While the adoption of cloud computing in the enterprise offers all sorts of benefits, too many organizations are forgetting about risks. Yes, the benefits of cloud computing are very real. I’m certainly not a cloud denier (and they, like Flat Earthers, do still exist), but one costly misstep as you move to the cloud could quickly eliminate all of those benefits you were hoping for.

I’m not talking about security risks. Those have been dissected almost as thoroughly as the cloud’s benefits. Rather, I’m talking about business risks. If you move a mission-critical application to the cloud, what happens if you don’t have enough available bandwidth to meet demand? What if your headquarters now must suffer with the poor performance that your branch office managers have been complaining about for years?

What if your Internet service provider’s network is compromised? The service provider will invariably tell you that redundancy and failover capabilities are built into their systems, but should you trust them?

Recently, two severed fiber optic cables in Sprint’s network grounded Alaska Airlines for much of a day. What are the chances that two different fiber optic cables in two completely different regions of the country accidentally get cut on the same day? The chances aren’t great, but they obviously were greater than zero.

The risks don’t have to be that severe, however. Access to critical applications is essential to knowledge workers. If your ERP, CRM or SFA application performs poorly because you’ve moved it to the cloud, what have you gained other than headaches?

Legacy apps drag down cloud efforts

Even as we’ve started the full-scale migration to a cloud-centric world, most applications aren’t ready. Most applications were not designed to reside away from your office, outside of your firewall and in the cloud. Even newer versions of software that has “cloud ready” slapped on it are often not cloud ready. Sure, the software can be hosted in a cloud environment, but can it perform at a high level in that environment?

Remember, cloud-based apps must now perform over the chaotic public Internet, rather than over your tightly controlled, high-throughput LAN. Latency, dropped packets and congestion can all seriously degrade app performance – routinely.

To overcome this problem, deep-pocketed organizations invest in expensive MPLS lines, WAN optimization equipment or CDNs. These solutions aren’t cheap. Remember all of that money you hoped to save when you moved to the cloud? How quickly it can evaporate.

If this has you thinking that you should just go back to the on-premise status quo, I have more bad news for you. It’s not going to be possible. Many software providers are moving to a services model, meaning you won’t be able to host your own versions even if you want to, and, meanwhile, organizations of all types are becoming more and more decentralized.

Outsourcing is no longer a trend, but a mundane reality. Telecommuting is on the rise. Even traditional office workers often work a day or two a week from home. The workforce is becoming more and more mobile. And newer businesses, in order to compete with incumbents, are avoiding the overhead of big office spaces, instead opting for small headquarters, smaller branch offices, and many mobile and work- from-home employees.

In other words, the cloud is shifting the cost center away from apps (which will be consumed on a pay- as-you-go model) to networks, which are consumed on an overprovisioning one.

Putting your head in the sand and hoping that the old ways will suffice will simply let more nimble competitors catch you, pass you, and pick your carcass clean during your bankruptcy.

The cloud may be the problem, but it’s also the solution

Fortunately, cloud computing is not a static innovation. If applications can be delivered as pay-as-you- go services, why can’t network services be delivered the same way? Why can’t you avoid purchasing expensive WAN optimization equipment or signing up for expensive CDN contracts, instead consuming each as needed?

The answer is simple: as in any other established space, the incumbents resist change. WAN optimization vendors make money off of that expensive equipment. MPLS providers want you to buy those expensive links. CDN providers thrive on those expensive contracts, which are all the more juicy when you under-utilize the services.

However, that doesn’t mean that change isn’t coming.

These services will move to the cloud – they already are starting to – and the incumbents know it. Many will even tell you that they are actively preparing for this new reality, but until they face enough competition from either startups or some underperforming incumbent needing to shake up their business model, the market leaders will be slow to move. It’s just the way these things work. After all, Saleforce.com, AWS, Dropbox and NetSuite weren’t overnight hits, and they faced serious skepticism on the road to success.

How to prepare for the future

In an economic climate as cut-throat as this one, where a startup in India could just as easily threaten you as one in Silicon Valley, it’s not enough to simply embrace change. The winners will help drive it.

If your cloud-based applications are underperforming, don’t believe the experts when they give you a limited list of options, all of them expensive, all of them more suited to yesterday’s technologies and work patterns. Network virtualization is happening. It took a while for server virtualization to mature and move out of the science-project phase. Now, every business under the sun understands the importance of server virtualization. Fortunately, the adoption phase for network virtualization should be much shorter. After all, it’s a heck of a lot easier to be the second group traveling over the Oregon Trail than the first.

So, the first thing you should do is simply not accept the status quo, and distrust anyone who tells you there is no other way.

Second, if vendors try to sell you expensive hardware or lock you into inflexible contracts, push back. Remind them that everything from expensive CRM suites to engineering testing and development has moved to the cloud.

Why haven’t application delivery services? Why haven’t network services? Why haven’t WAN optimization and CDN services?

Chances are the vendors you’ll work with already have these things on their roadmaps anyway. Why not force their hand?

Finally, negotiate for a win-win. In the end, embracing these new delivery models will not only be in your best interest, but in theirs too. After all, if they drag their feet and insist on stale business practices, you won’t be working with them for very long because they simply won’t be around to work with.

Resisting change in an era of innovation is the risky choice. It’s a cutthroat, winner-take-all economy we’re in, and those who are too risk-averse are actually the ones who end up courting risk. It’s a cruel irony, but your less risk-averse competitors won’t have much sympathy as they steal your market share.