Case study: Sprout Social opts for IaaS from day one to support on-demand growth

Sprout Social offers a social media management and engagement platform for social businesses. The Chicago, IL-based startup wanted to focus on developing and refining its product, so the founders went with Infrastructure as a Service (IaaS) from the outset. The decision freed the company from much of the IT planning that consumes resources in traditional environments.

Going to IaaS on Day One

Aaron Rankin, cofounder and CTO of Sprout Social (pictured, left), explains why the company chose IaaS. "Early on we were a bootstrapped startup. When you are paying your own money, you don't want to have to spend more than you have to. We did it for the kind of flexibility that it offers. We didn't want to predict exactly what our IT needs would be from Day One, or even a month out. If you don't have to plan something, then it's nice to figure it out on demand — that's what the Infrastructure as a Service offering gave us. We could basically do everything on demand." Another reason Rankin cited for choosing IaaS: the horsepower relative to how much you pay.

Technology solutions

Sprout Social started with Cloud Sites (formerly Mosso). According to Rankin, "Back then, three or four years ago, there was nothing like it with the amount of scale for the cost, and it was turnkey as it is. So it's kind of like a whole website stack in a box, and you don't have to worry about setting up all the components."

Early on, Sprout Social's growth outpaced the scalability of systems, so moving databases onto large, dedicated systems was the quickest fix available. The Sprout Social platform is now based on a flexible hybrid infrastructure from Rackspace. The infrastructure includes:

• Hybrid hosting
• Dedicated servers
• RackConnect
• Public cloud
• Cloud block storage
• Cloud backup
• Cloud files

The Sprout Social platform processes and stores large amounts of heterogeneous data at varying frequencies. By design, the system keeps similar data together, making it easier and more cost-effective to address complex storage and throughput requirements.

"At our company, we have two engineers who manage all of our servers," Rankin says. "They happen to be former Rackspace employees. We do have an account manager at Rackspace, a tech lead, and a few other folks who are dedicated to our account. If we have a serious issue, they get pulled in at all hours. Very early morning conference calls when hardware has failed and that kind of thing."

Better growth through IaaS

Rankin talks about the company's early days, "So, as a technology startup, we weren't thinking too much about infrastructure. We were all focused on the product and getting people to use it. As is kind of typical, we knew we had outgrown the hardware we were running on because the site was consistently slow."

He adds, "It was clear from our own usage of the site, and monitoring reports from users, that we had to grow our infrastructure. Back then, we weren't very calculated about it — it just crept up on us." Sprout Social now has more formal site-monitoring mechanisms and processes in place.

Opportunities vs. cost savings

"We haven't been doing anything to chart our operational savings, but that is something I am starting to think about now, especially in the last year," says Rankin. "Our growth has been great, and our bills have kept up with that. At the moment, we are starting to look at Rackspace's private cloud offering, because it would give us a lot more control. It should actually help us cut our bill quite a bit."

He adds, "The rationale behind us using it [IaaS] was not cost savings directly. If we knew exactly what we needed and procured that hardware ourselves and did a capital expenditure, or even leased it from Rackspace, we could probably save money. But we are on it because we want to change things on-demand. A team of my developers may tomorrow tell me that we need ten machines for something with cloud servers you can literally just rent. We can do that immediately. There is no procurement process or lead-time. Just turn them on, and they are ready to go. So there is, for sure, the business value in that. It's more of opportunity creation than cost savings."

Lessons learned

Rankin and his team have learned a number of lessons from going IaaS first. He starts by noting that "The cloud is not perfect. People love to make cloud puns, like the cloud is always stormy."

Rankin continues, "Unless you are actually using the entire physical machine you are running on and how much it takes to do that because it's not transparent, then you have neighbors. You don't know who your neighbors are, and they don't know you. Their stuff might be misbehaving. When that happens, it will affect you."

"You have to always build your systems to be even more fault tolerant, even more resilient — even more than if you had a complete purview of [what] your hardware situation was and what systems are running alongside other systems. We've been bitten by the cloud so many times, but we've just built for it, and we've embraced some technologies that were built for it — technologies that expect hardware to fail or underperform and account for it on the fly."

"The cloud is not a silver bullet," Rankin notes.

Conclusion

A hybrid cloud platform from Rackspace enables Sprout Social to focus on product development and its core business, while enabling the startup to scale its infrastructure to meet the demands of its growing customer base.