How telling a story can help build the case for data and analytics

We’ve all seen the recent data surrounding the COVID-19 pandemic: the daily updates regarding case numbers and the devastating fatalities. Alone, the statistics appear confronting. However, it’s not until we consider the data as part of a broader picture or newsworthy story that it starts to reveal important and very compelling information:

• How many people caught the virus in one country, versus another?
• How quickly did the rate increase in one country, versus another?
• Is there any other data that may affect the case numbers in a particular country – e.g. the number of tests conducted, or how fatalities are reported?
• Which age groups are most at risk of catching the virus?
• What impact did early border closures or quarantine measures have on each country’s case numbers?
• How likely are we to see a second wave of infections based on experiences in other countries?

Over the last few months I’ve consumed an endless array of newspaper articles which have delved into many of these statistics relating to the pandemic – all highlighting some interesting trends and asking critical questions.

Obviously, the pandemic itself is an extremely compelling story, and in the current climate, people are hungry for statistics and insights. However, it also indicates the value in constructing a ‘narrative’ when presenting data or analytics of any kind.

To illustrate the importance and value of this ‘data storytelling’, I’d like to share a personal example. I feel very fortunate to work at Modis, as we have such a strong focus on community projects – and recently, we have been working with several government and agencies and departments to develop capabilities that may be used to reduce the incidence of family and domestic violence in Australia.

With one particular project, we leveraged analytics to better understand the leading factors of a domestic violence incident: specifically, for high-harm, repeat offenders. This involved analysing a wealth of data – from the 1960s through to 2015 - and we found some very surprising and interesting results.

However, I soon found that when it came to sharing this information with relevant stakeholders, my presentation lacked impact. I spoke about some of the statistics that we had uncovered regarding offence rates, their frequency, the types of offences, their percentage accuracy and even spoke about the algorithms that we used to come up with this information. On their own, the statistics were very compelling – but I could tell my audience’s attention was wavering.

So, I decided to change to a story-based approach . I structured my findings according to a very simple but reliable storytelling structure: time/place + who/what + surprise + message. It went something like this:

Over the last few weeks (time/place), I’ve been working on an important project to help us better understand domestic violence in Australia.

We wanted to see if it was possible to predict the likelihood of a domestic violence offence occurring, and we’ve found some interesting results (who/what).

When we analysed the data from the 1960s through to 2015, we discovered something extremely interesting: weather is really influential! in fact, out of nearly 500 factors we identified that may trigger a domestic violence incident, the amount of rainfall and the phase of the moon are consistently in the top 12 (surprise).

Using this information (with the complete results of our analysis) we can recommend intervention approaches in advance of an event occurring – and hopefully prevent it occurring in the first place. (point)

While this is just one very specific example, businesses of all kinds can very much benefit from data storytelling such as this. On their own, statistics don’t tell us much. However, when they’re shared as part of a compelling story, they can have a much larger impact. For stakeholders looking to drive a new initiative, or convince board members to back a particular initiative, telling a story with data is particularly crucial.

Creating a compelling, fact-based story is not always as easy as it sounds. Here are a few common principles to keep in mind:

• Always use real, everyday language: don’t try to make the story too technical or convoluted. Use normal, authentic terms that your audience can very quickly and easily understand.
• Be short and sharp. Don’t drown your audience in dry, boring detail. Pull out the most interesting and relevant statistics, and focus on those. You’re telling a short story, not writing War and Peace.
• Use visual tools. According to Gartner’s article, ‘visualisation’ is a key part of the story telling process: ‘Data and analytics storytellers must choose a fitting visualisation based on the kind of data they want to present and the audience to which they want to present it. Arranged into a time or conceptual sequence, these visualisations can be shaped into a narrative to help reveal findings, trends or underlying patterns.’
• Know your audience. It may sound obvious, but be sure to always craft your story according to the people you’re presenting to or sharing information with. ‘The sales team may love the story of the gifted salesman who snatched the contract from a competitor with a single well-chosen data point to the prospect’s CEO while in an elevator. However, this will not have the same appeal for the finance team, who wants to hear about predictable outcomes from efficiently executed processes and contract negotiation,’ says James Richardson from Gartner. Put simply, give your data stories a solid premise, a good structure, and make them snappy and interesting. Put the detail in the appendix, and pull out the facts and stats that are the most meaningful and relevant.

If you’re interested in learning more about creating data-led stories in your business, or want to discuss how Modis can do more compelling things with your data, please don’t hesitate to get in touch.