Culture Change in Social Housing

Two housing associations merging into one culture

When Peabody and Family Mosaic merged, the immediate focus was structural - combining departments, aligning systems, reorganising teams across 66,000 homes. But the harder question was cultural. How do you bring together two organisations with different ways of working, different relationships and different identities, and build something genuinely shared?

Peabody is one of the oldest and largest housing associations in the UK and a member of the G15 group of London providers. Family Mosaic was a substantial organisation in its own right, with its own history and its own way of doing things. Both had run change programmes before - around coaching culture, lean thinking and customer experience - that people had invested in.

Why culture isn't something you can install

Most merger programmes answer the culture question by defining a new set of values and cascading them. But culture doesn't work that way. It isn't something you define and install - it's an emergent property of how an organisation works. It's the product of its systems, its service patterns, its team practices and the thousands of small interactions that happen every day.

The CIPD describes this as the difference between culture and organisational climate: culture is the deep pattern of shared assumptions, while climate is what people experience through the policies, practices and behaviours around them. You can't mandate culture directly. You can shape the climate that produces it.

The gravitational pull in a housing association merger

A housing association merger makes this harder in ways that go beyond logistics. People lose things that don't show on an org chart - the informal networks they relied on, the rhythm and identity of their team, the shorthand that made daily work predictable. Everyone is already stretched by restructuring and uncertainty about their own role, which is exactly when people retreat to what's familiar rather than reaching toward something new.

And there's a pull that most mergers don't acknowledge openly. Without deliberate attention to post-merger integration, the larger organisation's way of doing things becomes the default. "Taking the best of both" is easy to say and hard to do when one side's systems, processes and leadership are already in place. The result is absorption rather than integration, and people on the absorbed side feel it even if nobody says it. With Peabody the larger partner, this was a live risk rather than a theoretical one.

The opportunity inside a merger

Mergers also create something that's hard to find at any other time: permission. Both organisations can be honest about what wasn't working. Practices normally defended because "that's how we've always done it" are suddenly open to question, because everything is changing anyway.

Things that are hard to challenge in normal times become fair game - service processes nobody designed deliberately, communication habits that don't serve customers, siloed working, performance frameworks that measure compliance, and previous change programmes that didn't land but nobody wants to say so out loud. A merger is one of the rare moments when genuine cultural shift is possible, if an organisation is willing to use it that way.

Why culture change works differently at 66,000 homes

In a 50-person organisation, culture lives in who you know and how your colleagues behave. At 66,000 homes it can't travel through personal relationships - it has to live in systems. In how services are designed, how teams communicate, how decisions get made.

That's what made this approach to culture change in social housing different from a values-led programme. The work had to change the conditions that produce culture, not describe what the culture should be. And it had to do that while protecting customer experience through the period when internal disruption was at its highest, which meant culture change couldn't be a separate initiative competing for attention. It had to be woven into the work of improving services.

Culture change in social housing built from the ground up

Most culture change programmes in social housing follow a familiar pattern: define the values, cascade the message, train the behaviours, measure compliance. It's logical, well-intentioned, and it rarely produces lasting change - because culture isn't shaped by what an organisation says it values, it's shaped by what people experience day to day. The gap between the two is where culture programmes quietly fail.

The case for values, taken seriously

That critique is easy to make and worth resisting for a moment, because the values route has a real argument behind it and a merger is where that argument is strongest.

Two organisations becoming one need a shared language fast. People need something to point at when they ask what kind of place this is now. Values are quick to produce, easy to communicate, and they give leadership something visible to stand behind at exactly the moment people are looking for reassurance. In a small organisation, or a genuinely new one, values can work - they travel through relationships and there's no accumulated practice contradicting them.

The problem isn't that values are wrong. It's that at this scale they arrive as adjectives, and adjectives don't tell anyone what to do differently on Monday. "Caring" doesn't resolve whether you send the standard arrears letter. A merged housing association of thousands needs a shared language - but one made of questions rather than adjectives, because a question has to be answered in context and an adjective can be agreed with and ignored.

From values to practice

So the objective was to ask "what does good look like in how we actually work?" rather than "what do we want our culture to be?" That's a shift from values as aspiration to values as practice - from what's on a poster to what a customer experiences when they call, or how two teams work together to solve a problem.

That meant designing culture change people could see, feel and use in their daily work. And it meant accepting that we couldn't prescribe how everyone should work. What a repairs team needs is different from a care and support team, which is different again from a contact centre. The aim wasn't uniformity. It was synthesis - getting a whole organisation using the same conceptual framework to talk about its work, without dictating the answers.

Starting with what customers and colleagues need

Finding the pattern under the noise

The starting point for the culture change was evidence, and there was a lot of it: customer data and complaints, service performance, employee insight from both organisations, and time with frontline teams, senior management and the CEO.

Most programmes handle that material in two streams. Customer experience is one workstream, employee engagement another, and each produces its own set of findings, its own priorities, its own plan. It's a sensible division and it hides the most useful thing in the data.

