What is social proof?
Social proof is a concept that describes how our behaviour is often influenced by how we see other people act. We look to others to determine how to behave and which decisions to make in a given situation. This explains why restaurants keenly sit patrons by the window when it’s quiet and why nightclubs will unnecessarily keep a queue of people waiting to be admitted outside[1]. We make our decisions in reference to others’ behaviour. Marketers have successfully used this for decades and it explains testimonials, celebrity endorsements and why we are told what percentage of consumers are happy with their purchase.
Plenty of research has documented this phenomenon. In the 1950s Solomon Asch ran his conformity experiments which tested the extent to which participants would be influenced by the (wrong) decisions or answers of others. Participants were asked to match lines of the same length and state their answers. Unbeknownst to the target participant, the other group participants were confederates/stooges who were told to give the wrong answers. This documentary footage shows the surprising level to which participants are influenced by the behaviour of the group.
How is the concept used in behavioural science?
Social proof is one of the strongest concepts and methods that behavioural scientists employ to influence desired behaviours. It has been used to increase rates of tax collection, energy conservation and voter turnout, to name but a few.
Perhaps most famously this insight was used in the UK by the Behavioural Insights Team (aka Nudge Unit) in letters to taxpayers informing them that 9 out of 10 people pay their tax on time[2]. They tested three messages saying that the majority of people had paid on time either nationally, or in the recipient’s postcode or in the recipient’s town against a control group with no message about the behaviour of others’. This led to increased payments represented hundreds of millions of pounds in the Exchequer’s pocket. It was one of the first experiments that proved the efficacy of behavioural approaches to public policy and the value of running trials to definitively know which interventions work most effectively.
Social proof has also been used to increase voter turnout in the August 2006 primary elections in Michigan, USA. Three variations of key messages were sent in letters to citizens:
Stating “Do your civic duty – vote”
Stating whether members of the household voted in the 2004 primary and general election and saying that the record will be updated and circulated after the 2006 election
Stating the household voting record and including information on the voting records of neighbours – signalling that in the future their neighbours would know whether they voted
Variation #1 had a small effect compared to a control group receiving no information, but #2 lead to a 16% increase in voter turnout and #3 led to a 27% increase in voter turnout[3]. This small intervention, inserting a line of text into a letter, led to dramatic increases in the desired behaviour.
A final classic example is on encouraging environmentally-friendly behaviours in hotels. A study sought to understand how social proof influenced the reuse of towels by hotel guests. One group received a general message about the value of protecting the environment, the second group’s message said that most hotel guests reuse their towels, and the third group’s message said that guests who had stayed in the same room had reused their towels. Against the control group, the second group were 26% more likely to reuse their towels, and the group told about reuse in the same hotel room were 33% more likely to[4].
So far, these examples have focused on highlighting positive, desirable behaviours that we want to see. But the insight also holds in the reverse – highlighting undesirable behaviours can send the signal that the negative behaviour is widespread and can encourage the bad behaviour. Robert Cialdini famously found that norms could affect theft levels in the Petrified Forest National Park, Arizona by indicating whether the behaviour was widespread. A sign “Many past visitors have removed petrified wood from the Park, changing the natural state of the Petrified Forest” saw theft levels of 7.92% whereas the sign “Please don’t remove the petrified wood from the Park, in order to preserve the natural state of the Petrified Forest” saw theft levels significantly lower at 1.67%[5].
So, we are far more easily influenced by the behaviour of others than we suspect. And we can utilise this to try to encourage desired behaviours in the responsible business realm. As the petrified wood example shows, we need to be careful about what signals we are sending. We do not want to “inadvertently to reinforce a negative social norm by emphasising the prevalence of an undesirable behaviour”[6]by signalling that most people are failing to act appropriately.
Social proof is often leveraged (for good or evil) in contexts that have serious human rights implications, including where business is involved. As Andrew Woods argues in his excellent article on the promise of behaviouralism for the international human rights regime, “social proof is also perhaps the most compelling explanation for the sudden cascade of demonstrations in Egypt in 2008”[7]. He argues that the use of Facebook enabled people to see that others were protesting (or planning to protest), they had more visibility of the political preferences of others and could see that they were supported by their peers in opposing government policies. This led to the protests that overthrew the Mubarak regime. Currently, there is a debate around the extent to which Facebook also enables the reverse - the dark side of social proof - by normalising discrimination and supporting ethnic hatred, for example in Myanmar. Andrew Woods also describes the role of social proof in the Rwandan genocide and how central it was to the propaganda strategy of the génocidaires, with radio broadcasts serving: to “inform listeners of the behavior of others (other Hutus are killing Tutsis); to inform listeners of the prevailing norms (killing Tutsis is socially sanctioned); and to invoke shame and guilt on the part of those Hutus who were not hard at “work.””[8]
We also see how social proof functions in business contexts positively or negatively. The Wells Fargo scandal – where bankers opened millions of fake and fraudulent accounts in customer’s names in order to meet unrealistic sales goals – had all the right policies in place…
“there was no shortage of internal publications that advised Wells employees on how to conduct themselves, including the Wells Fargo Code of Ethics and the Wells Fargo Team Member Handbook, which warned if an employee engaged in “manipulating or misrepresenting sales [gaming] in an attempt to meet sales goals or receive compensation” immediate termination could result”[9]
But employees looked at the incentive structure in place, saw how employees not meeting targets were let go, saw how reports to management about difficulties meeting targets fell on deaf ears and saw that the response of peers was to engage in gaming behaviour. The social proof and the social situation led employees to a rational conclusion of how to behave.
