Housing data leaders: Why making AI work depends on your workforce

The Housing Ombudsman has claimed that data is the silver bullet to solve the top 3 issues in the housing sector. But does this tell the full story?

For all the noise about “proptech” and “AI”, the uncomfortable reality is this: most UK housing providers are still firefighting the same complaints, repairs and damp and mould issues they were dealing with a decade ago. AI can help, but only if people understand data well enough to trust it, challenge it and act on it.

Why data and AI literacy are now business-critical in housing

Regulation, scrutiny and tenant expectations have all shifted faster than most operating models. The Social Housing (Regulation) Act, Awaab’s Law and strengthened consumer standards have hard-wired damp, mould, safety and tenant voice into the core of landlord accountability.

At the same time, Tenant Satisfaction Measures (TSM) data shows a sharp drop-off between general satisfaction and complaints handling. Surveys show that 5 out of 7 residents are happy with services. But only 36% are happy with complaint resolution.

This is the new reality for data leaders in housing: your organisation will be judged less on your AI experiments and more on whether data-informed decisions tangibly improve damp and mould responses, repairs performance, and how heard residents feel when they raise issues.

That is impossible without a workforce that is confident with leveraging data and AI, not just surrounded by it.

The AI misconception: tools first, people later

If you work in data or digital, you’ve probably seen the same pattern:

• Invest in new case management or asset systems, then discover the data going in is inconsistent or incomplete.
• Launch a predictive model for repairs prioritisation, only to find planners override it because they don’t trust the outputs.
• Roll out new analytics dashboards to managers who still ask for a spreadsheet or use their “gut feel” before they act.

Data Literacy Academy attended the Housing Innovation Show in Birmingham on February 5th, and many of the keynote speakers made the same point repeatedly in different guises.

When housing providers succeed with AI and advanced tech, they start by building the strategic capability and culture, not by buying the cleverest system.

Across sessions on digital twins, asset lifecycle management, GIS mapping, income management AI and agentic AI, three themes keep emerging:

1. Technology is no longer an IT side-show. It is reshaping core operations, from Awaab’s Law compliance to predictive maintenance and income collection.
2. The sector is suffering an acute skills gap and “brittleness cycle”, with limited capacity beyond daily firefighting. Workforce readiness for AI is now called out as a critical risk.
3. Agentic, autonomous AI systems are “useless without high-quality, standardised data” and strong governance foundations. The sector will keep lagging behind other industries if it doesn’t improve the basics.

In other words: if your staff are not data- and AI-literate, your AI strategy becomes an expensive way of proving your culture is not ready.

Where AI can add real value: Repairs, capacity and processes

Housing providers are dealing with rising repairs demand, ageing stock and limited contractor capacity.

Advanced scheduling and predictive tools are starting to change that.

One of Data Literacy Academy’s customers, Raven Housing Trust, shared how they are “shifting from transactional scheduling to real-time reactive optimisation”. They’re using data to optimise and absorb increased demand, plus improve tenant satisfaction.

Asset Lifecycle Management tools now provide real-time insights, predictive maintenance capabilities, and long-term asset planning to reduce costs and enhance resident satisfaction.

And we saw how video diagnostics and AI-driven triage (e.g. Help me Fix) are already resolving up to 20% of repairs remotely, reducing unnecessary callouts, cutting carbon and boosting first-time fix rates.

But these benefits rely on basic capabilities: teams must understand why certain jobs are prioritised, what data the models use, and how to flag when the data is wrong. That is data literacy in practice.

Damp, mould and Awaab’s Law

Damp and mould are now the litmus test of whether a landlord is serious about data, compliance and tenant safety. Under Awaab’s Law, landlords face strict deadlines for inspection and repair, and regulators are watching not just outcomes but the quality of evidence.

The sector is already exploring AI-backed solutions:

• AirTrap XL uses “lab-verified artificial intelligence” to support more informed and consistent decisions in damp and mould management, helping leaders strengthen compliance and improve living environments.
• Other sessions focus explicitly on using data insights to maintain regulatory compliance under Awaab’s Law, and on moving from reactive repairs to proactive, data-led risk management.