Read together, the two converge. What customers said they wanted turned out to be, in different words, what staff kept pointing at as the thing they wanted more room to do. Customers described being treated like a case rather than a person; staff described processes that made them sound like one. Customers wanted to sort something out in one go; staff were frustrated at handing people between teams. Neither group was asking for a culture. Both were describing the same handful of things from opposite sides.

That overlap is where the real work was, and finding it took time. It meant staying with contradictory material longer than felt comfortable, resisting the urge to organise it, and looking for the shape underneath rather than the categories on top. But what comes out of that is worth the patience: a truth that holds for customers and colleagues doesn't need selling to either. Nobody has to be persuaded of it, because they said it first.

That's the investment we chose to make. Most programmes don't, and the reason is understandable rather than lazy. A rich framework - four pillars, sixteen behaviours - looks like it honours the complexity of what you found, and building it shows visible progress every week. Distillation looks like the opposite: weeks with the same material, and you come back with three sentences. It looks like less. It's the harder thing to do and the harder thing to sell.

One distinction matters here, because it's where the argument usually gets misread. Simplicity belongs in the framework, not the delivery. Behind three questions sat real machinery: nine workstreams covering quick wins, cross-departmental sessions, team support, leadership, the Heads of Service toolkit, performance management, internal engagement, customer voice and measurement. That isn't a contradiction. A simple idea needs solid delivery precisely because it's carrying so much. What you're avoiding is complexity in the thing people have to hold in their heads, not complexity in the work.

The three questions

Three areas came out of it, and they became three questions:

1. Is our service designed around the needs of our customers?
2. Are we working well with other teams and colleagues?
3. Are we being human and kind in the way we communicate?

They look obvious. They're supposed to - that's what a universal truth looks like once it's been found. The work was in the finding, not the phrasing.

Each one sits on the overlap. "Human and kind" is the clearest to trace: customers weren't asking for customer-centricity, they were describing, over and over in different words, letters that felt hostile and staff who sounded like a process. Take away the vocabulary of a few hundred individual complaints and that phrase is what's left. Because it came from customer data rather than a workshop, it arrived with an authority no internally generated value could have.

Four tests decided what stayed. This is the transferable part:

• Is it true everywhere? If it only holds for frontline teams it's a policy, not a truth. A finance team has to be able to answer it about the colleagues it serves.
• Can a team act on it this week? "Be more innovative" can't be acted on. A truth that works leaves a team able to change a form, a letter or a handover before the next session.
• Does it come from evidence rather than aspiration? Anything we wanted to be true, rather than found to be true, came out. That removed most of the first list.
• Does it make a team look, or let them agree? The real test, and the one most candidates failed. The enemy isn't disagreement, it's quiet compliance. "Do we care about our customers?" gets a yes from every team in the organisation and changes nothing. "Is our service designed around the needs of our customers?" can't be answered without going and looking at the service. The question has to do the work - if it can be satisfied by nodding, it's a value, not a question.

Three areas passed. Everything else, however true, went.

Built through sessions, not launched as a programme

The programme had a name - People First - a brand, and everything needed for a launch. We deliberately didn't launch it.

No all-staff email, no firework display. People First was introduced through action-focused workshops with teams, always opened by the CEO or a senior sponsor, always focused on practical application rather than abstract ideas about culture. It was soft-launched with the executive and a range of frontline teams rather than the whole organisation at once, so that when it did spread, it spread on evidence that it worked rather than on an announcement.

The sessions took away the ambiguity of "culture" and replaced it with something concrete: how things work here. Instead of asking people to change how they felt about the organisation, the sessions asked how the work could be done differently in their context. That's a conversation people can engage with regardless of how they feel about the merger, the leadership, or the idea of culture change.

How the culture change actually spread

The challenge we needed to address early was this: if you don't have a big launch, how does the work spread?

The answer was simple, if you make the content engaging enough.

People First spread by word of mouth. Through the water cooler, the informal networks that survive any reorganisation, colleagues telling colleagues it had been worth a day. Demand for sessions came from people who'd heard about them from someone they trusted, not from a communications plan.

That's why it could stay optional. We didn't need to mandate attendance, because peers were generating the demand - and a recommendation from someone at the next desk carries something no all-staff email can buy. It also cost nothing. The most effective distribution channel in the organisation was one that already existed and was free.

Most importantly, it sidestepped the "done to us" problem that follows so many change programmes around. A programme that arrives by decree is something happening to you. A programme you heard about from a colleague who found it useful is something you might choose. Same content, completely different relationship to it.

The feedback from those first sessions was consistent enough to be a finding in its own right, and it kept arriving unprompted weeks after the event. Some of it was about the work - people naming, for the first time, things about their own service they'd stopped noticing. Some was about the organisation, from people who'd been through years of restructuring and found that hearing directly from the top, in a room, changed what they thought was happening. Several said it was the best session they'd been to in a career measured in decades.

