I have grown a little tired of all the articles proclaiming the benefits of the cloud. If you don’t know those benefits by now, then frankly – where have you been? Saying that the cloud is cheaper than on-premise has become accepted knowledge. Saying that the cloud is more reliable than on-premise has more or less become fact. So why aren’t we all in the cloud already? Why are software companies still offering on-premise, and why are some still choosing on-premise over cloud, despite ‘obvious’ benefits?

This is by no means the backlash against the cloud, or a retro statement – but the argument is far from as clear-cut as the writers of the pro-cloud articles often suggest (and they are, let’s not forget, often cloud providers themselves). When might on-premise software be better than cloud-based? Why would anyone still do it?

Control & Certainty

Cloud problems certainly get plenty of attention. UK banks will know full well that down time has not just a financial cost, but equally a long-term reputational one. And they can’t avoid using some form of cloud. Those organisations who have the most sensitive data are often those who are not yet fully convinced of the control they can have over their data held in the cloud. Yes, they will be the ones with their photos on iCloud, but they’ll equally be the ones who retain the strongest grip over their own data.

There is a good degree of certainty over data held on-premise, which is one of the major reasons it is still offered. Pam Baker sums it up nicely in her post when she says that legislation and regulatory environments often hold people back from the cloud. Add to that the statistic that 42% of cloud projects are brought back in-house and the case for on-premise increasingly looks like one built around the weakness of cloud security.

Previous Investment In Infrastructure

Warren Butler’s post on cloud and on-premise CRM underlines an often forgotten element – the cost of prior investment in technology infrastructure. Imagine a situation where infrastructure are in place, and a significant investment has been made already – simply turning them off in favour of the cloud is not always a sound business call, and a business may want to ‘sweat its assets’ before turning to a cloud-based solution several years later.

This comes back to TCO – in this case, the TCO will potentially be lower, given the prior investment – making on-premise the more financially sound choice. Infrastructure can also equal resources – if there is a workforce in place to handle these structures, with internal knowledge, then it can make sense to forego the cloud in this instance.


Reliability is a moot point. There is plenty of evidence to suggest that cloud up-time is better than on-premise up-time – so how could reliability come into it? There is the reliability of spend, for one thing. Cloud suppliers are free to hike up their prices whenever they like (unless you stipulate contractually for them not to) – and on-premise negates that issue by providing a fixed basis. And there’s the reliability that comes from ownership – if you have the internal resources and knowledge to handle the systems and keep them up, then that’s more reassuring – for some – than externalising your data.


And finally, a key reason why many stay in-house: integration. While many of the big cloud providers have got much better over the last few years at integrating with, well, almost everything – internal systems, ERPs and CRM software need to talk to each other. There’s no one-stop shop, and integration is often easier when the solution is on-premise and there is a technical team to handle it.

We’re not all in the cloud because we’re not all ready. The cloud isn’t this wondrous solution that solves all your problems – for many organisations with existing structures that are hard to shift, it presents uncertainty and extra cost. And the most reassuring solution is on-premise.

In this case, moving to the cloud is a long-term project. We’re all fully aware of the cost benefits, and if the stats bear out, there’s less downtime in the cloud than with on-premise. It’s a five-year plan. It’s a phasing out of current structures, and a phasing in of a leaner way of working. We’re not all there yet. And who knows, maybe on-premise will become a fashionable retro statement one day. Right now, it’s still good business sense.