Social Impact and Brand Sentiment: an NPS Story
Net Promoter Score (NPS) has become the gold standard for companies to gauge consumer sentiment. Consultancies like Bain and marketing firms like Edelman have produced massive amounts of research showing the economic benefits of a high NPS including faster annual growth, lower customer acquisition costs, and greater customer loyalty.
But NPS can be a difficult metric to move, especially via marketing. The biggest contributors to NPS tend to be product quality and value as well as customer service. An engaging brand may help, though usually the brand becomes cool through high NPS not the other way around.
At Public Good, we are interesting in testing the connection between two ideas: the first is that consumers increasingly value brands that have positive social impact and the other is that inviting consumer participation in that social impact is a great way to empower customers and build brand advocates. While there is high quality research these ideas (the 2018 Cone/Porter Novelli CSR Report, for example, now identifies social impact as second only to product quality in consumer buying criteria, beating out price for the first time), this research is based on surveys that asks customers to describe their potential buying habits. Many are skeptical of this kind of research since the context of the questions being asked and the theoretical nature of the questions can lead to various forms of bias. Before dedicating more resources to social impact initiatives, data-driven marketers want to know if customers will be quite so enthusiastic about their social impact when they are looking at the supermarket shelf or the search results on Amazon.
We felt that with out platform which allows people to take action on social causes alongside brands, we were in a unique position to look for supporting evidence and NPS could be an ideal way to cancel out the potential biases. With all the work that has been done to prove the connection NPS to actual buying habits, we felt that if we could close the loop between impact and NPS, we’d be able to make a strong argument supporting the other surveys conclusions.
So this year we did just that. We started with a sample of 15,000 consumers who used our technology embedded on a major media company’s website to donate to hurricane relief in 2017. This media company has an NPS of -5 to +5 according to the multiple public sources, so we could use that baseline to determine if the consumers who participated in hurricane relief with the company viewed the brand differently as a result.
Our methodology was to create three Facebook target groups and ask them to complete an NPS survey on the media brand without stating who we were or why we were asking. The first group was composed of people who had participated in the hurricane relief effort. The other two groups were controls: one of them random, the other of people who had liked some content on the media company’s Facebook page (the closest proxy we could find for its general customers.)
The results were powerful and surprised even us. The two control groups both underperformed the publicly reported NPS statistics of -5 to +5 while the test group came in with a staggering 92 NPS, which is completely off the charts. The measurement was made on consumers who took action up to nine months before the survey was taken, so there was also evidence to support that the NPS impact lasted over time.
While the result does not rigorously prove causality — that consumers who took action were greater supporters of the brand because of taking action rather than vice versa — the causality is strongly implied by the fact that the data strongly support the Cone/Porter Novelli survey, which did ask exactly that.
We are very excited by this initial result and look forward to continuing to study the relationship between social impact and NPS. Our hope is that this link will enable impact oriented executives to make the argument that their social impact does well for the company and is worth continued prioritization and investment.
