Custom Viz is a fast emerging trend in the big data space. Data visualization teams in Silicon Valley are discovery the flexibility they gain by adding Custom Visualizations to their toolkit. This often means reframing the way one thinks. Instead of “how do I translate this requirement into Tableau”, we now think “what is the best way to deliver this insight.”
See my previous post on changing mindsets for more.
Viz engineers are no longer in a box, forced to use CSS stylings, controls, and navigational patterns that came off a shelf past its due. Suddenly, we have dozens of options when rendering something as simple as a Date Control:
- A simple slider
- Your humble form input field
- An input with a pop down calendar
- Multiple inputs with pop down calendars
- A multi level menu with sliding selectors and predefined date ranges
- A cascading menu
Now imagine these options paired with 100’s of styling themes. The final look and feel of a simple widget can have infinite permutations.
With great power comes great responsibility.
The Facebook Ads one above is my personal favorite. It is intuitive, has frequently used predefined ranges, and it’s pretty.
Given all of these options for a date control, how do you choose? What about navigation controls, menus, radio buttons, drop downs, validations, state, etc etc. Now you’re overwhelmed and stuck in analysis paralysis. Normally, your BI tool of choice has already defaulted all of this for you. While convenient, they were designed for broad and generic use cases. As a result, they are often unintuitive in many contexts.
Time to start thinking like a Product Manager! Notice I didn’t say “think like a UX or UI expert”. In most cases you will be developing a custom viz by yourself just the way you built Tableau dashboards. If you need a dozen people to build a custom viz, you’re doing it wrong. Choosing the right date widget is one of many design choices you‘ll be empowered to make. You don’t want to pigeonhole you’re mindset as a UI, UX, or Data Viz expert. You need to think about the whole lifecycle of your custom viz as if it was an App.
This means you need to understand your audience, your go to market strategy, the value proposition, on boarding, roadmap, time to MVP, and audience conversion. All of this in addition to UX, UI, and of course analytics.
Why does this matter? Dear friend, you are now in the wild world of modern web development. Once mastered, it’s liberating relative to your BI tool. As such, what should now constrain you is what constrains any decent Product Manager: the real world.
Ajo Abraham is a big data expert with extensive background in building beautiful big data applications and visualizations. For consulting requests you can email him here ajo@veroanalytics.com
