Although we spend huge amounts of time as researchers agonising over methodology, much less attention is paid to how we actually communicate the stories that emerge from research; I believe that the art of communication of research insights is something that really deserves much closer attention, and this series of posts is a small step towards this.
In the previous post in this series – Getting your story straight – I put forward some ideas for how to identify the key messages you want to communicate, then structure and express these in an effective and compelling way. Within this I talked about key principles like developing an ‘elevator pitch’ for your story, structuring your presentation around headlines and, above all, approaching the presentation from the audience’s perspective.
In this final post in the series, I will now talk in more detail about using visuals, graphs, video etc. to illustrate, dramatise and humanise your story. (I will not explicitly be covering infographics in this post; this is a subject in its own right that I hope to cover at a later stage.)
Use pictures to simplify and dramatise
Thankfully, the days of presentations illustrated by Microsoft stick men seem to have passed. However, in my experience, imagery is still used by most presenters primarily as a means of making their charts “pretty” (i.e. less dull). In fact, your choice of imagery should be as careful and calculated as your choice of words. An image should be there to clarify, dramatise and (ideally) simplify the point you’re trying to make. If it doesn’t do any of these things, then it shouldn’t be there.
There’s no excuse in this day and age for using crappy clip-art to illustrate your presentation. There are a vast array of sources out there for high-quality imagery. The most simple is, of course, Google Image Search. However, there are issues of copyright with the use of any images you might find here, so we cannot recommend that you use this as your ultimate source. But it’s a good starting point for some quick inspiration.
If you’re prepared to spend a little money on the images, there are numerous online stcok image libraries out there. In my experience, about of the best for this purpose is Shutterstock. They have a good selection, and for most of their images there is a very modest cost involved in downloading a small-to-medium-sized image file, usually more than adequate for presentation purposes. (One caveat here, I always try and steer clear of those dreadful, cheesy fake images of models pretending to be real people that typically appear on stock image sites.)
If you don’t want to spend any money at all, then doing a Creative Commons image search (http://search.creativecommons.org), or looking through Flickr’s Creative Commons libraries (http://www.flickr.com/creativecommons/) are good alternative options.
Another option, of course, is to get your camera out. Nearly all of us have digital cameras at our disposal which are able to capture extremely high quality images. Why buy an image of a green traffic light, a tiny seedling, or a blue sky from a stock library, when the real thing is probably sitting there just outside your window?
And the best option of all is to capture images as part of the research process itself – scrap art exercises, respondents’ drawings, photos of the groups in action, etc.– and use these to illustrate the debrief.
Choose the right graph for the right comparison
You would be amazed just how often I sit in presentations and see graphs whose meaning is obfuscated because the wrong type of graph has been used to illustrate the data. Although most people would immediately know that a pie chart is the best approach generally to presenting component data (i.e. the size of each part as a percentage of the whole), there is much more confusion about the use of column, bar and line graphs to illustrate rankings, time series and frequency distributions.
You could write a whole book about the effective use of graphs to communicate data, and indeed Gene Zelazny has done so with in Say it with Charts; this might not be the book to take on holiday with you, but it’s a very useful reference guide. Drawing on this as inspiration, here is a bluffer’s guide to what type of graph to use with each type of data.
- For component data (the size of each part as a percentage of the total) a pie chart would normally be your chart of choice.
- For ranking items, the best option is usually a (horizontal) bar chart.
- For displaying a time series (how things change over time) or frequency distribution (how many items fall into a series of ranges) you should generally be using a (vertical) column or line chart. Which of the two to use would usually be dictated by how many data points you have to plot (lines are better when you have lots of data points).
- And if you are looking at a correlation (exploring the relationship between two sets of variables), then an XY scatter graph would be the most common choice; if the data is relatively simple, a twin-axis bar graph might be preferable, as these are easier to read.
Lastly it’s worth noting that sometimes it actually better to show a data table in preference to a graph. This is particularly true when the data is simple, when precise figures are of vital importance, or when the audience is used to seeing data in that particular form.
Using colour to enhance meaning
I suspect this may be one of the areas in which I’m going to upset a few people, because many major organisations (including some of our clients) fall foul of what I consider to be best practice by having pre-defined colour schemes for use in all their charts. These may be a great way of reinforcing a corporate identity, but they are lousy way of communicating research data.
The use of colour in charts is not to make them look pretty, or to remind people who you work for, it should be there to highlight key points, and/or to help people understand the point you are making by tapping in to common usage/associations of colour. So, for example, if you are comparing data across the nations of the UK, it would make sense to colour Scotland blue, Wales red, etc., rather than use the colours of your company logo. Similarly, as a rule, it makes sense to the tap into common associations of colours (red equals ‘warning’, blue equals ‘cold’, etc.).
Learn some new software
We live in a multimedia world, and we are dealing with a sophisticated, media-literate audience when we present – especially so if you are presenting to a media organisation.
We also have amazing technology at our disposal nowadays: photo editing, animation software, TV quality digital video and more are all within the range of ‘amateur’ users these days. Yet, more often than not, we limit ourselves to two-dimensional, monochromatic, corporate issue cookie-cutter presentations.
In my early days in advertising I can remember having to learn to use PowerPoint for myself, rather than hand-writing slides then getting a secretary to type them for me. This was the best part of 20 years ago. Just think how much computer technology has moved on in the meantime, yet most of us are still using the very same tools as we did then.
It’s not even that the world of research is averse to learning how to use new software. Most of us have been on courses to learn how to use Telmar, TGI, SPSS or similar. I’m not saying that we shouldn’t be learning how to use those packages, but why aren’t we also being trained as a matter of course to use scanners, digital cameras, video cameras and editing software. I would argue that these should be just as important to our job.
With this in mind, The Knowledge Agency has invested heavily over recent years in developing the skills and resources to deliver high-quality multimedia as an integral part of our research offer. We use high quality DSLR video cameras to film research groups and interviews, industry-standard production tools such as Apple’s Final Cut Pro to edit the footage and export high-quality presentation and web video, and web design software such as WordPress to create micro-sites that allow audiences to access that data in new ways.
This has opened some interesting new methodological avenues to us, such as our specialism in video ethnography, but it also has broader applications to our daily work. We shouldn’t just think of video as a means of recording the research process (and therefore being a form of research data); it is also another means of communicating research insights.
Who says that your narrative, data charts, etc. are best expressed through a PowerPoint presentation? It may be that a short movie is the best, or a valuable complementary, means of communicating that research story, irrespective of whether or not video was used as part of the research methodology.
Lastly, however you choose to communicate your research insights, it’s always good to have at least one piece of ‘killer’ content in your debrief.
“Always have one killer chart—the one that, if your audience does nothing else, you want them to remember.”
Don’t limit yourself to the chart types provided by PowerPoint or Excel; create your own custom chart type, use a graphics package to create a bespoke graphic to communicate your key point or, if you can’t do that, just draw it.
“Drawing graphs live in front of your audience brings them to life in ways that computers cannot do. It also displays mastery of your material. Technology is only beneficial if it is an action mechanism.”
John Beaumont, MD, Energis Squared
Creating this type of memorable moments can be labour-intensive, so less is more. Save it for the most important point in your presentation.
What I’ve gone through here are just a few of the ways in which it is possible to go far beyond the standard, dull research presentation. It’s certainly not a comprehensive list of tricks and tips – I’m sure there are many others that I’m not even aware of. However, I can assure you that if you start adopting some of these, your research will be more interesting to your audience and will have greater impact, without doing the research itself any differently (that’s another story!).