Again, these tools are only powerful if:

• Frontline teams know what environmental and case data is being captured and why.
• Caseworkers can interpret risk scores and communicate them clearly to residents.
• Leaders understand the limits of the models and can challenge bias or blind spots.

Without that literacy, AI becomes another black box. And in a context as sensitive as damp and mould, blind trust is not an option.

Complaints and tenant experience

Now let’s take a look at complaints, where a staggering 74% percent of tenants are unhappy with how these are resolved. What can the industry do to improve this reality?

Here are a few practical solutions that organisations need to work on implementing. They of course all require tech and people to operate in sync.

• Natural language processing can classify incoming complaints, detect themes (e.g. recurring damp in a block), and triage cases to the right team faster.
• Federated case data (across repairs, ASB, arrears and previous complaints) can provide a single view of the resident and their history, reducing the need for them to repeat themselves and allowing more empathetic decisions.
• New diagnostics tools showcased at the Housing Innovation Show allow residents to show issues over video, helping landlords hit inspection deadlines and improving first-contact resolution.

But if complaint handlers cannot understand basic dashboards, comprehend risk flags, or explain why an AI-assisted triage decision was made, the technology will feel like a shield rather than a service. That erodes trust, both internally and with residents.

The missing piece: a data- and AI-literate workforce

Most data leaders in housing are already investing in:

• New systems and integrations
• Data warehouses or lakes
• Dashboards and reporting tools
• Point solutions for repairs, income or compliance

What is often missing is a deliberate, strategic investment in data and AI literacy that reaches far beyond the data team. This is exactly the gap the Housing Innovation Show agenda surfaces in sessions on workforce readiness, AI skills and cultural transformation.

A data- and AI-literate workforce is not made up of “mini data scientists”. It is made up of people who can:

• Capture better data at source: Frontline staff understand why consistent coding of repairs, damp categories and complaint types matters, and they enter data in a way that machines and humans can both use.
• Interpret and question insights: Managers feel confident reading TSM dashboards, spotting anomalies in complaint resolution times, and asking the right questions of predictive models.
• Use AI responsibly: Staff know when to rely on AI suggestions (e.g. triaging standard repairs) and when human judgement is essential (e.g. vulnerable residents, complex disrepair).
• Communicate decisions: Case handlers can explain to tenants how data has informed a decision in plain language, strengthening transparency and trust.

Crucially, the Housing Innovation Show includes a session explicitly tackling “workforce readiness for AI”, warning that informal AI use plus low digital confidence create compliance and efficiency risks. That’s why partners like Data Literacy Academy support the 80% of people who are often left out of upskilling programmes, but are crucial in enabling the solutions so many tenants would benefit from.

Reframing the assumptions data leaders often hold

If you’re already a CDO, CIO or CTO, it’s helpful to gently challenge a few common assumptions.

Assumption 1: “Literacy is training; we’ll do it after the system goes live.”

Reality: literacy is a strategic enabler that needs to be designed in from the start.

• Awaab’s Law and building safety require demonstrable, auditable decision-making. Training people after the fact to “use the system” is not enough. They need to understand the data value chain, how their actions intersect with this trail, and what outcomes they create.
• Agentic AI and autonomous systems panels explicitly call out the need to build a reliable “data foundation” for the next 3–5 years of digital investment, which of course includes people who can steward that data day to day.

Treat literacy as part of your core risk and transformation strategy, not a nice-to-have.

Assumption 2: “If we fix data quality, AI will work.”

Reality: data quality is a behavioural problem as much as a technical one.

• Without literacy, the same inconsistent behaviours that created bad data in your legacy systems will replicate in your shiny new platforms.
• Sessions on “disconnected systems” at the Housing Innovation Show describe how fragmented processes and workarounds make it hard to evidence compliance or hit Awaab’s Law timescales. This problem is rooted in human habits as much as in architecture.

A literate workforce is your most powerful data quality tool.

Assumption 2: “AI is about reducing headcount.”

Reality: in housing, AI is about freeing scarce capacity to focus on what only humans can do.

• Income management projects using AI and ML show “measurable improvements in arrears recovery, workflow automation, and officer capacity”, freeing officers for proactive tenant support.
• Video diagnostics and remote resolution tools reduce unnecessary visits so operatives can focus on complex jobs and vulnerable residents.

