FEATURE22 March 2011

Do researchers understand visual design?


Are researchers mastering the art of presenting their findings visually? Designer Andrew Doyle and researcher Andrew Tharme go head to head.


Andrew Doyle
Holmes & Marchant

During the Crimean War, Florence Nightingale spent months gathering statistics on the causes of death in military hospitals.

When she got the chance to present her case for sanitary reform to the government, she knew she couldn’t risk them missing the seriousness of the problem.

So instead of reeling off bamboozling figures, she presented her findings in the form of graphs and a very famous modified pie chart. The results were impossible to ignore, instantly understandable and just as compelling today as they were in the 1850s. You can see at a glance that most soldiers had died not from battle but from communicable disease, and how better hygiene measures instantly turned this around.

In a letter to Sidney Herbert, who chaired the Royal Commission investigating deaths in the Crimea, Nightingale wrote that “none but scientific men ever look into the appendix of a report.” She wanted her findings to get through to the “vulgar public”.

That was 150 years ago. Nowadays, the public has access to more data than ever before. But to counter this, a new science has emerged – that of research design. Watch any TV news programme or look at any news site and you’ll see a range of innovative presentations of facts and figures that might otherwise seem uninteresting.

People have realised that stats are inherently boring, and the more we have, the less we notice them. A good infographic (and we’re talking about more than the odd pie chart or graph) allows you to make sense of the data that we’re bombarded with. People are increasingly expecting information to be presented in this way – after seeing an animated infographic on the internet, no stack of figures can ever compare again.

And yet, how is the vast majority of market research still presented? As tables, reams of figures and the odd pie chart or graph. And let me tell you, your clients aren’t Nightingale’s “scientific men” – they are regular people, and consumers too. They are also drowning in data and need help to turn it into useful information and insights that they can act upon.

Well-presented data also allows marketers to do their job better – to communicate these crucial findings internally to peers and the board.

While market researchers are great at producing research, their output is often (necessarily) dense, detailed and dull. Their raison d’être is the accumulation and interpretation of data. They do not seem to see it as being within their remit to make that data easy to understand and absorb.

Even when they try to make sense of their data, they do a job to make designers wince, with all the worst that PowerPoint has to offer: clip art, 3D text, drop shadows and every colour available, leaving the viewer even more confused.

Market research companies don’t currently employ enough people who are design-literate. Instead, their staff are just data-literate. These people are necessary – but you now need designers too.

That’s why companies are going to designers to make sense of the 200-page PowerPoint print-out that otherwise is going to sit in a drawer, unread and unacted upon.

And never mind terrible graphs or injudicious use of clip art – being ignored is the worst thing that can happen to carefully thought-out, expensive and maybe even business-critical research. If your research isn’t easy to understand, you’re throwing money down the drain – and currently, market researchers simply cannot make it easy to understand.


Andrew Tharme
Managing director
SPA Future Thinking

The urban myth of researchers as dull and dusty is all-too prevalent, so I can see where Andrew Doyle is coming from when he suggests that designers are better placed to bring data to life than the data gatherers themselves.

But he’s mistaken. Although there are research organisations still out there that are driven purely by the process and the resulting charts Doyle so deplores, they are a dying breed.

Many more research specialists today are turning themselves around in understanding the need to present their findings and insights more palatably and in more useful formats. We’re not only masters in the art and craft of often complex number-crunching; we have an intrinsic and hard-won understanding of the markets in which we work, coupled with knowledge of how they can be improved through research.

We’re also becoming equally adept at presenting our intelligence within a context that is meaningful and relevant to the myriad stakeholders to whom we communicate. Having worked collaboratively with clients on any given project, we know what the business objectives are and how to deliver findings in an appropriate format.

Those ennui-inducing charts are kept to a minimum. What we want is to have conversations with people and help them achieve their business objectives. That’s our bread and butter. At SPA Future Thinking we’ve even done a gameshow format to communicate data that was particularly complex and needed a high level of interactivity and involvement to be properly understood.

That notwithstanding, it’s important not to lose sight of the main goal of research in the commercial context: to provide substantiated insight.

Obviously communicating this information clearly and effectively is a critical part of the process and researchers have a cornucopia of data visualisation, infographics and other tools to help us do this.

This makes it all too easy to get hung up on the delivery channel rather than the intelligence itself. Data visualisation guru Vitaly Friedman, a man who clearly hasn’t forgotten the importance of design to research, said: “To convey ideas effectively, both aesthetic form and functionality need to go hand in hand, providing insights into a rather sparse and complex data set by communicating its key aspects in a more intuitive way. Yet designers often fail to achieve a balance between design and function, creating gorgeous [executions] which fail to serve their main purpose — to communicate information.”

So I don’t think research has forgotten the importance of design. I believe today’s research professionals are better placed than many designers to get their core insights through to clients in a way that’s not just memorable but useful.


13 years ago

I am not sure what planet Andrew Tharme recently arrived from, but in the last 10 years I have spent in research I have seen very little evidence that researchers are getting better at conveying ideas, insight or data through good design.The infographics display at the Research 2011 Conference is revealing - partly because there were so few submissions. Bringing data to life is a vital and exciting thing to do and an essential part of research, but most of the best entries in the exhibition were from designers and communications experts, not researchers. Having said that, in my view the best way to improve the situation is for researchers to work with, or employ people who are trained and expert in design, rather than attempting a DIY job (which often results in the mess that Andrew Doyle describes). Let researchers research, let designers design. Let them do it together!

Like Report

13 years ago

If we’re honest with ourselves, research agencies have rested on their tried and trusted Powerpoint charts for far too long . I certainly would not argue that some agencies have been guilty of crimes against design too - some of the Powerpoint decks I’ve witnessed certainly do come complete with multiple fonts (Comic Sans anyone?), font shadows and Clipart graphics. However, I’m not so sure that this stems from an inherent lack of understanding of visual design, more that priorities traditionally lay elsewhere. I think that today these priorities have shifted slightly and, most market research agencies are now sitting up and taking notice of data design. Go along to any research conference today and you will still undoubtedly come across stuffy presentations with 15 bullet points on a slide, followed by the obligatory 25 bar charts complete with eye-watering colours. Yet, you’ll also likely encounter some beautifully designed image laden presentations that communicate the message brilliantly without a single bar chart in sight. Here at Kantar Media we’re integrating new platforms and new methods of data delivery into our offer. Our qualitative outputs regularly incorporate professionally shot and lit video content, our conference presentations are rarely bar-chart based, we have bespoke online dashboards and are developing new animated online tools all the time. Indeed, Kantar’s recent announcement that we’re getting David McCandless on board as a creative advisor again shows the commitment we are placing on turning the tide for the better. Now, I’m as big a fan of a good infographic as anyone, but some of the (admittedly, very pretty) submissions are very light in terms of the actual data that they hold. Research agencies (even the most design literate) need to ensure that the deliverables they offer match their client expectations. Many research buyers, contrary to popular belief, still want their stats, still want their deck of data tables and powerpoint charts and would feel a little hard done by if all they got from their agency was a beautifully designed infographic. Personally, I’d love to see better data visualisation from Market Research agencies and fully believe that we as an industry need to up our game in this area.

Like Report

13 years ago

Re: Lucy's point - much depends on the MR agency partner in question. Some are simply better than others. But also consider the audience client-side. We are often presenting at holding board level. These people are used to navigating themselves around the P+L and Balance Sheet. Financial information is the dominant focus on most board agenda's. There is a desire - and a matching intellectual capacity for, data and detail. Visual data coceptualisations can be seen from a senior audience's pov as a bit patronsing or even apologetic ("we'll make it easy for via data visulaisation"). Most client stakeholders like a bit of visual theatre and story-telling but this is nothing new. So know your audience and cater accordingly. Or if you like, customise your presentation design.

Like Report

13 years ago

The example of Florence Nightingale's coxcomb diagrams is an interesting one. They actually distort the data because the area on the chart is not proportional to the frequency of observation (the area of the segments in the outer rings is disproportionately large and the inner rings disproportionately small). Whilst I'm all in favour of good design, and yes, we should try harder, all too often snazzy visuals end up misrepresenting the data and misleading the audience. There is a reason why bar charts have been used so much and for so long!

Like Report

13 years ago

It seems to me one of the major stumbling blocks to greater use of data visualisation is that it takes a lot of time and money to create them, and neither are typically available on the typical research project. David McCandless does his visualisations by hand in graphic design tools, and there are multiple iterations before arriving at the final article. Andrew Tharme claims that " researchers have a cornucopia of data visualisation, infographics and other tools to help us." Really? I'd love to know what they are as I've not been able to find them. Until these tools really exist, I wonder whether the most fertile ground for research is the middle ground - which lies in better designed versions of standard chart formats, that draw the eye to the key takeouts. This is a skill in itself that shouldn't be underrated. Too many powerpoint decks are just churned out by junior researchers with little attention to this.

Like Report

13 years ago

Whilst I would love to learn more about data visualisation as it is a great alternative to the rigidity of PowerPoint, I've also wondered why we need charts in the first place. If the researcher is confident enough, they should be able to pick out a story from the data, present that and stay away from using bar, line, infographics altogether. Granted, the complexity of the data will suggest whether this is achievable or not, but food for thought. Alas, as a few comments above have highlighted, an agency role is to match outputs with client requirements. There's no point delivering data free or data visualisation style presentations if the audience simply doesn't want it! If anyone can point me to a good designer though I'd love to chat!

Like Report

13 years ago

While good design is not unimportant, it is irrelevant if you aren't saying something worth hearing. The bigger challenge for the industry is NOT better display of compelling conclusions but rather creation of conclusions that address clients' business and marketing issues. Yes, we could and should do better in presenting our conclusions but too many very interesting visuals leave me quoting Clara Peller and asking "Where's the Beef?"

Like Report

13 years ago

As a graphic designer with a specialty in information design and an interest in science, I fully experience the intersection of art and science. Art needs science and science needs art. When I flip through journals like Science, I cringe when I see the ill-suited color palettes, thick line strokes, excessive boundary boxes, and a host of superfluous elements in their data visualizations. Information overload is forcing our culture to confront data and present it in an intelligent, visual way. Check out work by Edward Tufte. Or David McCandless. As Lucy wrote, let the researchers and designers work together! Enable this partnership, share grants. Extraordinary results. Don't believe me? Check out this project: http://www.echromi.com/ Believe me? I am available for freelance work. Email nattieliz@gmail.com for more details and my portfolio.

Like Report