Since its founding just four years ago, Cologix has grown to 20 data centers across eight markets interconnecting some 350 networks. They started focusing on tier two and three market opportunities before it became popular, blending acquisitions with organic expansions to rapidly establish themselves as a national player in the interconnection-focused data center world. Here to give us an update on Cologix’s approach is CEO Grant van Rooyen.
TR: Cologix has expanded rapidly since its founding four years ago. Has your viewpoint on the market opportunity evolved much since then?
GVR: Nothing has changed with respect to our strategic focus regarding the types of markets we like and an appropriate balance between the interconnection franchise and robust colocation for clients that need more than just access and meet-me rooms. We've now done nine acquisitions, and I think that gets most of the attention. But what gets less focus is that we've also constructed a number of data centers ourselves and have dramatically built out the portfolio in support of those interconnection franchises in the eight markets we operate in.
TR: Your strategy thus far has been to acquire its way into new markets and then invest organically to expand from that foothold. Why do it that way, and do you plan to continue on that path?
GVR: Absolutely. The reason that acquisition is necessary in our business plan is that the interconnection sites that are the most valuable and are of interest to us were established a long time ago. They created the density of networks and value to their markets from an interconnection perspective many years ago. You can't just walk into a new market, like Minneapolis for example, and decide you want to be an interconnection franchise, deploy a significant amount of capital, and expect the 75 networks we have to just march across the street for a new meet-me room. If you are an interconnection-focused franchise, which we are, and you want to expand into other markets, in almost every case to get that franchise you have to buy it. That has been our strategy on the way into markets. Once we are in, we look for the invitation from existing, incremental customers to deploy meaningful expansion capital to build out capacity around the core interconnection franchise.
TR: Why do you think we are seeing more M&A interest in data centers and interconnection businesses in smaller markets? Has it affected your approach?
GVR: Folks over the last five years have overlooked what are great businesses in very robust network markets in fantastic economies, whether that be in Canada where we moved into quickly four years ago, or cities like Minneapolis, Columbus or Jacksonville. We have thought for some time that folks have missed out on these opportunities. However, our style is to meet people that own great businesses, build relationships with them, and give them an opportunity to transact their businesses for fair value to us.
We are quite comfortable going out and finding businesses that may not be for sale and convincing them. That frankly has been the primary driver of our strategy. We don't sit back and wait for our friendly investment banker to call us and tell us they have something for sale. The better businesses in our opinion are the ones that are not for sale when they catch our interest.
TR: Do you think the need for closer enterprise access to cloud resources and better content distribution is driving demand in these markets?
GVR: One hundred percent. For lack of a better term, the 'thickening' of the edge of the Internet as it's reflected in the growing importance of what folks would call tier two and three markets is something we see every day in our meet-me rooms and in our sales numbers. Whether it is traditional carriers or newly emergent networks, they are increasingly pushing into the types of markets we operate in, and we're seeing that accelerate. We also have a number of exchanges that sit in our data centers that facilitate the peering fabric. So we see it both on the physical interconnection side in our meet-me rooms and also the exchange side, where we are watching traffic flows increase exponentially.
This acceleration is directly correlated with the fact that networks, content providers and media companies are discovering that these are important markets where they need to have PoPs. It was part of our thesis that these types of markets would become more important and relevant for interconnection and for the distribution of content, and that networks would be forced to better originate and terminate traffic efficiently.
TR: Another new piece of the data center markets that has emerged over the past year or two is Open-IX. What is Cologix’s view of that effort?
GVR: We are a member. Our exchange and interconnection philosophy is very consistent with what the emerging and evolving philosophies of OpenIX are. It's still very early for what they're trying to do, and we are keen observers of that evolution. At a macro level, anything associated with creating greater opportunity for networks and content to interconnect is a good thing. Having said that, we have been supporting open, member-run Internet exchanges since well before Open-IX emerged as a storyline. Today our customers can access four independent exchanges across our platform, and three of the top four independent exchanges in North America (TORIX, MICE and QIX).
TR: Your most recent acquisition of Colo5 in Florida included both complementary assets in Jacksonville and a pure colocation piece out in Lakeland. How does that kind of asset fit with your plans?
GVR: Colo5 was an interesting opportunity for us. A year ago, we bought the interconnection site at 421 West Church in Jacksonville. That gave us the meet-me room and dense network hub, which we like even more now that several promised new submarine cables have come to life, opening up a potential for Internet traffic to bypass Miami on the way to South America. These are real routes that are proving very popular. Once we did the 421 West Church deal, we wanted to build out meaningful capacity to respond to the demand we see in the Jacksonville market. We had two choices: build significant capacity within Jacksonville or see if an acquisition could fill that role. In this case, the Colo5 team had built a wonderful business there with a class-5, hurricane-rated, purpose-built data center. As part of the deal we picked up the building, all the land, strong management expertise, a wonderful roster of enterprise, content, and network customers that called that data center home, and meaningful capacity to expand both space and power.
The Lakeland asset was Colo5’s extension beyond Jacksonville, and they had just opened the site. To some degree it came along with the deal. But the more we understood how customers view Florida and how they view their disaster recovery plans, the more we realized Lakeland, which sits halfway between Tampa and Orlando on the highest point in Florida, was very interesting to our existing customers. So while it wasn't the primary driver for the acquisition, it was a very nice follow-on asset and one that we have a lot of interest in. It's fair observation to call it a bit different business for us, but we're finding a way to make it work very nicely with our Florida strategy.
TR: Are you looking for new acquisitions? Where might we expect Cologix’s next new market to be?
GVR: I've given up utilizing my crystal ball to try to predict the next place we will go to. If you'd asked me four years ago whether these eight markets would be the ones we are in today, I would have been wrong. But we'll always take the tour for assets that are in the marketplace, and when an opportunity arises we like to move quickly and confidently.
Before we started buying these assets, we spent a year making sure that we had a comprehensive view of the assets throughout North America. That groundwork pays off in spades when it comes to moving quickly on opportunities. We're big fans of staying focused and highly disciplined in how we go about building our business, acquiring companies, and most importantly integrating them. It's very easy to buy something; it's a lot harder to comprehensively, thoughtfully, effectively and efficiently integrate them. The source of pride for us is not the buy; it's successfully integrating it and operating it as well as we do the rest of our facilities.
TR: How difficult is the integration of assets like this? What’s the hardest part?
GVR: An area that the telecom industry broadly and certainly the data center and network business has been sloppy on is on making sure that the data and inventory of the infrastructure from a system perspective is at the forefront of the integration. If you don't have it in your system, it's tough to manage it, it's tough to bill it, and it's tough to drive the automated efficiency that one would expect. While integrating a 100,000-square-foot data center with 150 customers may be on one level a lot easier than integrating a 20,000-mile network and all the sites it runs between, if you aspire to have accuracy over your data and inventory, as well as the efficient command and control over your infrastructure, then those integration projects are very complex. But they allow you to build a better business. And we hold ourselves to a high standard. We expect to be billing out of our billing system within a handful of months after doing a deal, and nine times out of nine we have done so.
TR: Thank you for talking with Telecom Ramblings!
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!Categories: Datacenter · Industry Spotlight