When I was a policy analyst working with data to address issues in public policy, I was so excited to hear that my organization was getting *a certain dashboarding tool*. We saw demonstrations of great visualizations and examples of how companies were using these tools to improve their workflows and revolutionize their businesses. I would finally be able to kick Excel to the curb and make my decisions based on beautiful charts with data that was always up to date.
As anyone who’s been through this experience knows, it didn’t turn out exactly the way that I expected. We bought the tool, installed it, and after six months, we finally had data that was clean enough for a complete dashboard. A dashboard that refreshed once a week. A dashboard that enabled me to make about 1% of the decisions I needed to make on a weekly basis. A dashboard that showed me information that I already knew—info like total sales and customer numbers. A dashboard that showed me how this year’s data looks in contrast to last year’s, without any context of whether that’s good or bad. But most importantly, a dashboard that let me go to a sheet with a detail table and export that info to Excel so I could do the other 99% of my job.
But good news! IT informed us that they began collecting the data we needed daily and would soon update our dashboard more frequently. They would add new metrics and better charts. Awesome! Within a month, they delivered on their promise to refresh the dashboard more often, add new metrics with more contextual comparisons, and create a few new trend charts. This was great because I could then do 2% of my job with that dashboard, doubling my previous efficiency. And now, the spreadsheet I exported to Excel to do the other 98% of my job had five new columns.
Why does this story resonate with so many people? It’s because performing meaningful data analysis is a complex process. That process should be continuous, allowing for users to include, analyze, or discard multiple variables as necessary. Ideally, this process accounts for the consideration of factors that come from diverse sources, whether another system or from the analyst’s personal experience.
Enter self-service tools. Self-service tools have started to become very popular in recent years. These tools allow users to take a governed (or sometimes ungoverned) data model and create their visualizations typically using drag-and-drop functionality. Some tools even allow users to add external data. In our next post, we’ll be talking about how modern enterprise self-service tools, while they have their niche in business productivity, still don’t hit the mark when it comes to effectively driving insights with data. It’s the reason we created an entirely new type of tool, Knarr, to enable users to focus less on building pretty dashboards and more on getting to more insights faster.