Segment mapI worked on an example recently of four brand measures presented by market segment over time – up to four time periods and up to ten brands. In the PowerPoint, each segment was on a different slide with the figures in a table and colour coding used for significant changes.

I’ve presented this (real, anonymised) data here all on one slide, taking a more visual approach. I believe we learn a lot more from a data visualization which brings it all together like this, than broken up onto different PowerPoint slides.

I’ve taken the view that we don’t want to waste space with content that doesn’t add anything to our understanding of the market. So I’ve removed the actual percentages, although they are there in the form of screen tips for people who want to know, and it’s always possible to add a follow-up slide to a report which contains the actual data.

I’ve also shown significant differences with a black border and stronger colour – no reason to show increases in a different colour from decreases as it’s going to be obvious anyway.

  • I’ve arranged the segments so that the higher value segments, eg people prepared to pay more for a higher quality brand, are indicated and are up and to the right. We can see straight away that brands 1 and 2 have much better perceptions amongst those key people
  • We can see that the brand 4 has an impressive trial ratio compared with its quality perceptions. People don’t think it’s great, but they use it occasionally anyway, while, brand 3 has good quality perceptions but low trial indicating distribution strengths and weaknesses there.
  • With significant differences shown, we can see which segments show more variability over time – the convenience and value seekers, as we might expect, and also the heavy users. There’s little change amongst Willing to pay More, which makes the fluctuations within Reward Seekers, an area where marketing intervention may well make a difference, all the more interesting.

We are only going back a year so can’t see a long term trend, but taking this approach would enable us to go back much further in time if we wanted to, because it frees up a lot more space on the screen.

In my view, one vital component of a data visualization that adds value is the quantity of data. This is about using design to bring meaning to large quantities of data which would be lost if the data were simply presented of itself. Being creative with graphics instead of being bound by what default PowerPoint can offer allows us to bring much more meaning and understanding.