Whenever someone comes to me with a request for a CapEx purchase, I always ask the same question first. Do we really need this? I’ve been overseeing finance at software companies for going on two decades now, and I prefer OpEx to CapEx every day of the week. I know I’m not alone in this line of thinking, yet this wasn’t always the case. Here are a few reasons for the shift.
CapEx carries far too many hidden costs.
We all know that CapEx bills itself as a simple proposition. You pay X for a piece of infrastructure, depreciate that infrastructure over three to five years, then start over again. But it’s never that simple. You have to factor in all the costs associated with deploying, maintaining, and managing that infrastructure. In my experience, when you add infrastructure, you add headcount, because you need people to care for that hardware. These hidden costs have a serious impact on your organization.
OpEx lets you focus on what you do best.
Why manage and maintain something outside your domain of expertise or core business when you can have experts do it for you? Our company, Nasuni, is in the business of providing a better cloud file storage platform. We opt for cloud applications whenever and wherever possible because if we’re spending excess time and resources on our own internal infrastructure, then we’re not using those resources to drive the business forward. Maintaining hardware and bringing the technology in-house adds complexity, costs, and headcount. I’d prefer to have as many team members as possible staying focused on our core mission.
Relying on CapEx can slow innovation.
One of our brand-name, global enterprise clients first came to us in 2015 because their business was at risk. They’re in game development, and their competitors were outpacing them, innovating faster. One of the reasons was that they were still stuck in a CapEx model. Their files were siloed on NetApp boxes on three different continents. They saw the potential of a cloud-native, global file system that would unify all these locations, and Nasuni has had a transformative impact on their business, allowing them to reclaim market share and maintain their leadership in a highly competitive industry.
OpEx is more efficient and cost-effective.
The second question I used to ask whenever someone came to me with a CapEx request was how they knew whether they were buying the right amount. Rarely did anyone get it exactly right. The conservative thing to do was overestimate, but then the company would end up paying for infrastructure that was underutilized. Nasuni clients, on the other hand, generally start at a comfortable level, then expand their license by 30% to 40% within the first year. One of our large enterprise clients has expanded from 150 TBs at the start of their deployment to 2.5 PBs today.
The applications are mature.
If you were to look at the evolution of cloud over the last ten or fifteen years, you’d see that the basic applications shifted first. They replaced non-mission critical applications, were easier to deploy and use, and companies could save a lot of money by adopting them, so they switched. The mission-critical apps followed along a little later when companies became more comfortable with cloud technologies and better understood the operational benefits of mission critical cloud applications. Today we’ve turned a corner. Enterprises are not only moving their mission-critical apps to the cloud, but they are trusting the cloud for the most valuable asset in an enterprise — its data. And they’re doing so for the same reasons they embraced expense- or HR-related cloud apps. The cloud offers a simpler, less expensive, more agile alternative.
Cash, Control, and Survival
I’m not the first one to detail these advantages, so why is there still a reluctance to shift from CapEx to OpEx? In the past, financial analysts and investors put a lot of stress on a company’s EBITDA number, which eliminates depreciation and thereby makes your profitability shine a little brighter after a capital purchase. Today the health and prospects of a company are measured more by free cash flow, not EBITDA. In that sense, the OpEx model is far more appealing because you’re not spending all that money upfront. You’re spreading it out in appropriately sized increments over the life of the service and preserving cash.
Another reason I’ve come across is the reluctance to give up control. At my last company, we worked with a major retailer that had spent years building a powerful, consumer-focused commerce experience. They liked having control over the site and keeping it internal, but as the business grew, so did the demands on the website. When business units came to them with requests to modify or add functionality to the site, they recognized that they could not keep up with the pace of change of innovation and were slowing their business down. It was then they realized they’d be better off letting an outside provider manage the platform. Their expertise was outdoor wear, not building commerce applications. And although they were sacrificing the ability to control the fix if and when something went wrong, surrendering that control to a vendor whose sole focus was ecommerce allowed them to satisfy more internal requests and drive even more business.
Finally, in my early days at Nasuni, we used to experience pushback from large enterprises because they feared moving mission-critical files to the cloud. No one gets fired for buying more IBM, they’d say. At the time, in 2015, this was reasonable. Today, some of the largest companies in the world use Nasuni to store, protect, and globally synchronize their mission-critical files, and they’re accelerating innovation, speeding time to market, and driving new business because of it. Now, as my colleague Andres noted recently, relying on hardware is the riskier move and, financially, the less prudent one.