Perspectives

It’s harder to blame a dashboard than an analyst

Even the best technology can fail if you don’t take user behaviour into account

I recently had a conversation with a Nugit user that reminded me of an incident a few years ago. This was when I was leading a project to develop a dashboard for a large retailer. At the time, the client was wasting tons of man-hours every week manually copying/pasting their digital media results into Excel as due diligence on their media spend. Our brief was to help them do away with the spreadsheets, automate all the data wrangling and create reports that would be available to different business units.

We delivered on the brief perfectly, ran training sessions for 30 or so staff and got everyone onboard. The system was working as it should and the client, a senior executive, was happy. No one could see any reason why the dashboard would fail — but it did.

Nobody was logging in. Not the brand managers, not the marketing managers. Instead, they kept calling our teams to request for reports. So this was the question I posed to the Nugit user mentioned earlier: why don’t people use dashboards?

He made an interesting point. It doesn’t matter how well a dashboard is built or what fancy features or impressive capabilities it comes with. The problem is on the other side of the fence. It’s human nature.

In his opinion, when you pull data from a system, the onus is on you to interpret it correctly and communicate it accurately. But when your data source is an analyst—an actual person—there’s security in knowing someone else handed you the report. You get to ask questions, clarify doubts and, if anything goes wrong, you get to blame them.
To a certain extent, we can’t fault them. A lot of data that we’re expected to work with is unfamiliar and complicated. It’s unchartered territory and anyone would feel a lot safer being guided by someone who knows the way.

If human nature dictates that people would rather turn to people, not systems, for help, then we have quite a challenge on our hands here at Nugit. We build awesome stuff that automate tedious, time-sapping tasks. We present the results beautifully and intuitively. We make all kinds of reports readily accessible, anytime anywhere. But are we really addressing what our users need? Can we create a system that plays the role of an expert, holds people’s hands and makes them feel comfortable enough to trust, and act on, the insights presented?

Well, we plan to find out. We’re going to be running several interesting experiments over the next few months, so stay tuned.