As I'll be getting started on a project involving displaying dashboard data on call centre wallboards I thought I'd solicit your experience as part of my preparatory research.

If you've been involved in a similar project what were the main challenges? How did you approach the task and how different was it to a website / mobile app project?

If you haven't been involved in such a project how would you prepare?


I've not worked on one... but I had a chat with somebody who did something similar and the problem she mentioned to me was the temptation to put too much information on the board.

It's a great big space - and there's room to put a lot of information there if you want to. However she told me it was much more effective to keep the message big and simple so that you can pick it up the message from your peripheral vision.

This makes sense to me as you get the same thing with large non-electronic displays. Big simple charts that give a more ambient information displays seem to be most effective.

You might try googling around the topic of "information radiators" for some ideas in that direction.

| improve this answer | |
  • Cheers Adrian. That sounds sensible and corroborates some of the points I read in a good book called "Information Dashboard Design". Now to check out that link... – Rob Feb 9 '10 at 8:43

A few additional tips

  • Make sure you know who is reliant on the KPI's v who has sponsored the project. You will find Call Centre Floor managers can probably already access more info then on a wall board. It is likely that the Call Centre Team Managers / Team Leads will be more reliant on the wall board info as it is probably there only source of call volumes etc.
  • I have previously used change in font, blinking text, move to bold to indicate SLA breaches; this tends to take care of colour blind issues as well as works on older style wall boards
  • Work out how much info the ACD (call centre call routing switch) can provide and if it is easier to set SLA breach triggers on the switch rather than in the MIS engine (I have found some switches that provide SLA breach triggers built in so no need to code in the MIS engine)
  • Look to get the 'true' end user involved as soon as possible and if possible encourage them to communicate with the project sponsor; this will ensure that the sponsor realises that what you are doing will be truly useful to the company.
| improve this answer | |

First bit of advice: usability test the designs, but we all knew that, right?!

Second bit of advice: make sure you find out exactly what drives the department, and nail the display of those KPIs and metrics.

Resources & Preparation

The must read book in this area is Stephen Few's "Information Dashboard Design". It's a goldmine of information, and how-tos, and how-not-tos. Also read Tufte's Beautiful Evidence, as it explores sparklines in great detail.

I would know your technology too. Are you building from scratch, or using a tool (a horrible, horrible tool...) like Xcelcuis? If it is the former, then great, there are loads of fantastic tools like jQuery Sparklines that you can easily leverage. If it is the latter, then know the limitations and prepare for frustrations.

Another useful addition is Color Oracle, which works brilliantly on a Mac (can't comment for the Windows version). It simulates colour blindness on screen, and saves you exporting images elsewhere. It also forces you to rely less heavily on colour, and add in extra stimuli for RAG states.

One final tip: design & preview on the screen/projector the final output will be displayed upon. It can be quite a shock when you transfer from a high quality desktop/laptop screen to a projector/TV screen, as colour disappears and things that looked great turn mushy.

| improve this answer | |
  • Thanks Alastair. IDD is certainly a goldmine. I'll try to get my hands on Tufte's book too. I think we're going to use Navaho's MediaCAT, which I understand is a very flexible system and one that we'll be tailoring to fit. Testing out the designs and drilling into the KPIs was always the plan but I hadn't considered projecting the final output and simulating colour blindness so I'm glad you chimed in. – Rob Feb 9 '10 at 15:51

There's a case study of a wallboard Panic put together here:


It's worth a read as it contains usability insight accompanied by implementation notes.

| improve this answer | |