You and the 10 Stages of Data Readiness for Analytics


The term “data readiness” means different things to different people, especially with respect to analytics. In this article we’ll take a high-level tour through various stages of data readiness in order for you to evaluate what’s right for you and your enterprise when engaging a new analytics project.

Credit: Pixabay
Credit: Pixabay

If you total up all the stages with which you feel proficient, a score of 7-10 means you’re “ready” for analytics. How do you stack up?

Stage 1: Business Case. The enterprise must first have an idea of what it is that they are trying to measure: does the enterprise have a clearly defined use case for the use of analytics? What business problem needs to be solved? Unleashing analytics tools on data without goals usually will result in achieving little.

Stage 2: Infrastructure Readiness. The enterprise must be prepared to invest in the infrastructure to easily access and analyze data. There also should be an investment in analytical software tools that will enable these tasks. Further, the enterprise needs to make an investment in training, recruiting, and dedicated staff possessing the technical skillsets to leverage the data.

Stage 3: Cultural Readiness. The enterprise thought leaders must be ready to make data-driven decisions as opposed to just using intuition or “gut feeling.” The enterprise should have internal processes and appropriate culture to embrace insights exposed by the analytics team.

Stage 4: Operational Readiness. The enterprise should be ready to measure success and identify KPIs and other metrics to evaluate progress towards the organizational goals with analytics. Further, the enterprise should develop a clear propagation plan, i.e. a way to bring analytics insights to those who need them.

Stage 5: Ingest. If you’re weighing in on your state of data readiness, you’ve most likely recognized the competitive advantage of data to your enterprise. Your enterprise undoubtedly has a wealth of data, but it may be in formats that are inconvenient for analytics. Further, you’re eager to take advantage of your data assets to gain useful insights, the question is whether you have all the data you need. You need to set up your data ingest pipeline so you can work with large, open data sets and combine your internal data with external information to make your insights even more meaningful. Satisfying the ingest stage means you’re ready to build your data lake with the data you need, ready for analytics.

Stage 6: Analyze. Your enterprise is ready to start extracting insights that will affect the bottom line. Predicting the future is no easy task. Without a rigorous process in place, conclusions seemingly backed by data can fall short in practice, costing you money. You may be interested in getting a handle on evaluating marketing campaigns, determining customer sentiment, taking a deeper look at workforce performance, etc. You’ll need a comprehensive background in statistics and predictive modeling to transform your data into reliable and usable conclusions.

Stage 7: Governance. You’ve embraced data as a critical part of your enterprise’s strategy. You regularly collect a variety of data essential to operations, but you encounter barriers to analysis including dirty, inconsistent, missing, and unconsolidated data. Perhaps the security measures of your organization are prohibitive to an otherwise straightforward collection process. Whether you need to improve speed, quality, or security of your data, you need to define a data strategy that works for your unique enterprise. Strategies range from better documentation and workforce education practices, to an entire overhaul of warehousing and analytics processes.

Stage 8: Real-time Insights. Data-driven decisions are commonplace in your enterprise. Your business is scaling, and you’re ready to move from real-time data access to real-time insight. You need to deploy tools, accessing real-time insights, freeing up more of your enterprise’s resources. You’ll need experience in algorithm design that will help transform your existing practices into a data intelligence product.

Stage 9: Big Data. You’ve built your data pipeline from the ground up. You have your pulse on real-time insights. Your enterprise has deployed a data intelligence tool; however, it’s reaching its performance limit with the amount of data you continue to collect. You need to scale. You need experience in parallel distributed computing to take your analytics to the next level.

Stage 10: Data Storytelling. Valuable new insights lose their power if the message isn’t properly communicated to enterprise decision makers. This stage of readiness requires you to have data scientists in place who are experienced in refining the results of analytics into a concise and actionable form. In a very real sense, data visualization is as much of an art as it is science at this stage. Finesse with the end result of analytics means your hard work with capitalizing on the value of your enterprise data assets is optimized.

So, what's your organization's readiness score?