How could social proof be leveraged in the responsible business field?
It is clear that social proof can be better utilised in driving better business behaviour. And it is already used (intentionally or intuitively) to build health and safety cultures within companies. There are a few areas where social proof might be leveraged more effectively.
At the responsible business field-level we want to beware normalising bad behaviours. Highlighting low levels of corporate action on human rights – for example, “only 54% of companies are complying with the UK Modern Slavery Act statement requirements”, or “4 in 10 companies ‘failing’ on human rights” - may send the signal to companies not to bother – their peers aren’t. It is not the norm. There may be plenty of other reasons to highlight corporate inaction, for example to inspire consumer pressure or galvanise government regulation. But if the aim is to encourage companies to actually change their behaviour, it may be more effective to reframe messages that emphasise corporate inaction or bad behaviour and instead focus on examples and key messages that highlight the desired behaviour as a norm. Practitioners often use this approach when trying to secure support or resources from senior leaders by gathering best practice examples of what peers are doing and presenting these as the standard of expected behaviour, ignoring the laggards. Further, the narrative around SMEs is often that responsible business practice is harder for SMEs given their unique characteristics compared to MNCs. But social proof may encourage us instead to highlight the positive examples and stories of entrepreneurs driving rights-respecting behaviours within their companies and to create networks of SMEs with similar experiences to share to act as a beacon to others.
For business practitioners, leveraging social proof might mean paying greater attention to how messages and signals can be perceived by employees or suppliers. For example, we may want to beware sending overly regular reminders of a corporate code of conduct, which may signal widespread non-compliance. Instead, timely reminders at periods where transgressions are more likely (e.g. when trying to meet year-end targets or before social events) may trigger the good behaviour reminder at a desired time without normalising the bad behaviour. Practitioners may benefit from efforts to emphasise desirable behaviours and many practitioners already utilise this, for example by including stories of employees noticing something amiss and reporting it, leading to the company preventing a problem, in newsletters or large employee gatherings. Practitioners also may want to highlight good practices in different parts of the business and use transparency to drive competition between different subsidiaries, operations or facilities/sites.
Practitioners may also want to consider how training formats can normalise norms and behaviours. In-person training is not just a conveyor of information “but also a stage for business leaders to help establish and shape social norms”[10]. Whereas online training creates an academic understanding of the material, watching colleagues role-play training scenarios better activates behaviour moderation by allowing us to observe and imagine conforming to our peer behaviour[11]. Social proof is better leveraged by seeing how our peers behave in similar situations.
I hope that these initial ideas for how to leverage social proof resonate. I’d love to hear how you are using social proof in your work and your experiences. Let me know at katrynwright[at]googlemail.com
[1] B. Ross and O. Mahmoud (2018) ‘Change for Good: Using Behavioural Economics for a Better World’, p.199
[2] Cabinet Office (2012), ‘Applying Behavioural Insights to Reduce Fraud, Error and Debt’, p.22
[3] O. Service and R. Gallagher (2017), ‘Think Small: The Surprisingly Simple Ways to Reach Big Goals’, p.151
[4] F. Gino (2018), ‘Why It’s So Hard to Speak Up Against a Toxic Culture’, Harvard Business Review
[5] R. Cialdini (2003) ‘Crafting Normative Messages to Protect the Environment’, Current Directions in Psychological Science Vol. 12. No.4., p.107
[6] O. Service and R. Gallagher (2017), ‘Think Small: The Surprisingly Simple Ways to Reach Big Goals’, p.154
[7] A. Woods (2009), ‘A Behavioral Approach to Human Rights’, Harvard International Law Journal Vol. 51 No.1, p.60
[8] A. Woods (2009), ‘A Behavioral Approach to Human Rights’, Harvard International Law Journal Vol. 51 No.1, p.60-61
[9] B. Mclean (2017), ‘How Wells Fargo’s Cutthroat Corporate Culture Allegedly Drove Bankers to Fraud’, Vanity Fair
[10] M. Bedan (2019), ‘Behavioral Science Lessons from the Wells Fargo Scandal: Focusing on the “Regular Apples”, The Anti-Corruption Report Vol. 8, No. 3. P5
[11] Ibid.