloud, and when they build a private cloud they spend more with us.
The only thing that ends up being a negative for us is when enterprises don’t know what to do. I think F5 saw that a little bit two or three years ago when SDN came about.
There was a little bit of a pause because customers didn’t know what SDN was going to be and what to do about it, so there was a bit of a delay while they figured out the architecture, figured out, “Is F5 part of my SDN plans or not?” Once they went through that things went forward again for us.
Right now, all the guys who are building private clouds, that is a catalyst for us and they’re using our new iSeries platforms, our virtual editions. Where we get a little bit of headwind is when they pause because they don’t know if they’re going to go private cloud, go public cloud?
Which applications are going to move? Am I going hybrid? Part of what we do is try to advise customers and help them make decisions and make these migrations.
Ok. In closing, let’s shift to a personal note. Tell us something about you that people would be surprised to learn.
Very few people know that my lifelong dream was to be a full-time chicken farmer. I failed miserably at it.
Wow. Didn’t see that one coming.
When I started making a little bit of money in technology I did create a chicken farm in my home country in Togo and we got up to 12,000 chickens. But I must have been a lousy farmer because we were losing money and in the end I shut it down. I think it’s a blessing that I didn’t become a full-time chicken farmer.