Social impact is an essential part of ESG, itself becoming more central to the mission of housing providers. While for the ‘G’ and ‘E’ there are clear regulations, the ‘S’ requires dialogue with the public.
The need for public engagement varies between the different elements of ESG. Governance (G) is highly regulated, and the environmental (E) aspect is at least partly covered by regulations, although benefiting from public engagement and recruitment of multiple stakeholders. However, when we come to defining the social (S) impacts, dialogue with the communities affected by change and development is imperative.
Discussing ESG can contribute to organisational learning, calibrating internal and external expectations and building trust with tenants. This may apply to all ESG elements but when it comes to social impact, there simply is no credible social impact framework without public engagement.
Circles of engagement
Our data on local development and regeneration demonstrates how social impacts and the perception of gain can be complicated and dependent on multiple variables. Data from dozens of public engagement exercises shows that there are differences in the perception of benefits that relate to factors such as the type of interaction respondents have with the area where change will take place.
We have large amount of data from past projects and we recently conducted a ‘deep dive’ into our data on a medium-size town in a northern UK conurbation. 70 per cent of local residents were positive about the changes presented on Commonplace, rising to above 90 per cent for visitors to the area; here are two quotes from residents:
- “I’m positive about the proposals but the main issue is on nice sunny days, anti-social behaviour around [location] is rife.”
- “Proposals should make the area more inclusive; I feel that some areas are ‘out of bounds’ and that isn’t what the area is supposed to be like. The previous history of [location] was as a more diverse and vibrant population which isn’t the case now.”
Social benefits are both an aggregate and a balance of different populations’ requirements. The viability of local uplifts and gains frequently depends on attracting new footfall to workplaces, hospitality venues or shops, or attracting new home-buyers and tenants to an area. Public engagement is essential for discerning these differences and presenting them back to the different populations. Because most developments affect populations differentially, a digital presence is essential for reaching everyone affected, at the epicentre of a development and further away from it.
Digital reach and machine learning
Maintaining public engagement as plans take shape and are reiterated is essential to maintaining trust and explaining the balance found between the expectations and needs of different groups. These conversations and iterations can be more immediate and more transparent thanks to technologies, not just using the web for engagement but also using machine learning to accelerate the dialogue with communities.
Using machine learning to analyse and summarise residents’ feedback makes the reporting process much faster (days rather than weeks) and more cost-effective. It’s now possible, within the timescale of plan preparations, to maintain continuous engagement without needing substantially more resources than a traditional approach (which typically leaves yawning gaps between consultation phases). The ability to analyse free-text and voice comments can radically accelerate engagement and make feedback to residents fast and cheap.
For example, Commonplace has devoted considerable time and effort to training our machine-learning models to ‘understand’ written comments in specific development areas such as regeneration, public services and transport.
This isn’t trivial because our aim has been to extract not only what people are talking about but also the sentiments associated with their various comments and suggestions. This is done by having our (very much human) specialists categorise thousands of comments and feed these into our machine-learning models. Once the parameters have been coded and tested, the analysis of written comments is very fast, enabling rapid decisions and a quick feedback loop with the public.
We recognise that there are people (usually from specific demographics) who don’t want to use a website for their comments. In these cases, the beauty of machine learning is that any communications, such as an interview, letter, voice message or conversation, can be analysed and the resulting data collated quickly and reliably.
To conclude, we have proposed that:
- Successfully defining social value requires wide and deep community engagement.
- The eventual definition and benchmarks for social value will be a balance between the perceptions and expectations of different groups.
- Finding that balance and getting consensus around it requires reiteration and good two-way communications with residents in different circles around the epicentre of development.
- Such engagement and rapid feedback can be achieved quickly and at little cost with the help of online engagement and newly-developed machine learning for analysing resident and stakeholder feedback.
- This facilitates faster communication and reiteration, thereby trust and reducing the sense residents often have of being mere bystanders to decisions that impact them deeply.
- This package of process and technology can significantly de-risk development planning processes as well as help to ‘bake in’ elements that are of genuine benefit to communities.
Dr David Janner-Klausner is the co-founder of Commonplace.