“Keep it simple stupid”? Not if you’re asking for help

How can you convince someone to give you their time or money free of charge? Focus on the head, not the heart.

The toughest sell in business isn’t a sell at all, strictly speaking. Convincing others to donate valuable time or resources to your cause, without any tangible compensation, is the rarest and most prized of communications skills. That is especially true in our age of digital mass communication, when written appeals jostle for attention in our email inboxes and social-media feeds every day. For those few who have mastered it, the art of successful solicitation can come in handy in myriad professional contexts–from crowdsourcing and crowdfunding campaigns to recruiting volunteers for undesirable tasks.

Jiyeon Hong, assistant professor of marketing at George Mason University School of Business, recently published a paper in Marketing Science (co-authored by Paul R. Hoban of Amazon) shedding light on why soliciting uncompensated help is so difficult for most of us. These solicitations may flout one of the most well-known rules of business writing: namely, “keep it simple stupid” (KISS).

The assumption behind “KISS” is that readers respond most strongly to lean prose that makes minimal mental demands. But Hong’s research, including algorithmic analysis and a randomized controlled trial (RCT), suggests that simple, punchy writing is not always the most convincing for donors.

Hong’s algorithm builds upon the hierarchical attention network (HAN) model, a deep learning methodology that classifies documents by identifying their most salient features and mapping them to predetermined categories. “Hierarchical attention allows us to see which words and sentences are most detrimental and beneficial to success,” Hong says.

To train and test the algorithm, the researchers used data from DonorsChoose, a crowdfunding platform for public school teachers seeking support for classroom projects.

After training with more than 60,000 DonorsChoose projects from the period 2009-2017, the algorithm identified which sentences within the DonorsChoose proposals were the most pivotal to the outcome (i.e. getting funded or not). The effectiveness of the algorithm was then verified in an RCT, in which teachers revised the essays–half the teachers were given the algorithm’s recommendations, and half were not–and a group of undergraduates rated the revised work. Overall, the essays revised with the help of the algorithm were judged 4.5 percent more likely to be funded, a difference can be translated into additional funding of nearly $10 million.

Sentences classified as “beneficial” and “detrimental” to success by the algorithm exhibited a consistent pattern of characteristics.

“More concrete and specific content tended to be in the beneficial sentences,” Hong says. “Acronyms and insider terms also appeared often in the more persuasive sentences.” In direct opposition to the “KISS” rule, beneficial sentences were slightly longer and demanded more of the reader. Their average readability score was 9.51, compared to 8.72 for detrimental sentences. (Readability scores correspond to the grade level required to understand the sentence.)

For example, instead of describing diversity in vague language such as “Our school has a very diverse student population,” a beneficial sentence would be densely packed with detail – e.g. “Our school is a Title 1 school serving a diverse and vibrant student population: 80 percent are students of color and nearly half are English Language Learners.”

These findings added supplemental insights to official writer’s guidance on the DonorsChoose website, which stressed “Don’t use jargon,” “Tell a story” and “Let your students shine”. While steering clear of jargon may be a good rule of thumb, the algorithm identified exceptions to this rule–i.e. when use of jargon would strengthen a rational argument for help or funding. Similarly, storytelling and examples of student accomplishments should include specific details that pinpoint relevance and value.

Hong explains, “The readers are in cognitive mindset, trying to compare many options because they also have a limited budget. In this context, objectively delivering information can be more persuasive. Facts win out over emotions.”

Her explanations piggyback upon past research, such as a salient study from 2008 showing that in contexts demanding rational thought rather than emotional response, communicators should use language that appeals to the head, not the heart.

Hong also references a 1984 study that identified a direct relationship between the number of arguments in a message and its persuasiveness. Additionally, a 2011 paper explored how conditions that force us to work harder to retain information can aid long-term learning, a phenomenon the authors termed “desirable difficulties”. Both studies point to scenarios where a more complex written presentation may sink in better than simple, easily digestible language.

“We conclude that the most successful appeals for help will not be those that make the simplest and tightest arguments. Instead, they will a) expose the reader to a modest amount of desirable difficulty; and b) put forth a detailed case that is low on emotional coloration,” Hong says.