Page 92 - The EDIT | Q1 2017

92
Thoughtleader
you will have experienced this frustration which is a naturally occurring phenomenon when you ask your team to tell stories from the data.
The reason being; your seasoned data analyst...
...doesn’t naturally gravitate to emotional human storytelling, they love data and are far more comfortable in the world of data and science.
...sees crafting a story around the data generally as an unnecessary waste of time and that insights or facts should stand on their own.
...typically believes it’s the job of others to see the value of these insights they have uncovered.
The blame for this attitude does not rest solely
on the analyst, as they genuinely believe that a strong data insight should be sufficient to drive
their audience in the right direction and to make a business decision. The disconnect here is that the analyst believes business decisions should be based on logic and reasoning. Unfortunately, in the real world business decision are almost entirely driven by emotion, the decision gets rationalised afterwards. Next time you are trying to get the data analyst team to go that extra mile and to create a story from the data, try to reframe the task with your new understanding of why the data analyst doesn’t see your point of view on what’s required. It’s not that they don’t want to be good at telling data stories — it’s just that their left-brain filter sometimes makes
it difficult.
There is a clear role for leadership to train data analysts into becoming better data storytellers
and creating an environment whereby analysts are trained, mentored and better guided — and given the incentive (and headspace) to bridge the current data storytelling gap.
The journey towards compelling Data Storytelling
So what can you do to improve your (or your team’s) data storytelling ability? Let’s assume you have the necessary technical competence to interrogate the data. Collectively, your team also possesses the business acumen to know what the big picture is and what it is you are trying to solve.
Sadly, there is no magic solution. There is no “plug and play” to create compelling stories with data. It is a skill — and like any skill, the practitioner needs to practise and practise to become better.
Based on a review of existing frameworks, there
are several processes which commonly appear in effective data storytelling literature. One of the better ones is an article written by Daniel Waisberg — an Analytics Advocate at Google entitled “Telling Stories with Data”. In his article, he presents a six- step framework for effective data storytelling:
1. De netheAudience
It is important to start with the audience, otherwise the outcome of the analysis won’t be as interesting to whomever you want
to help. An executive and a data scientist will probably be interested in very different types of information.
2. CreateHypotheses
Without a great idea, it will be extremely hard to get people to hear your story. A great place to learn more about creating good ideas is the comic strip The Shape of Ideas.
3. Sketch
Sketching will help you going from abstract to concrete. When you let yourself think without constraints you will understand the essence of the idea more easily.
4. GetData
It is vital to be detailed and cautious when requesting the data, both the Devil and God are in the details.
5. ExploreData
How do you go about analyzing and visualizing the data you have in your hands? During this step you will filter, sort, group and turn the results into beautiful charts.
6. TelltheStory
Once you have analyzed and visualized the data, choose the best storyline in order to present the results. A very helpful book on the subject is Show and Tell by Dan Roam.
THE EDIT ISSUE 05 | Q1 2017