Guide to telling stories with data: How to share analytics insights
A comprehensive collection of articles, videos and more, hand-picked by our editors
Storytelling will be the hot new job in analytics in 2015. That was the bold prediction made back in 2014 by Robert Morison, research director at the consultancy Stryve Advisors in Houston. To support his claim, Morison pointed to the unique example of a medical insurance services firm employing a journalist full-time to "tell the story of analytics" -- not exactly overwhelming evidence of the job trend.
As it turns out, Morison was on to something. According to Howard Dresner, chief research officer at Dresner Advisory Services LLC, there's recently been a big push from the collaborative computing vendors (think Yammer, Tibco and Dropbox) to build out tools that include data storytelling features -- and there's been big interest from customers as well. What constitutes data storytelling features? Such features as enabling "scripting of formerly static presentations (e.g., PowerPoint) with live data for more interactive discussion and drill down," Dresner wrote in a recent report. And Automated Insights is doing a different type of data storytelling: using technology to generate earnings reports and sports stories of the type that, until now, have been written exclusively by journalists.
In this podcast with SearchCIO, Dresner breaks down the data storytelling trend and whether these new features need to be on the CIO's radar. Short excerpts from the interview are below. To hear the podcast in full, click on the player.
What is data storytelling? And what prompted its inclusion in your recent collaborative computing report?
Howard Dresner: What if I simply want to send you a presentation and have you go through it and understand a point I'm trying to make? That's where data storytelling can be useful. And many of the business intelligence tools are now including some semblance of data storytelling features for exactly that purpose -- so that I can send you the 'story,' if you will, and step you through the process with a workflow, with some annotations, with some images, with some animation, too, and allow you go through my workflow -- my story -- step by step.
Could you talk about some of the key features you think of as exemplary data storytelling features?
Dresner: We looked at six different key areas. At the top of the list are things like author highlighting: shapes, colors, line art, annotation, etc. [Other areas include] being able to define the navigation or the flows of the analytical objects; allowing users to interact with objects in the stories; being able to control the navigation or have buttons to control the flow of that presentation; being able to play, rewind, pause; and allowing the author to record and synchronize their voice with various analytical objects or flow of the overall story.
What problem is this solving -- what are business users running into that they need this kind of technology to communicate or collaborate?
Dresner: These could be great training tools, first and foremost. But, also, how many people really understand the data? One of the things I talked about forever ago was this notion of information democracy, where everyone ought to have access to insights to carry out the various tasks associated with their job and also to align them with the strategy of the organization.
And that's really hard to do because people have a different level of knowledge and skill as it relates to analyzing data. It's not something we're born with. And so using data storytelling would be a great way to leverage a very knowledgeable person -- an analyst type -- and distribute that more broadly, one, as a training tool, but to also help imbue people with the knowledge they need to help them do their jobs.
I recently heard the CEO of Automated Insights talk about applying their technology to the business world -- taking business data and essentially generating a news story for a marketer or a sales rep. Do you think something like that would catch on?
Dresner: It's sort of like the emergence of autonomous cars, which are really cool. Will there be a market for self-driving cars? Absolutely. However, my guess is you will not find [an automated] Porsche Turbo. Why? Because people don't buy those kinds of cars for the car to drive itself. So, you've got drivers and you've got passengers. I think the same thing is going to be true with business intelligence and analytics.
Where do the CIO and the IT department fit into the data storytelling equation?
Dresner: Certainly, there's some collaboration with IT because IT very often owns the infrastructure and the back-end system and is responsible for the physical data. But somebody has to be responsible for everything between the physical data and what the user consumes.
I think that's a competency center, where you have business analysts as well as the technology folks that are involved. And that competency center could live under IT, depending on its charter, or it might live within the business somewhere.
Consultant Tony Bodoh laid out tips for building effective data storytelling presentations. Read about it on SearchBusinessAnalytics, a sister site to SearchCIO.