Written by: Phil Goerdt and Mike Fuller
Several years ago, I was working on an enterprise reporting project at a Fortune 15 company at about the same time that cloud-based services were really starting to catch on. During one of our meetings, a colleague of mine on the client side asked me to “explain what the cloud is”. It was a fair question on her part, and I’m confident she wasn’t the only one in the room wondering; they had invested a lot of time and money in on-premise technology and hadn’t yet entertained the idea of utilizing the cloud. I gave her my best attempt at defining the term: remotely managed applications, hardware, compute and storage. For having little experience with that flavor of tech, she seemed satisfied with my answer. This definition does concisely capture what the cloud is about, albeit rudimentary.
This particular organization has since made the move to the cloud, at least in-part, and they aren’t the only ones. Especially in the world of data and analytics, “Cloud” is ubiquitous with terms and buzzwords describing cloud services showing up everywhere: SaaS, PaaS, IaaS, DWaaS, and any other acronym describing an as-a-service offering. Even Serverless Computing is now a thing.
Cutting through all the terms and buzzwords, it’s easy to see that there are many advantages that come with leveraging cloud services. Offloading the administrative tasks, pay-as-you-go pricing models, and simplifying your technology ecosystem are a few quick wins we can achieve with the cloud.
A significant reason we have found hosted services to be advantageous is that our clients typically don’t like paying for analytics consultants to spend hours and hours installing and configuring software. This has traditionally been the case, with long project timelines to conduct installations, configurations, patching, and other set-up tasks. Additionally, these on-premises systems typically require on-going maintenance to stay up-to-date (not to mention up-and-running) but cloud means most maintenance tasks can become someone or something else’s responsibility, right? So is the cloud all it is ‘chalked up’ to be? Can we really free up developer and admin time by moving to the cloud, or is it a classic case of sales > delivery?
Red Pill Analytics recently partnered with a client that chose to use cloud applications for their entire BI stack. They went with Fivetran as the data replication utility, a cloud-based product that replicates data from source applications into a data warehouse. Snowflake Data Warehouse, built exclusively for the cloud, was selected for their target data warehouse. And last but not least, Looker as the data modeling and front-end reporting tool. Looker offers cloud or on-premise options; our client chose the cloud offering to round out an all-cloud BI stack.[Full disclosure: Only light-weight transformations were required for this particular use case so we were able to use a combination of database views and derived tables to satisfy the requirements. We have implemented several solutions like this and have found they are generally acceptable. For a more robust solution, an intensive transformation layer most likely requires an additional piece of technology not mentioned here.]
Given that all three products mentioned are hosted services, there were zero installation tasks and minimal configuration. We landed on site and started loading data the same day the project kicked off. After a few days, we had a good start on the data model and we were creating reports and dashboards by the end of the week. This quick win helped us illustrate tremendous value to the business instead of spending weeks or months provisioning servers, installing software, etc., etc. Back to the question at hand, we typically find with that cloud-based products allow projects to get off the ground faster and limit up-front costs for our clients. In our minds, this is how most analytics and data projects should be.
We liked the idea of leveraging the cloud for analytics so much that Red Pill has teamed up with Fivetran, Snowflake, and Looker to bring you an immersion course on how to do this for yourself. We call it the Jump Program: one eight-hour day of learning and building, from database to dashboard. We think it is a pretty great idea, but perhaps the greatest thing is that it is free to attend. No hidden fees. No secret cover charges. No drink minimum. No presentations for timeshares. Just a group of people willing to share what they know about what is possible with the next wave of analytics. Kevin and Stewart will be there. Representatives from the vendors will be there. So, what are you waiting for? Register here for the Jump Program on January 11th, 2018.
I guess there is such a thing as a free lunch, after all.
