Telling (visual) stories within stories

“Graphical excellence is that which gives to the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space.”Edward Tufte, The Visual Display of Quantitative Information

The Ken has been writing longform business stories for its subscribers since 2016. This means we deal with words. Lots of them. Our stories average 2,500-3,000 words. We agonise over their meanings, their transitions and their relevance.

But we also deal with numbers. We’re business and financial journalists, after all. 

It’s not just us though. People love numbers too! Why else would we be so obsessed with the calories we burn per day, the number of ‘likes’ on our social media posts or the daily pollution air quality index? Numbers matter.

That is why we challenge ourselves every day to present numbers from our stories in ways that add meaning, context and interest. Like the words we use, our visualisations and infographics tell their own stories. 

A graphic for us isn’t just a pretty chart that adds to the story—it is a story all on its own. Just as our stories must make sense even if you take out our graphics, I try and ensure that each of our graphics tells a story even if you remove the text it was attached to.

This is why the process of conceptualising and designing them starts while the story itself is being put together. It is never an after-thought. 

Take one of our recent infographics, for instance. It was for a story on Godrej Nature’s Basket, one of India’s most ambitious attempts at high-end grocery retail.

At first glance, the company’s revenues seemed to be rising steadily (except for a small dip in the last year). But when we plotted its revenue growth on the same timeline, you can hear the alarm bells. 

This represents our belief that the visual representation of information is integral to the process of writing deeply meaningful narratives. It declutters information and makes it easier to comprehend. As research suggests, our brains register images within about 13 milliseconds of seeing them and continues processing them even after they are no longer visible.

The journey of information from the raw stage to the final infographic is not an easy ride though. Finding numbers to support your argument is the first problem that any writer faces, but that’s a story for another post. 

Design hiccups

One of the challenges we faced recently was while visualising data for a story on venture capital (VC) funds in Southeast Asia. 

The raw data consisted of funds by 8 VC funds over the past 8 years, meant to be plotted on the same timeline. Our writers Jon and Ka Kay had painstakingly collected the data by skimming old news reports.

The values ranged from $2.2 million to $300 million and none of the VCs had more than five funds, which meant the values on the same timeline would look irregular. The timeline was important because it was an indicator of fund sizes increasing over a period of time. Plotting bars on the same “x axis” was confusing to the eye and led to misinterpretations.

I asked Jon how he wanted to visualise the information he’d given me. Here’s how he drew his idea on a piece of paper.

We normally shy away from plotting time on the “y axis” because traditionally time has always been plotted on “x axis.” Going the traditional way was not an option because our page layout offers a fixed width for charts. 

But limitations are good. Because they lead to outrageous experiments and force you to think out of the box.

So, taking a cue from a radar chart, I plotted something that looked like this.

In this version, the transition in fund sizes over a period of time is visible but each VC’s progress does not stand out. The usual practise would be to designate distinctive colours to each VC but in this case it would mean eight colours i.e. one more than the rainbow. So we chose to go with different shades of green. Needless to say, the colour green represents money.

One might argue that choosing similar looking colours does not help and might make it hard for some readers to distinguish between them. The usual litmus test for designers is to convert the charts to grayscale.

You can see here that a lot of colours look similar to each other, which is why labelling names along with each individual VC’s timeline helps for a reader who may not be able to distinguish between colours. If you are planning to add legends in a separate location your chart, we’d suggest not going with similar looking colours. Also, try not to have more than five legends.

Designing charts friendly to the colour-blind is new to us. We’re ashamed to say, we did not give it much thought until a subscriber wrote to us about it recently. But, we’re happy to say that we are working on it and will get there soon.

By the way, here is the final version of the chart.

Beyond numbers 

Our visuals are not always based on numbers. Or even data that writers send me. Some of our charts explain strategies and business models of the start-ups we cover. 

Like this one:

This chart accompanied writer Ruhi Kandhari’s story on one of India’s largest e-pharmacies, PharmEasy. It explains the business model of three different online pharmacies popular in India in less than 100 words. 

A similar chart was created by writer Pranav Balakrishnan to explain how fiber internet works, based on information that was shared by some of his sources at Reliance Jio.

That sketch does not explain how internet reaches people. Thankfully, the missing information is not new and is publicly available. So Pranav, after a lot of reading and research shared a visual that best explained it.


Another layer of information was added to the above visualisation i.e. the descriptors for terms mentioned in the chart. Here is a look at the final visualisation.

Network diagrams offer an entire class of charts. Like this one created to capture the Agritech funding ecosystem.

Designing a network chart manually is like finding one’s way in a labyrinth. You never know at which turn you may hit a dead end.

(A quick tip: It always helps to plot the chart on paper first. I made the mistake of skipping this step while making this chart, leading to me plotting it over and over again.)

Here’s an older version of this chart. Notice how noisy it looks compared to the previous one?

Leave a Comment