Using Frameworks to Find the Story in the Data

What is the most important tool in communicating research findings to clients in a way that creates impact? Ask a roomful of people and the answer ‘Storytelling’ will be almost universally put forward, supported by comments about visualizing, emotion, and business focus. However, there is much less advice on how the stories we tell should be found.

Some people seem to have an intuitive skill in looking at data, in rearranging it, and finding the key message that can be crafted into a compelling story that delivers the message from the data to the client. However, in most cases these intuitive problem solvers struggle to teach others their technique. These intuitive people also struggle to produce results when they have to work as part of a team, although they can be quite successful if they head the team – to misquote Carly Simon, you probably think this quote is about you, don’t you!

In my experience, most of the teams who reliably find useful stories in the data employ a frameworks approach. A framework is a systematic process to guide the analysis of information, that helps ensure that no key steps are missed, that people do not jump to conclusions, and that the story is consistent and coherent.

There are a wide range of frameworks being used, and most teams should probably evolve their own version that suits their strengths and needs. However, I think the following elements are essential components.

1. Framing the problem, if we really understand the problem then we are a long way to solving it. Talk to the client, what does success look like and what actions will be taken based on the outcome of the research?
2. What is already known/believed? Too many projects act as if they were an island, they should draw on what is known and take account of existing hypotheses.
3. Assemble the data sets so that they are verified, assessed, and in a comparable form. For example, are translations and transcripts available for each country, has the quantitative data been standardized/normalized in ways that allow the signal to be distinguished from the noise?
4. Find the main story. Too often researchers start by using modern interrogation tools to look at differences between subgroups. If you look at the detailed picture first it is much harder to find the big picture, and much slower.
5. Find relevant exceptions to the main story, where relevant means relevant to the client and their business question.

During stage 5, findings that are uncovered can be put into two broad categories a) relevant and important, b) not relevant but important or useful. The b) group should be communicated to the client, but not as part of the story. For example these can be sent to relevant people inside the client or can be delivered in a follow-up report.

After the message in the data has been found the next two steps are the crafting and the communication of the story.