The Problem
I joined the Metrics Team in March of 2020 just weeks after the V1 launch of the product. The Metric was designed to give users a simple way to follow a number on a data viz. It had so much potential to be so much more and I was given a team full of ideas but a product that had no clear vision. We charged ahead into the depths of working from home and had a really hard time bringing our ideas together in this new work environment.
I started educating myself on KPI's. Reading college business textbooks and posts from industry leaders. I created a Deep dive document that outlined several of the mechanisms our business customers use to structure KPIs.
Humble Smart
It was clear that my experience building a design language system would come in handy here. What if I could build a KPI language system? What would that look like? My experience working with custom views taught me a lot about the metric architecture because the metrics were built on the custom view engine. I had a lot more to learn though. I had never built a complex authoring experience for Tableau and I had to learn how to create parameters for the Metric to express complex information in a glanceable way.
Jobs to Be Done
As I learned how to turn metrics into the coolest clocks on Earth, I started working on a research plan. The first step was to hold a Jobs to Be Done Assumptions Workshop. I brought the team and leadership together to map out the JTBD at each stage of the KPI lifecycle. We then opened up a conversation about the gaps in our flow and discovered that we were very aligned on the biggest problems but unclear about what comes after that.
Speeding Up
This is when my timeline for vision work sped up. I wanted to do a large-scale user interview cycle to validate our assumptions about the JTBD. Just as I had done for the custom views. Out of the blue, our team was moved into a new org, and the leadership wanted a vision much faster than would allow for my research. The cool thing was that I had done a hackathon two years prior about combining several features and the re-org brought all of those feature teams together. Even better, my research prepared me to create the vision our leaders were looking for.
The designers from the new org held an org-wide design sprint. We interviewed all of the leadership, dev managers, and PMs from the new org and concatenated the interview notes into an org-wide Jobs To Be Done board. At this point, my mind was so full of ideas I was ready to burst. The designers from each area identified scenarios that they wanted to prototype and we all got to work.
Conclusion
In the end, I designed an invention that was much greater than me. This process allowed me to synthesize all of the brainpower of our leadership and the customer input they had gathered with my research and the input I had gathered from the team. When I presented the design sprint findings to the developers they said "Joe I haven't seen anything like this at Tableau in many years." and "Usually when a designer presents something of this magnitude it's impossible to build, but I could cost this right now. This is totally doable."
