Housemark AI maturity framework
The research demonstrates that AI adoption in social housing is not a binary choice between using AI and not using AI. Organisations are progressing at different speeds, with maturity shaped by leadership, strategy, governance, workforce capability, data quality and the extent to which AI is embedded in day-to-day work.
Drawing on the survey findings and Housemark’s sector expertise, this report proposes a practical four-stage AI maturity framework. The model is intended to act as a benchmark and guide for progression rather than a definitive scorecard.
Stage 1: Experimenting
AI use is informal, individual and often driven by enthusiastic early adopters rather than organisational direction. Activity typically focuses on content creation, research, summarisation and personal productivity.
This is currently the most common stage identified in the research, with 26 respondents describing their organisation's approach as individual experimentation. Separate findings across the full respondent base show that 11 respondents reported having no AI strategy in place and 46 said they were not currently measuring AI's impact. These wider findings provide context for the maturity framework but should not be read as characteristics of the 26 respondents in this stage specifically.
Individual experimentation
No AI strategy
Not currently measuring AI’s impact
So, how do we move from isolated experiments to coordinated organisational learning?
Signs of progress include clearer ownership, basic guidance and mechanisms for sharing successful use cases. The next step is to turn individual learning into coordinated activity.
Stage 2: Coordinating
AI adoption is becoming more organised. Teams are beginning to share good practice, governance is emerging and organisations are starting to identify specific use cases where AI can add value.
Nineteen respondents described their organisation as having reached team-level adoption. Separate findings across the full respondent base show that 27 respondents were developing an AI strategy and 39 said all or most staff had access to AI tools. These measures provide wider context and are not attributed specifically to the 19 respondents at this stage.
Reached team-level adoption
Developing an AI strategy
All or most staff have access to AI tools
The question is, how can successful use cases be scaled beyond individual teams?
Signs of progress include clearer ownership, basic guidance and mechanisms for sharing successful use cases. The next step is to turn individual learning into coordinated activity.
Stage 3: Operationalising
AI is increasingly integrated into operational activity and business processes. Organisations begin using AI alongside organisational data, reporting systems, workflows and service delivery functions.
Seventeen respondents described their organisation as operating coordinated AI initiatives. Across the respondent base as a whole, 41 respondents, 62%, agreed or strongly agreed that AI was delivering measurable value and 43, 65%, agreed or strongly agreed that AI had improved productivity. These reported benefits cannot be attributed specifically to the 17 respondents at this stage, but they show the wider context in which more coordinated adoption is developing.
Operating coordinated AI initiatives
Agree or strongly agree AI is delivering measurable value
Agree or strongly agree AI has improved productivity
So, how can operational benefits be converted into measurable organisational outcomes?
Signs of progress include clearer baselines, outcome measures and links to strategic priorities. The next step is to scale use cases that demonstrate value.
Stage 4: Transforming
AI is treated as an organisational capability rather than a collection of tools. Leadership, governance, data, skills and measurement frameworks are aligned around clear business objectives.
The research suggests that some of the foundations associated with transformation remain uncommon. Only seven respondents reported having a formal AI strategy, just three said they measured AI impact consistently and 21% agreed or strongly agreed that they could demonstrate return on investment. These measures do not combine to provide a definitive maturity score, but they indicate the scale of the gap between current activity and more mature adoption.
Formal AI strategy
Measuring AI impact consistently
Agree or strongly agree they could demonstrate return on investment
How can AI become a repeatable source of organisational value and better service outcomes?
Signs of progress include consistent measurement, governance and organisational learning. The next step is to refine, scale or stop AI activity according to the outcomes it delivers.
What the research tells us
The findings suggest the respondent base is spread across multiple stages of maturity. Access to AI tools was widespread and practical use cases were reported across housing organisations. However, many respondents described organisations that are still developing the foundations needed to support wider adoption, including:
- Stronger Data
- Clearer Governance
- Workforce Capability
- Measurement Frameworks
The challenge for the next phase of adoption is not simply gaining access to AI technology. It is helping organisations move confidently from experimentation to coordination, from coordination to operationalisation and ultimately from activity to measurable impact.
Housemark expert insight:
Where does your organisation sit?
One of the strongest messages from this research is that AI maturity is not determined by the number of tools an organisation has purchased. Housemark believes meaningful progress depends on how effectively technology is connected to organisational priorities, supported by leadership and translated into practical use cases.
For organisations at an early stage, the priority may be to establish clear ownership, give colleagues practical guidance and create ways to share what is being learned through experimentation. The aim is not to stop people exploring AI, but to turn isolated activity into organisational learning that can be assessed and repeated.
As adoption becomes more coordinated, the questions change. Leaders need to understand which use cases are worth prioritising, whether the organisation has the data and skills to support them and what proportionate governance is needed. Successful experiments should not automatically be scaled. They should be connected to a clear business or service objective and assessed against the outcome they are intended to improve.
Organisations should also avoid treating the four stages as a race. Different services may progress at different speeds and an organisation can be advanced in one area while still developing foundations elsewhere. The framework is therefore less about assigning a score and more about identifying the gaps that matter and the next practical step.
For more mature adopters, measurement becomes increasingly important. Leadership, governance, data, workforce capability and evaluation need to work together so organisations can distinguish activity from impact and decide which applications to refine, scale or stop. For many housing providers, the next step is therefore not buying another tool, but strengthening the foundations that allow existing technology and proven use cases to deliver sustainable value.
Jonathan Cox
Chief Data Officer, Housemark