Structuring and Sharing Findings
Market research reporting usually contains one or both of the following:
- A report that indicates what the data means. Most commonly this is conducted in a PowerPoint-style document, except when preparing reports for government agencies in which case a Word-style document is more common.
- A data dump, which, as the name suggests, is a large amount of data so that the user can draw their own conclusions about what it all means. These are commonly delivered by using:
- Decks of crosstabs, typically provided as PDFs, Excel or Word documents.
- Interactive dashboards.
- Data files, so that the user can interrogate the data as they wish.
Typically, where only a data dump occurs the recipient of the data dump will end up writing their own report.
There are a number of common poor ways of structuring research findings:
- According to the order of the questionnaire.
- Like a 'story', with lots of bits of information, building to a conclusion at the end.
- Thematically, with similar types of information grouped together (e.g., demographics, brand usage, attitudes).
- As a pyramid, with the key bits of information at the top, less important information underneath, and so on.
Ordering the reporting by the structure of the questionnaire is most readily seen to be inappropriate: a questionnaire should be ordered with the goal of maximizing the efficiency and quality of data collection and it is unlikely that such an order will be the best way to interpret the results.
The idea of structuring the report like a story has a certain appeal, because it creates a sense of theater and can keep the audience engaged as they wait to see how the story ends. However, the problem with this approach is that it gives the audience insufficient context when interpreting results and they end up either being irritated, or, if reading the report, skipping to the end anyway. The reason for this is that to correctly interpret any research finding you need to know what conclusion it supports.
Structuring information thematically is also less helpful than it often appears. Ordering information thematically assumes the person who is reading the information is thinking in terms of the themes that are chosen, whereas they will generally instead be thinking in terms of decisions that need to be made from the information. For example, if you are trying to understand the potential for a new product, splitting the data up in terms of behavior, attitudes, demographics and the like forces the reader to try and integrate the information together.
The solution to all of these problems is to structure research reports in a pyramid-like structure. Ideally, information should be structured with recommendations at the top of the pyramid. Where it is not possible to make recommendations, then insights should be presented at the top of the pyramid.
Consider the two slides below. Each adopts a pyramid-like structure. The chart below contains some data which the user can read if they wish, but the core conclusion is presented at the top of the slide and thus the user can read this conclusion and decide whether or not it is useful to read the information in the slide.
The chart below also shows its key conclusion at the top. In this case, the key conclusion is a recommendation. If the reader wishes to find out more information they can look at the graph, which shows a driver analysis. The red call-out box shows a more nuanced conclusion, and more detailed conclusions about differences by demographic groups are shown in quite subtle ways, so the ordering of the information to the reader is very clear.
The entire report
The basic idea of adopting a pyramid structure also applies to the overall ordering of information in a report. For example, if we are conducting a new product study, our second slide (after the title page) might contain only a single word:
Or, if a study leads to the central conclusion that prices should be raised, then the first slide may just contain:
Then, the next slide may lay out the core reasons for raising prices:
And the ensuing slides would provide the additional detail for each of the points on the slide above. Again, the benefit of such a structure is that the reader can quickly work out what the key conclusions are and then decide how much effort to invest in getting across the detail.
From time-to-time it is not possible to create a pyramid structure. It may be that you have insufficient understanding to work out what the data means. Or, perhaps the audience does not expect you to make recommendations. Either way, the challenge is then that the research report will appear as something of a data dump and then it becomes important to find ways to create engagement. Standard techniques for creating engagement are:
- Creating pretty presentations.
- Using montages of photos to represent key findings.
- Getting cartoonists to summarize findings as drawing.
- Including short videos, sometimes created by actors but typically recordings of real consumers.
- Creating infographics (for example: www.greenbook.org/PDFs/GRIT-Infographic-2013.png).
- Conducing workshops of the research findings, giving people the objective of creating new strategies based on the findings of the workshop.
- Minto, Barbara (2008), The pyramid principle: logic in writing and thinking: Financial Times Prentice Hall.