Daniel D. Gutierrez, Data Scientist

Daniel D. Gutierrez is a Data Scientist with Los Angeles-based Amulet Analytics, a service division of Amulet Development Corp. He's been involved with data science and big-data long before it came in vogue, so imagine his delight when the Harvard Business Review recently deemed "data scientist" as the sexiest profession for the 21st century. Previously, he taught computer science and database classes at UCLA Extension for over 15 years, and authored three computer industry books on database technology. He also served as technical editor, columnist, and writer at a major monthly computer industry publication for seven years. Follow his data science musings at @AMULETAnalytics.

Data Prep and Data Quality

Yes, there are ways to minimize the amount of time that data scientists spend on data preparation.

The Value Proposition of Streaming Analytics to the Enterprise

Streaming analytics, which is drawing an increasing amount of interest, helps enterprises by visualizing the business in real-time, cutting preventable losses, automating immediate actions, and detecting urgent conditions.


Re: Let's start at #6!
  • 9/28/2016 4:17:16 AM
NO RATINGS

@rbenci I think it may be the term, "analytics" that leads your clients to think this. Getting into analytics for them is jumping straight into the analysis. I do remember clients giving us a huge pile of data expecting it to be culled, though there was not much effort to consult us about the first five steps.

Re: Let's start at #6!
  • 9/16/2016 7:51:38 AM
NO RATINGS

Yes, as indicated the named stages all are very important for getting to the results. It's always tempting though to skip over one or more to get to something seemingly "more important" or at least easier to acheive.

Re: Let's start at #6!
  • 9/15/2016 7:08:39 AM
NO RATINGS

Good point on the vale of process, and what it should mean. It's telling that a client's data visualization was managed by the IT group.

Let's start at #6!
  • 9/14/2016 1:13:38 PM
NO RATINGS

Most of my clients want to jump right to Stage 6.  Perhaps they already bought a visualization tool, and had someone from IT "play with it" to create dashboards.  Then they wonder why they didn't get game-changing insights!  This is a great list to show that it takes a process --and QUALIFIED analysts/data scientists-- to achieve credible, reliable insights.  Bravo!

Re: 10 Stages
  • 9/9/2016 9:07:45 AM
NO RATINGS

I agree Seth. I like this list too, particularly with governance positioned before scaling as mentioned in Stage 9.  Thnking about governance for analytics involves understanding how data is consumed within an organization and viewing how data is combined - identifying where PII is inadvertently used is a good example.

There's always a temptation in business to scale something up - increase production on a product, raise prices when demand is high. But analytics forces a reasoning on how resources are used.  This description of stages is a great demonstration of why that reasoning is crucial for long term success.

Re: 10 Stages
  • 9/8/2016 4:05:38 PM
NO RATINGS

Thank you Jamescon.  I'll have to remember that.  "Good enough for government work."  

I looked up a quick history on that. "Originated in World War II. When something was "good enough for Government work" it meant it could pass the most rigorous of standards. Over the years it took on an ironic meaning that is now the primary sense, referring to poorly executed work."

Re: 10 Stages
  • 9/8/2016 3:07:22 PM
NO RATINGS

@Seth. Your tale about just getting the project done in time brings to mind the old line that probably is still popular around construction sites. "Good enough for government work." The wall doesn't have to be perfectly plumb, the paint can be a bit streaky, the lawn a bit weedy. Good enough, let's get lunch!

Re: 10 Stages
  • 9/8/2016 1:39:50 PM
NO RATINGS

I over heard someone on transit today saying all that matter was that the project was completed on time, not that it was good. 

I heard the same thing from a friend a few years back working on a project that everyone expected to fail.  

Often just getting it done and worrying if it works later is the definition of success.  

Re: 10 Stages
  • 9/8/2016 1:30:42 PM
NO RATINGS

Absolutely, data injest is often the most Helter-Skelter stage in the organizations I consult with. The reason? The data sources often are the result of years of evolution, not always the most objective process in many enterprises. It is difficult to tell thought-leaders that their prized data assets aren't all they should/could be. This makes injest not the most straight forward process.

10 Stages
  • 9/8/2016 12:45:53 PM
NO RATINGS

These 10 stages could be applied to a variety of other projects as well.   I particullary like the ingest phase and think it is overlooked most of the time.  Often projects fail so thinking what needs to be done if it is a success probally doesn't enter into the minds of many.  If they succeed and get all this great data are they ready to process it? 

INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +