Hidden persuasion: Beware of propaganda and manipulation in data narratives

For many of us our job requires us to persuade others to do what we are suggesting. To garner support for what we believe, think, or want. Not as some megalomaniac, but rather and more often to work towards a common goal, to support a common interest. But persuasion must not be about manipulation or control and the brutal enforcement of one person’s idea over another. To be successful and beneficial for all, it has to be the meeting of minds, the arrival together at an agreed way forward. When Blaise Pascal suggested that persuasion must be eloquent and that it “requires the pleasant and the real; but the pleasant must itself be drawn from the true”*, he wasn’t only talking about how we persuade but also the integrity of that act.

When we set out to persuade it’s all too easy to fall into the ‘dark arts’. These are the techniques and approaches we have all learnt, if not directly by working in politics, advertising, PR or propaganda, then by exposure to them. By now we all know the power of storytelling for persuading others. Ten years ago digital advertising and marketing agencies were repackaging themselves as ‘digital storytellers’ (there’s even a LinkedIn group called that). Subsequently storytelling has become one of the hot must have business skills. This is evident in trends like the rise of the TED talk and the realization that ‘death by PowerPoint’ doesn’t help anyone. Today there’s a huge amount of advice on how to be a ‘killer storyteller’, how to make every presentation resonate with the audience, how to be a TED talker.

If you are actively involved in data visualization and analytics then storytelling is seen as an essential part of how you share data insights. You simply have to ‘tell stories with the data’. It’s the key to leveraging the hot engaging and persuasive power of the narrative to support the cool, neutral facts in the data. The narrative helps you frame it, position its impact and stress the importance you have assigned to it. For many years it was the opposite of this, at least for journalists. It wasn’t about ‘data storytelling’. Instead the cool numeric ’facts’ were used to legitimize and shore up an argument, to add a convincing air of expertise, often ignoring the original context or any scientific rigor. It’s that practice that brought forth the fabulous phrase “lies, damn lies and statistics” (the history of which is a great story in its own right 1).

Many of us have chosen to ‘flip the script’ and now present the data as the story. We do this by placing the data via visualization upfront, in an attempt to lay the facts bare. However, when we present data we must be careful not to engage in the “prettying of falsehoods”*. We must keep a close eye on the integrity of the act, creating an empathetic engagement with those who we hope to communicate with. Unfortunately we often fall into the same traps and transgressions that all storytellers must navigate. We know that for our stories to persuade they need to be compelling and captivate our audiences. We are told this over and over again. What’s more we intrinsically know this as we are all readers, listeners and watchers. We are all the consumers of other people’s stories and we know what we want from that experience. As authors and storytellers we look to frameworks such as Dan and Chip Heath’s S.U.C.C.E.S.S (Simple, Unexpected, Concrete, Credible, Emotional, Stories, Stick 2) to help us up. We work hard on the craft, the packaging of our facts and our delivery. But if you are competing in a narrative arms race where everyone seems to be trying to present the next TED talk then sometimes you may find yourself turning to the dark arts I mentioned earlier.

In one way or another when we set out to persuade we set out to sell, whether that’s an idea, a fact or a falsehood. In the 1957 book The Hidden Persuaders, Vance Packard wrote about how the Advertising and Public Relations industry persuades people to buy things by leveraging subconscious or ‘hidden’ needs. The book offers a simpler view of people and marketing than we have today. It’s written before the advances in behavioral economics and the deeper understanding of motivation we now have, but the eight hidden needs he identified still ring surprisingly true. They are the buttons that get pushed, that nudge us towards this product or that and have underlined many of the narratives and stories of the commercial world. When we create our business presentations or construct data visualizations it’s unlikely that we are embracing these deep psychological tools. We may not be trying to convince our audience that unless they believe that “variance in spend on X has a significant influence on long-term profitability” they will be doomed to be alone in the world. We rarely consciously construct our stories to address our need for emotional security, reassurance of worth, ego gratification, creative outlet, love objects, sense of power, roots or immortality 3. But it’s there under the surface. What Packard identified was the subtlety, power and sophistication of the tools we use to persuade people. The more we feel the need to persuade via a single authorial voice (to tell someone something), the more we are likely to leverage these hidden needs. Even if we don’t set out to actively manipulate, we employ tricks and techniques to help our ‘facts’ resonate with our audience making them feel important and engaging. We do this so that the audience remembers the value that we have assigned to the story. And it’s not just down to the storyteller, our audiences are complicit in the exchange, as soon as we take up the role of the narrator, the audience’s expectation is set; “here we are now, entertain us”. We instantly bring forth this dynamic; storyteller and audience, us and them. We want the audience to accept our story, they want to be engaged and we want to give them what they want. And that, according to Scott Berkun is the “moral trap of storytelling”4. Just how far do you go in giving them what they want?

As my colleague James Richardson points out in Spinning Data Yarns often we unwittingly become unreliable narrators. Where even unrecognized by ourselves, we craft the story to suit an implied reader, to fit our agenda, to support an existing belief. Whether that’s by deliberately leaving things out to help ‘clarify the message’, or making ‘best guesses’ when we don’t have the supporting data, or skipping over or lingering too long on one fact over another, or even simply changing the order of things to add drama and a more powerful finish. These activities sound pretty devious, but are very much part of good storytelling. They are key to a great narrative and reflect the structures and rhythms we have come to expect, the whodunnit, the three acts or the hero’s journey.

Ultimately the only way to avoid these dark arts is to step away from the monologue and the narrator, the authorial voice. If it’s truly about a common endeavor then it must also be about collaboration. That requires shrugging of the trappings of the single narrative. It has to be a real exchange, a debate, a shared experience of discovery. Where the facts and insights are teased out, questioned and validated together. Where the narrator has no special status and it’s not about the authority of their position or voice. Where we can get back to the scientific rigor that we believe our data brings to the table and be open to Carl Sagan’s ‘baloney detection kit’ 5, where we all get to:

• Check the facts or get independent verification,
• encourage debate from knowledgeable proponents for all points of view,
• ignore authority,
• create multiple hypotheses for a single idea,
• question your own ideas and be willing to kill your darlings,
• quantify our findings and open up the data,
• examine every link in the chain of argument,
• employ Occam’s razor and select the simpler of two equally well suited hypotheses,
• and always ask whether the hypothesis can be falsified.

All this is about debate, not story, not presentation. Only through this active engagement can we get to the eloquence that Pascal spoke of, because as Pascal pointed out 400 years ago:

“People are generally better persuaded by the reasons which they have themselves discovered than by those which have come into the mind of others.”

* Thanks to Maria Popova’s brilliant Brain Pickings for the Pascal references https://www.brainpickings.org/2015/05/20/blaise-pascal-pensees-persuasion/

1 — https://en.wikipedia.org/wiki/Lies,_damned_lies,_and_statistics

2 — Made to Stick: Why Some Ideas Survive and Others Die, Chip & Dan Heath

3 — The Hidden Persuaders, Vance Packard, also see http://changingminds.org/explanations/needs/packard_needs.htm

4 — Scott Berkun — http://scottberkun.com/2015/the-four-lies-of-storytelling/

5 — The Demon-Haunted World: Science as a Candle in the Dark, Carl Sagan

This piece originally appeared in Qlik’s Research Digest (download a PDF of the full digest here : The Art of Persuasive Communication in the Workplace)