Visualize Multiple Dimensions of Data in 2D Using Marimekko Charts

Readability and information density are often at odds. I have found, however, that the Marimekko (or simply mekko) chart popular in NXP’s strategy group balances both demands. We use it to simultaneously convey competitor market share by segment and the size of the segment

Yes, it conveys three dimensions using only a two-dimensional plot! It does this by varying the width of the bars in a standard stacked bar (column) chart. I’ve usually seen 100% bars; i.e., the vertical axis is percentage. The bars have no space between them, yielding a chart that looks like a fabric used by the designer Marimekko, hence the name. (Honestly, the chart looks more like a Mondrian painting if you ask me, but there’s another chart type called Mondrian that’s more like a treemap.)

Here’s an example of a mekko chart in action for a mythical pharmaceutical industry in which four competitors--Alpha, Bravo, Charlie, and Delta--supply three types of products--pills, potions, and poisons. At a glance we see Alpha is the largest supplier of each of the three product types and pills are the largest product market. Additional conclusions can readily be drawn, such as Charlie is the second largest supplier of potions and does not supply poisons.

We can also pivot the chart if we’re more interested in how each company’s revenue breaks down by product type, as the figure below emphasizes. The data is the same as above, but pivoting facilitates seeing the relative importance of different product types to each company. For example, the above chart shows that Alpha and Bravo are the same size in the pills product market, but the chart below shows that pills are a greater proportion of company sales for Bravo than for Alpha. Similarly, whereas the chart above highlights Charlie as a distant third in the pill market, the chart below makes it clear that pills are greater portion of its sales than they are for the tied-for-first supplier Alpha. Similarly for potions, Charlie is the second biggest supplier (above) overall, and potions are its biggest seller.

As with all data, we should use the mekko charts the same way a drunk uses a lamp post: not for illumination but for support. If we’re Bravo, we may have a thesis inspired by various information that Charlie is the bigger threat in pills than Alpha. The charts above can support this: Charlie may be #3 in pills, but pills are a greater portion of its revenue than they are of the larger Alpha.

If you're a director of business insights or director of strategic marketing, I recommend introducing your team to mekko charts. The holistic view provided by such a visualization tool is essential to making and communicating good decisions.

Making mekko charts is easy in Excel, but not as easy as it should be. They’re a special case of tree map, but (a) Excel only just added tree maps in the 2016 version and (b) Excel’s implementation is half baked. Instead, I recommend the Peltier Tech charting add in (https://peltiertech.com/). Mr Peltier has cleverly hacked the pre-2016 chart types to generate mekkos and other types. He tells how to manually do what the add in does, but it’s an arduous process. Spend the money for the add in and save yourself time (and get some other charting tools).