Visualizing Data or Painting the Pig Organization Strategy
Having worked in this space for a whole lot of time, I have indeed come across many sorts of "Feedbacks", "Comments" and "Improvements" on the work that we do to make executives and the likes, visualize a story in their data. The point being, everyone essentially has their own version of what may be effective and to add to this, these opinions are not easy to ignore, given where they come from. So how do we handle this?
Data visualization is in more than one ways the moment of truth for Analytics. All the ETL, the number crunching, the statistical modelling, the design thinking etc., all of this culminates in this single set of charts that drives home the decision. Does keeping all this work in the boundaries of a Tableau, QlikSense or Power BI for that matter does justice?
Now that we have a glimpse of what it is like to be a data visualization guy, let me put forward my favourite lines up front that Henry Ford once said -
"If I had asked people what they want, they would have said faster horses"
Let's put this up front and build a perspective there off. The definition that my education has taught me spells something like - "A dashboard is a visual display of the most important information needed to achieve one or more objectives that has been consolidated on a single computer screen so it can be monitored at a glance".
Now let's get down to business and look at a few myths:
- Visualization is about story telling - Let's think about this in a slightly different way. "The speckled band" is one of my favourite stories. I read it about 3-5 times at first and later about 10 times more spaced over 15 years. The point being a story can only keep you interested during the pitch or may be a fortnight after it is delivered.
- It's about co-creating with the customer - To this, I agree. But the subsequent question is to what extent? Let's remember the quote and let's remember designing data visualization is a science and not a form of art. Take the inspiration but not the design.
- It has to be jazzy - This is the biggest pain I have lived with. Again, let's look at it this way, I met my dermatologist for some facial infection and while I was at it, I had a full blown make up done (not that I do make-up ever). I agree presentation is essential but jazzy is certainly not the word. I would go more for discipline as the way of representing.
- D3JS is simply criminal - It certainly is not. Once you want to deliver a certain message and the tool you are working with cuffs you then there is just not point in sticking to the tool of choice. The word of caution however is, this requires a certain level of accuracy and a certain level of standards to consuming open source visualizations. Will delve into this at a later post.
- You need Designer and UI/UX Specialists - It's definitely not an overhead if you are on that line of thought. But yes it is a big help if you have the right guy. Now who is the right guy? The answer is you will have to build one.
Let me pause this post here and let this sink in so that we can get back on board afresh. But what I would do is give my vision of what a data visualization is.
As a kid, we all played the game "Super Mario". There was this one brick that would be hidden somewhere on which if you keep hitting, it would keep minting the coins on and on. That was sort of a cheat code. That's what a data visualization is to me. Keep playing with it, spend hours on it and it keeps generating insights just as easy.