BUSINESS INSIGHTS

Jan 26, 2017

Getting Started with Data and Analytics in the Cloud

Mike Cornell Posted by Mike Cornell

As my colleague David Eldersveld pointed out in a recent blog post, doing data in the cloud can seem daunting.  There's a growing list of cloud providers to compare from, and each has their own set of tools, services, and even lingo.  There are also all-new concepts to learn, like infrastructure-as-a-service and platform-as-a-service.  So much to learn and evaluate, and so little time to do so (because if you aren't in the cloud, your competitor is... right?)  I'd like to share some advice that I give every client who is thinking about taking on a cloud data and analytics initiative (and even some who aren't thinking about it yet, but should be), so that they can navigate the hype and challenges of the cloud and have a greater chance of success and adoption.

iStock-539810882edited.png

Think Big, Start Small

We see clients have the most success and best adoption with a cloud data endeavor when they think big, but start small.  I usually refer to it as "tip-toeing" into the cloud.  This generally means to approach the cloud data landscape with big ideas about how and where it can benefit your organization, but to also carve out one or two smaller, manageable, very specific use cases to get started.  It's very difficult to succeed if you are trying to boil the whole ocean or lift-and-shift your entire infrastructure and processes into the cloud all at once.  Below are just a few of the advantages for thinking big, but starting small with your cloud data initiative.

Faster Time-to-Market

One of the general promises of the cloud is that less infrastructure procurement, server configuration, and software configuration should yield faster time-to-market.  When operating in a platform-as-a-service model, you can often be up and running with a Hadoop cluster or a SQL database in just minutes.  Pair that with a manageable (but meaningful!) use case where you are only standing up the services necessary for the use case, and you can likely be up and running with a viable solution in just a couple of weeks (not months).

Lower Risk

Starting with a smaller use case also helps to mitigate risk.  In most cases, you only pay for the cloud resources that you use, which means no hardware or software license commitments up front.  If you spend a few weeks building something in the cloud and you decide you don't like it, you can just tear it down.  In this case, you only incur costs for services during the few weeks you were developing.  This is drastically different from buying and provisioning physical hardware and a big expensive license only to find a few weeks into it that the platform was not right your organization.  

Model for Greater Adoption

Driving interest and adoption for a project or initiative that seems to be taking months and months to implement can be difficult.  People begin to lose excitement and even question the value of the initiative. However, a project that shows results and value in just a few short weeks gets people excited and wanting to do more.  This is also true for cloud data and analytics initiatives.  Building out a smaller, more manageable use case that creates value in just a few weeks can be huge for your bigger cloud strategy. The results of this foundation project can be demonstrated to other stakeholders and business areas which will lead to greater buy-in, adoption, and funding for future initiatives.

Building for Long-term Value

Finally, starting small doesn't mean you are abandoning your big-picture ideas and long-term vision.  In fact, it's quite the opposite.  When architected appropriately, the smaller project is simply laying the pipes (both architecturally and organizationally) for more data flows and use cases.  Nothing you build should be throw-away.  Think about both the cloud services you stand up and the use cases you build out as building blocks that will continue to interact with each other, working towards a bigger, long-term cloud data strategy and architecture.

Let us help

If you are interested in starting a cloud data initiative, but you aren't sure where or how to get things going, let BlueGranite help.  We can work with you to develop that long-term, big-picture cloud data strategy, and help you to focus in on 2 to 3 good use cases to get started.  Once we define a few candidate use cases, we can help you "quickstart" your cloud initiative with a 2 to 5-week engagement where we will implement one or more use cases and help mentor your team in the best practices necessary to continue to develop effectively in the cloud.  We also offer a great 3-day Cloud Analytics Workshop that can help bring your team up to speed on doing data in the cloud with an overview, assessment, and mini "kickstart" proof-of-concept.

Doing data in the cloud doesn't have to be scary.  Remember to think big, but start small, and you will be more likely to be successful in your cloud data and analytics venture.

Exploring RTVS
Mike Cornell

About The Author

Mike Cornell

Mike Cornell is a Senior Consultant at BlueGranite who is passionate about helping clients to solve business problems of varying size and complexity using data and analytics. Mike's specializations include big data platforms, cloud data platforms, advanced analytics, and data visualization and exploration. His technology interests include the Azure Data Platform, Hadoop Data Platform, Spark, R and Python for data analysis, Power BI, and SQL Server. Check out Mike's blog at http://www.datamic.net.