AI is a powerful way to augment human creativity and empathy, as it frees up time by reducing firefighting and enabling more considered, creative problem-solving that delivers a higher quality to tenants.

What a strategic literacy investment looks like

To put this into practice, data leaders benefit the most by treating data and AI literacy as a three-year capability programme, not a one-off course. Here’s a practical way to frame it in your organisation.

1. Start with the problems that matter

Anchor literacy in the issues that are already on your Board’s agenda:

• Improve complaints resolution satisfaction from 36% towards parity with overall TSM satisfaction.
• Reduce emergency repairs and repeat visits by increasing first-time fix rates.
• Demonstrate stronger compliance with Awaab’s Law and building safety requirements.

Define 3–5 specific metrics for each (e.g. days to first inspection for damp, percentage of complaints resolved at first stage, repairs jobs resolved via remote diagnostics). Literacy activities then become a means to those ends, not an abstract concept.

2. Segment your workforce, and tailor what “literate” means

Different groups need different capabilities:

• Frontline repairs and housing officers: focus on why accurate categorisation matters, how to use AI triage tools, when to escalate, and how to explain data-driven decisions to residents.
• Contact centre and complaints teams: reading and interpreting dashboards, spotting trends, using AI summaries as a starting point rather than a script, and managing bias and fairness concerns.
• Managers and heads of service: framing questions for analysts, understanding model performance, balancing efficiency and equity, and using data to redesign processes.
• Executives and Board: understanding the limits of AI, oversight responsibilities under new regulation, and how to link data investments to tenant outcomes and risk reduction.housing-technology+1

Data Literacy Academy’s approach takes current capability and roles into consideration when proposing cohort education tracks. Every cohort requires specific criteria of success, which is underpinned by our rigorous Value Realisation Framework. Where most training goes wrong, is that it assumes that learning translates into impact. But it often ignores the fact that change management, strong internal feedback loops, use case capture and ongoing executive engagement are all required to deliver measurable outcomes. While proving ROI will always remain a challenge, when it’s baked in from the start it delivers a clear roadmap of a before and after picture that far goes beyond learner engagement.

3. Embed literacy into real workflows

Avoid generic “data literacy” courses divorced from daily work. Instead:

• Use live repairs, damp and complaints cases in training, so staff see how their data entries flow into dashboards and AI tools.
• Pair new AI deployments (e.g. predictive maintenance, income management automation) with targeted learning sprints for the people who will use them.
• Create simple, visual data dictionaries and process maps that show which fields matter for Awaab’s Law, TSMs and regulatory reporting.

Customised learning is a key reason why Data Literacy Academy’s programme have such high success rates. By tailoring learning objectives to people’s lived experience, it results in better implementation of the education, stronger collaboration among learners, and a clear alignment of the programme to the organisation’s strategic objectives.

4. Build governance that people can actually use

Effective governance is not just policies and committees. What matters is everyday behaviour.

• Co-design “rules of engagement” for AI tools: which use cases are approved, what human-in-the-loop checks are required, and how to log decisions.
• Make model behaviour visible: simple explainability views that show why a repair was prioritised, or why a damp case was flagged as high risk.
• Align literacy and governance: ensure training explicitly covers how staff responsibilities connect to the Digital Golden Thread, Awaab’s Law and consumer standards.

A proactive safety culture can be nurtured by documenting storage to real-time data assurance, showing how digital twins work and integrating case management. But it requires people to understand and trust the data behind them.

Positioning literacy as your strategic advantage

For data leaders, the opportunity is to reframe data and AI literacy as a strategic investment that:

• De-risks AI and advanced analytics projects by improving data quality, governance and adoption.
• Directly supports regulatory compliance and evidencing under Awaab’s Law, the Social Housing (Regulation) Act and building safety regimes.
• Improves core tenant outcomes, especially in complaints resolution, repairs performance, and damp and mould, where the sector is under the most pressure and where TSM gaps are most visible.
• Builds organisational resilience in the face of growing workloads, skills shortages and public scrutiny, by empowering staff to use AI as a force multiplier rather than a threat.

The future of housing lies in turning fragmented data into connected intelligence and building the right foundations that allow AI to deliver real impact.  Those foundations are not just technical. They are human, and they start with data and AI literacy.

‍