How to find and describe the messages in the data
Posted by Ray Poynter, 16 May 2021
During the analysis and story finding phase, we want to turn data into findings, with clear descriptions. Once we have these clear messages, we can group them, look for the themes, determine what actions should be taken and then craft a story that will deliver the actions.
At this stage, the finding of the messages, try to write clear descriptions of what you find. Try to use few words, use simple words and simplify. For example, you might notice that “47% of undergraduates change their principal topic during their time at university.” Your working description might become “half of students change their main subject”. You will make decisions about precision and detail, IF that message makes it to the final draft.
Don’t Describe, Explain/Synthesise
Consider the following description:
“The man had quite a lot of white wine, Emmental cheese (grated), and Gruyere cheese (grated). He also had a small amount of garlic, lemon juice, cornflour, and kirsch. As well as a large number of cubed bread pieces and he seemed to be getting ready to cook using a large, heaving metal pot.”
Synthesis = “He was making a classic, Swiss-style fondue.”
Instead of repeating the description, try to answer one or more of the following:
- What is the output? (the fondue)
- Why is it happening? (find out why he is cooking)
- Who is it for? (who will be eating the fondue)
Using insight templates
One way to speed-up the process of message finding is to use templates, here are two that are in common use.
Template 1 – simple case
[Persona] [Action] [Thing] because [Reason]
For example, ‘Young patients want a second opinion because they don’t believe their regular doctor.’
Template 2 – with caveat
[Persona] [Action] [Thing] because [Reason] but [friction]
John eats ice cream because it makes him happy but it makes him overweight
You can take the insights from the templates and try to ladder them using the ‘because’ clause as the link, for example:
- Anne wants a second opinion because she does not believe her regular doctor
- Anne does not believe her regular doctor because she is scared
- Anne is scared because she is reluctant to tell all the details to her regular doctor
And so on …
Looking for Why?
Take a key finding and see why that might be happening. For example, we might find that ‘Patients often fail to be compliant, in terms of medication and processes.’ So, we look for other findings that might explain the why, for example ‘Patients are often nervous and so do not listen properly.’
Try using the five whys approach. For example:
- Q: Why won’t the car start? A: There is no electricity reaching the starter.
- Q: Why is no electricity reaching the starter? A: The battery is flat.
- Q: Why is the battery flat? A: Because something is draining the power.
- Q: What is draining the power? The car radio.
- Q: Why is the radio draining the power? A: there is a loose wire in the radio.
You can read more about five whys here.
Finding the lead
Journalists use the lead (the text in bold at the top of an article) to entice people to read the article. Researchers should use the idea of leads to get to the essence of a story.
There is a great example of the lead that was given by Nora Ephron, from when she was at journalism school. The teacher read an announcement and asked the students to write the lead for the story for the student newspaper.
The copy for the article was “Kenneth L. Peters, principal of Beverly Hills High School, announced today that the faculty of the high school will travel to Sacramento on Thursday for a colloquium on new teaching methods. Speaking there will be anthropologist Margaret Mead, educator Robert Maynard Hutchins, and several others.” The teacher checked all the suggestions, and none were good enough. The teacher then explained that the lead was ‘No school next Thursday’. This is a great reminder that the message is not always in the data, it is the implication of the data, the consequence.
Here is a great post about how to find and write leads.