How do you incorporate feedback and iteration into your dashboard development process?

'What actions do they want to take based on the data' is the key consideration here. Above all else, you must understand what real world business improvements should result from your dashboard. You should also strive to ensure that those actions can be tangibly tied to helping the company find customers, keep customers or reduce costs. Once you have figured out the real world actions and their measurable impacts, you can work backward to the technical requirements of the dashboard. If you don't do this and just accept a list of features or data points, or worse yet, an excel mockup, you are at risk of putting a lot of effort into something that misses the mark!

Before launching straight into gathering your data and designing your dashboard, make sure you spend as much time as you need to with the intended audience to get their requirements down to a tee. Don't just ask for a list of points to hit - ask why these things are being measured, what benefit does it bring, is there a better way to do it? Getting your audience to really think about what they want will mean when you do come to building the dashboard, the impact will be felt immediately. Measure twice, cut (or build) once!

Engage your clients (whether they are customers or the next downstream operation) on establishing dashboard measures and how they are defined. If your dashboard is only internal input type metrics such as quantities produced, numbers trained; then you are blind-sided. On-time delivery is more important to your customer than quantity made. Reduction in error rates is more important than number trained. Outcome measures carry the day.

Get this wrong and your users will stick to what they’re most comfortable with - the old long way, which will likely result in XLSX files being shared around the business. Get this right and consult with your users along the way and you’ll stand a far higher chance of success.

In my experience, having the right layout and visuals is crucial to creating effective and engaging dashboards. The layout should be arranged in a logical and intuitive manner, ensuring that the available real estate is utilized effectively. As they say, sometimes less is more, and it is important to avoid cluttering the dashboard with unnecessary information. In addition, when choosing visuals, it is important to keep in mind accessibility considerations. Select visuals that are not only visually appealing but also easy to read and distinguish between for users with color blindness or other visual impairments.

In all intensive purposes ,keeping it stupidly simple could bring in real value the dashboard was intended for. Human brain associates better with pictures - Showing numbers vs chart vs graph could be a consideration based on your audience. Hover over help for more info or tree view if your audience want to drill down to details.

One thing I've found helpful is to test the dashboard with a small set of users first, allow them to find and report any issues which can then be dealt with before a larger release. Nothing will harm confidence in the outputs more than problems as soon as stakeholders access your dashboard. At the same time, test your data sources and connections. Make sure refreshes are planned and executing without errors, updates should be smooth and imperceptible to users. Get the background work right and the user adoption will be smoother and feedback more positive.