Advancing Global Evidence for Local Practice

Background: Why we need localised evidence based insights ?

Evidence-based insights is widely regarded as the crème de la crème of strategic thinking for many governments, businesses and nonprofit organisations across the world. This idea is grounded in proposing programmes, products and services based on established scientific evidence of what works. As we embrace scientific methods in our strategy, we ought to consider the limitations of advancing a one-size fits all approach.


One of the most common insights from social psychology is the idea of ‘social proof’. The term, coined by Prof. Robert Cialdini, suggests that people are more likely to engage in a specific behaviour if they can see that people are doing it. When Wikipedia tested this message in their fundraising banner, “... fewer than 1% of readers give”, they violated the global evidence of positive social proof. By highlighting that less than 1% give (and in some instances “... less than 99% don’t give”) in their messaging, they were doing the exact opposite. The insight from social proofing suggests that they were normalising the fact that most people don’t donate which in turn would reduce donation rates.

When Wikipedia tested this message in their fundraising banner, “... fewer than 1% of readers give”, they violated the global evidence of positive social proof.

Many called for Wikipedia to remove this immediately or took to social media to express their grievances with this strategy. But it turns out, their online fundraising team was aware of this evidence-based insight and deliberately decided to test the opposite message on multiple occasions. They found every time they removed this message, their donation rates dropped significantly. The team hypothesised that there must be something in the localised donation context that alters the global rules of social proofing.

Case Study: How can we use "nudges" to improve voter turnout ?

In the 2016 local government elections in Auckland, the voter turnout was less than 40%. The Research and Evaluation unit at Auckland Council had conducted a comprehensive review of behavioural insights to identify new strategies to increase voter turnout in the 2019 election. This report revealed a wide range of promising messaging strategies from global evidence, such as, expressing gratitude to create positive feelings of reciprocity. Another study with Danish voters suggested that messages with duty-framed-as-losses were most effective (e.g. “Do not let others decide your everyday ‒ vote”).

Testing 12 x behaviour change strategies on Nudge Testbed platform with 433 participants


Our challenge was to consider all the global evidence to support Auckland Council and develop localised strategies that might increase voter turnout.  We applied the Nudge Testbed platform to design 12 global strategies in the form of messages and stress-tested this with 433 residents in Auckland to identify localised strategies to motivate different demographics to vote. The online experiment revealed that the most persuasive localised strategy was to get people to feel ownership over their right to vote, which is referred to as 'the endowment effect'. The losses people may experience as a result of not exercising their rights can actually motivate more people to vote.


The most persuasive localised strategy was to get people to feel ownership over their right to vote, which is referred to as 'the endowment effect'

However, for some underrepresented demographics, the most effective strategy was not always the most effective approach. For young Aucklanders (18 to 34 year olds) and Māori audiences, the message "Make sure the right people represent you" was a far more effective strategy. To achieve a sustainable democracy, a pivotal localised strategy is to inspire diverse candidates to stand for the local body elections. Celebrating younger candidates and diverse Māori representatives can inspire demographics that have lower levels of voter turnout.

Bottom Line.

The adoption of evidence-based insights should only be the starting place for advancing new strategies. Rather than blindly following global insights and expecting behaviour change to follow, we need to stress-test our strategies with our audiences for better outcomes. In order to do so, a new set of scientific tools can enable us to rapidly understand what resonates and what does not resonate with people in the localised social context.

Advancing global evidence: The next evolution to evidence-based insight needs to involve collaboratively designing with communities and testing which global insights from behavioural science supports changes in our local practice.

- This article is from an excerpt published in the Diversify Annual Compendium 2020.

Keep learning insights from behavioural science

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Learn something new every month. We deliver behavioural insights straight to your inbox.