How well do housing providers feel they know their residents?

“I believe our organisation has made significant strides… particularly through ongoing casework, direct engagement and data collection,” said one participant.

Another commented: “We are making progress in building the trust with customers, which is fundamental to enabling them to share sensitive information with us. So our information on customers is improving in breadth and depth – but we still have much to do.”

The biggest barrier to engaging with residents was clear. A total of 47% of respondents said their greatest challenge was disengaged or ‘hard-to-reach’ residents (see graph below). This came far higher than other potential challenges, such as not enough internal resources to handle demand (17%), and a lack of detailed data on residents (15%).

But providers are working hard to systematically understand tenants and improve engagement. Almost 67% have at least experimented with customer segmentation – whereby they engage with resident groups differently depending on their needs or preferences.

Of that group, 31% say they have established customer segments and deal with them accordingly. A total of 20% have experimented with segmentation for some projects/areas of their business. And 16% have tried segmentation but have not implemented it meaningfully. Only 5% say they have no plans or desire to introduce customer segmentation of some kind.

“In our own models, we found that the more defined the segmentation, the greater the value,” says Trevor Hampton, director of housing solutions at NEC Software Solutions. “We also found that customer needs and vulnerabilities change quickly and need continuous assessment and updates. This is not a point-in-time activity.”

Segmentation is, of course, informed by data, and the use of this to prompt customer-centric action is widespread – certainly when it comes to residents with vulnerabilities. A total of 88% of respondents said they gather data about vulnerabilities, and of that proportion, 70% use it to proactively support those residents. The other 30% use it to tailor services rather than for early intervention.

And while organisations may still be on a journey towards data maturity, they use a wide range of methods to gather data (see graph below). For example, 76% use resident surveys and direct engagement to gather information to support residents; 72% train staff to recognise signs of vulnerability, such as mental health issues or financial hardship; and 70% act on referrals from social workers, health professionals, schools or local authorities.

And in response, providers use a similarly wide variety of interventions to help vulnerable residents (see graph below). A total of 74% of respondents carry out home visits for isolated or older residents, while 82% provide financial advice and support relating to benefits and 74% offer accessible materials.

Survey results from the second instalment in this research series show that combining datasets is a consistent challenge for housing providers. And it continues to be the case here. For example, 44% in this survey say they treat vulnerability and housing data separately.

However, many are using growing datasets, where new data points – such as data from sensors – are added over time, to inform asset management decisions. Well over a third (39%) said they know which of their properties are at risk of hazards such as damp and mould, and house vulnerable residents accordingly. Another 26% said they actively monitor property conditions using smart devices for insight into fuel poverty or overcrowding, for example.

Data analytics technologies, including AI, offer housing providers the ability to predict issues like rent arrears before they arise and to act accordingly to prevent the issue from becoming a problem.

But given the sector’s relative slowness to adopt the latest technology, it is not a surprise that few are using this yet. A total of 19% of respondents said their organisation does use predictive analytics and/or AI to predict rent arrears, 36% said that they don’t, while 45% said they don’t know.

Of those that do use it, almost all (96%) said they use it to contact residents directly to offer early support once risks are flagged.

And respondents as a whole do see upsides to these kinds of predictive analytics tools: 81% are optimistic about early interventions to enable support and prevent arrears from escalating. And 67% said they liked the idea of income teams being able to prioritise cases based on risk, again enabling early intervention.

“The ability to identify an issue with a rent payment or a repair early can make a profound difference to the outcome for the tenant,” Mr Hampton says.

“Predictive modelling has become an advanced science, and AI tools have added to the quality of the predictions. We believe there will be a significant increase in the use of predictive models for early intervention and prevention over the coming years.”

Resident engagement should sit at the centre of service improvement and, like any strategic initiative, it needs measurable outcomes. In terms of how housing providers assess the impact of their engagement efforts, tenant satisfaction emerges as the most common benchmark, with 88% of respondents using tenant satisfaction measure scores to gauge success. Other indicators include complaints data (68%), improved access to properties for repairs or maintenance (48%) and lower incidence of anti-social or problematic behaviour (44%).