The problem: how to maintain action from your supporters?

The festive season provides valuable campaign opportunities for many charities to secure support from their volunteers and donors. The high levels of recognition and reputation many achieve at this time is important. But as they move out of these periods of high awareness, the challenge to retain and drive interest, support and donations for their causes returns. And the ever-increasing volume of emails, messages and campaigning noise makes this challenge even harder.

At Renaisi we have done extensive research with organisations that look to recruit, use and work with volunteers to achieve their social missions: from volunteering organisations like Vinspired, to social action charities such as City Year UK, and local volunteering initiatives like Our ParkLife. Whatever the context, we have found that this is often a challenge of individual motivation, and what it takes to close the gap between people’s attitudes and their behaviours. Or, to put it another way, why people don’t always do what they say they want to do.[1]

Crisis is well known for its Christmas campaigns, and 2016 was no different in terms of the scale of work the charity undertook to support homeless people. In response to campaigns like these, people sign up to become Crisis ‘campaigners’ and receive information about future opportunities to help tackle homelessness. However, despite their apparent willingness to support the charity, Crisis faces a challenge in encouraging further action from these would-be campaigners.

Could subtle changes to the wording used in campaigns increase support from campaigners?

The experiment

During a live campaign in October 2015, we used a randomised control trial to test the impact of different messages and calls to action on individual campaigners. We wanted to explore the extent to which these could influence their behaviour in support of a campaign to protect the homelessness prevention grant. This includes signing petitions, writing to politicians or perhaps moving online activity to local offline work.

We sent different email communications to six randomly assigned groups of campaigners. Our aim was to see which approach gained the most signatures to a petition and then, more ambitiously, which moved more people to write a personalised letter to an MP. As past behaviour is the best predictor of future behaviour, we block-randomised: ensuring that each group had equal numbers of people who had undertaken different volumes of campaigning for Crisis in the past. Ultimately, we were trying to find the right way to influence behaviour to get campaigners to do the more impactful activity (the letter) rather than just the petition.

The research

Our six different emails were a combination of testing two behavioural insights. One is that people give time and money to causes to gain a ’warm glow’.[2] This is the concept that is used and measured to evidence why people give: there is considerable evidence that we don’t carry on giving as much once we feel sufficient levels of ‘warm glow’. To test this, we asked two groups to just complete the petition, two groups to just complete the letter, and two groups to undertake both by sending a second email after they had been asked about the petition. Even those who weren’t asked could complete a task, however, as both options were accessible through the campaign website.

The second behavioural insight, particularly important for campaigns that use email, is the role language plays in people’s commitment to achieve specific goals. The work of Dolores Albarracín has looked at how communication of behaviour-related messages can influence beliefs and behaviours. She shows that getting the language of any advocacy communication right can have a big effect on success. [3] These persuasive messages are at the heart of what any campaigning organisation does.

Introspection is also an important factor in a person’s motivation to achieve a specific goal, particularly though self-talk and encouragement. The use of the simple future interrogative (‘will I?’ rather than ‘I will’) to stimulate self-talk is found to be particularly effective in leading to goal-directed behaviour for individuals. [4] In our study we changed this to the third person of ‘will Crisis?’ to see how self-talk influences behaviour when mediated by a third party (the charity). We used this distinction to split our two groups that got each of the three email combinations, giving six variations in total.

The findings

We found that the most successful approach was to send a single call to action, but not to use the future interrogative tense. Stating that Crisis will do something and asking for the campaigners’ help was significantly more likely to drive action. Multiple emails saw even the most committed of campaigners do less (if you’ve given me a route to gain warm glow already, I’m more likely to ignore your next request) and the use of a questioning tense appeared to lower motivation for all groups. Comparing the most successful approach to its opposite (multiple asks and the questioning tense), there was a highly significant difference that would have resulted in hundreds fewer petition signatures and letters being written.

We see that as suggesting that the use of a third party (the charity) to drive an individual goal (to campaign on an issue) does not lead to a person fully adopting the goals of that third party in the way that the charity understands them. This underlines that one’s motivation is incentivised by a mix of the goals of the charity and their own, and must be understood as such. Our experience from other work is that many organisations do not spend enough time considering the motivations of individuals, and therefore their attempts to influence behaviour do not always succeed.

Advice for charities

In my view, this has three significant implications for charities:

  1. Strictly prioritise encouraging behaviours which are effective in advancing your cause.
    (And therefore it is essential that they know what is impactful before asking somebody to do it). In the instance of this study, we would suggest that it is a waste of “warm glow” for a charity to ask a committed campaigner to sign a petition, when they would probably do the more impactful behaviour of writing the letter.
  1. Be careful about confusing supporters or weakening the impact of campaigns with multiple asks from across your organisation. This is particularly risky when multiple departments (campaigns and fundraising) are using similar email lists. This experiment showed a gap of several days between emails as lowering activity, but more targeted and precise communication helped build relationships. Other research has showed that you can build engagement through getting the right timing of communication.
  1. Learn from and value those who consistently complete actions. The experiment tested for, but did not find, evidence to support a particular language to communicate with that highly active cohort, but it absolutely underlined their importance to the charity and its campaigning activity.

If you would like to know more about how experiments or behavioural research can help you improve the design of your volunteering work, then get in touch.

John Hitchin Renaisi CEO

[1] In the UK, the annual Community Life Survey is the best guide to understanding levels of pro-social behaviour amongst the UK population. It has, for the last four years, tracked annual changes in the levels of volunteering, charitable giving, social action (defined as people getting together to support a community project in their local area) and civic engagement.

[2] Andreoni, J., (1990), ‘Impure Altruism and Donations to Public Goods: A Theory of Warm-Glow Giving’. The Economic Journal, 100, 464–477.

[3] Albarracín, D. and Wyer, R. S., (2001), ‘Elaborative and non-elaborative processing of a behavior-related communication’. Personality and Social Psychology Bulletin, 27, 691-705.

[4] Senay, I., Albarracín, D., and Noguchi, K., (2010) ‘Motivating goal-directed behavior through introspective self-talk: the role of the interrogative form of simple future tense’. Psychological Science, 21(4):499-504.