Does Size Matter (in the Cloud)?

By Deirdre Mahon, CMO, Cloud Cruiser


Cloud is the new normal.  Well that’s one conclusion if you listen to the keynote address from the AWS summit in Chicago or the recent MSFT earnings call.  Revenue is up and growth rates are going in the right direction for all stakeholders (or shareholders).  I would argue every company is doing something in the cloud today.  Well, if we’re honest, every consumer or smart-phone user is using some form of cloud service. Many smaller or mid-size companies are seemingly more innovative when it comes to the rate of cloud adoption and I can only conclude this is for two obvious reasons: less bandwidth in-house and faster time to deployment — or as we marketeers like to say “value”.  Bigger enterprises are certainly adopting, but like big liners, they move a lot slower and it often feels like cruise control speed.

Regardless of size, if you peel back the cloud adoption layers inside any single company, the types of services being used can vary considerably.  The easiest and most likely “first-step” services tend to be for the purpose of development and testing, or perhaps using the cloud for back-up of an on premise environment or seasonal storage, including archival.  The economics, scale and speed make these types of services “no-brainers”.  The next hurdle most cloud users overcome is leveraging the growing range of development tools to build new applications and environments from scratch.  Collectively known as PaaS, this can span big data databases such as MongoDB, and perhaps some application performance monitoring tools like NewRelic or AppDynamics.  After a deeper level of maturity and comfort settles-in, running production apps in the cloud is where you reach a point of complete reliance on service levels for uptime and security.  This can be running custom-developed applications or out-of-the-box solutions such as Office 365.

Regardless of where you are on your cloud journey, no doubt you have faced increasing challenges as adoption levels grow across the business with users consuming different types of services and perhaps from different cloud providers.  As usage expands, your expenditure grows and as with any budgetary investment, you need to prove value and a worthwhile return to the business.  According to Nucleus Research published in April ’16, the cloud ROI multiplier – the relative return on investment delivered by cloud applications versus on premise ones –is on the rise. According to case studies published in the past few years, Nucleus found that cloud application projects deliver 2.3 times the ROI of on premise ones – up 24 percent since 2012.  With the added benefit of ‘going green’ and saving on environmental factors, cloud is an easy decision.

Once you make the strategic decision to move to the cloud, getting there is not so easy.  Many hurdles must be overcome, such as which cloud is best suited to which workload, at what rate and how to institutionalize policies and governance for all users.  There is no such thing as a free lunch, even if it’s cheaper than the former on premise legacy approach.  Important questions quickly come up – how do you get visibility on exactly what is being consumed?  How do you get ahead and start to forecast? Unfortunately, the trusty spreadsheet becomes the main tracking method, which before long proves inadequate. As soon as you “hit the wall” and can’t proactively figure out who is using which services inside the business, how much of them and at what cost – you call in the experts and try to get ahead before services come to a screeching halt.

Many who have gone down the cloud road already would say – “it’s about the size of your cloud investment”.  While that’s an important measure, I would argue other factors come into play which can cause much confusion and, worse-again, inability to prove value if left unwatched.  Size of spend is certainly important, but let’s face it, size is relative. An organization that spends ~ $500,000 annually on public cloud services might consider it valuable if they are operating a revenue-generating business leveraging those services.  To another organization, this level of investment might be excessive and may not directly impact the bottom line, so it is therefore harder to measure and demonstrate any concrete return.

As a first step, you must determine why you are using the cloud services and what outcome you expect from that investment.  Faster time to value is easy to claim, but unless you actually meter and measure as you go, you will have very few data points to report on.  There are many other metrics to track — such as specific rates and how they vary across vendors — but more importantly, planning for the actual units or volume of those services, which has proven quite difficult to do.  With cloud, you don’t really know until you start using.  With a historical perspective, the future certainly becomes easier.  Tracking historical detail is critical in order to get control.

Below are the top 5 analytics you need to proactively manage your cloud services and determine value:

  1. Cloud spend by account, department, subscriptions…
  2. Cloud usage by resource, project, geo…
  3. Trends over time, seasonal spikes
  4. Budgets & forecasts by usage and spend
  5. Resource inefficiencies such as under-/over-provisioned resources

Cloud is easy, fast and certainly valuable.  But if you can’t measure it, you will quickly have a problem.  Those problems will come in the form of a business owner or finance leader asking “Why is our cloud bill going up every month?”  Or “Why are we using all these VM’s? I don’t have any visibility into who is using and for what project.” And, most importantly, to improve business planning: “How much should I budget for next year on these cloud services?”

If you can’t come up with clear answers, it really doesn’t matter what your cloud size is.