But the mechanism carries a real risk, and it's worth naming because it's the reason this route isn't a free option. Word of mouth only works if the thing is genuinely good. A mandated programme can be mediocre and still reach everyone; an optional one that relies on peer recommendation has to be worth recommending, every session, from the first. If those early workshops had been unremarkable, nothing would have spread and there would have been no fallback - we'd deliberately given up the mechanism that would have carried a weaker approach.

Principles agreed before anything started

Before any of it ran, we agreed five principles with the steering team. They weren't communications material - they were a decision framework for the arguments we knew were coming:

• Everyone will have a voice, and a chance to make a difference
• It's about internal and external customers alike
• This will be co-designed with people, not imposed on them
• It won't be one-size-fits-all - teams customise with support
• The long-term aim is to give teams the tools to do this for themselves

Agreeing those upfront is what made it possible to say no later, when the pressure came to mandate attendance or standardise what teams produced.

Two levels, one loop

The work ran at two levels simultaneously.

Leadership worked on systems and environment - the structures, policies and conditions that shape how people are able to work. This is where the systemic blockers sit: approval processes that slow things down, communication templates that feel hostile, performance frameworks that measure the wrong things.

Teams worked on their own area - taking the three questions and working out what they meant in practice. What does "designed around the needs of our customers" look like for a housing management team? A contact centre? A care and support team?

That split matters because it addresses the most common reason culture change stalls. Work only with teams and they hit systemic barriers they can't fix. Work only with leadership and you get policies that look good on paper and change nothing at team level.

The two levels were joined by a feedback loop. As teams worked through the questions, they surfaced what needed to change structurally, and that went straight to leadership. Leadership made changes. Teams saw them happen. Ten high-priority challenges were picked as early wins - things that could be fixed quickly without disruption - precisely so the loop had something visible to show early.

What came back up that loop was more useful than a survey would have been, because people weren't rating the organisation they had. They were describing the one they wanted to be part of:

• Measures that reward the experience they were being asked to create, so that reducing a customer's effort and hitting your numbers pull in the same direction
• The trust to write to a resident in their own words, rather than through enough layers of review that the letter comes back unrecognisable
• Teams that finish what another team started, instead of each optimising for its own targets
• Ways of handling arrears and enforcement they could stand behind when they knocked on the door
• Managers who treat time spent on this as part of the job

Read as complaints, that list is uncomfortable. Read as a specification, it's a gift - it's the merged organisation described by the people who would have to work in it, in enough detail to act on. None of it is fixable by a single team, and none of it would have come out of a questionnaire. It came out because people were asked a concrete question about their own work and given somewhere for the answer to go.

That's what built trust. Not the asking, but what happened next - a loop where leadership demonstrated the culture it was asking for rather than describing it. It does mean a leadership team has to be willing to hear that some of the answer sits with them, including in how they measure. That's the entry price for this approach, and it's higher than a values workshop.

Small changes, deliberately

The challenge to every team was small: what can you do in your own part of the organisation to begin shifting how you work?

That was deliberate, for three reasons. Aggregation - hundreds of teams making small changes is the system change; no single team has to transform anything. Disruption - during a merger, small changes are manageable, and a transformation programme stacked on top of a merger overwhelms people. And most importantly, it bypasses "someone else's job" - when culture change is only about big structural moves, most of the organisation concludes it isn't their problem to solve.

Underneath all three sits a principle that was agreed at the start and did more work than any of them: the people closest to a problem should have the power to fix it. Most organisations agree with that in the abstract and are structured to prevent it. Making it real is what turns a small change from a suggestion into something that actually happens.

The pilot sessions alone produced more than fifty ideas, and the range is the point. At one end, automatically refunding leaseholders who ended the year in credit rather than making them ask - money returned to residents who were owed it and would mostly never have claimed it. At the other, a positivity jar on a desk.

The positivity jar is the one worth dwelling on. It was one person's idea, in one team, and it worked so well that other teams picked it up and ran their own. Reception ended up with a version residents could contribute to. No programme would have commissioned that. Put "introduce a positivity jar" in a change plan and it dies of embarrassment in the first review. Owned by the person whose desk it sat on, it became something else entirely, and then it travelled on its own.

That's the argument in miniature. The change wasn't imposed and then explained. It was made, locally, by someone who wanted to make it - and other people saw it working and wanted one. The culture was being demonstrated rather than announced, which is the only way it ever moves.

So "small" here isn't doing the work of "trivial". It's doing the work of "local". The person who knows the year-end credit is sitting there is the person processing it.

The three questions in practice

Customer experience and service design. The first question drove a focus on customer experience and service design. With every service being reviewed during the merger anyway, there was a natural opening to balance internal process against a genuinely customer-centred approach.

The move was from satisfaction measurement to customer effort - how much work it took someone to get what they needed. Effort is actionable in a way satisfaction isn't. If a process takes fourteen steps you can redesign it; if satisfaction is 6/10 you don't know where to start.

The decision that made this work organisation-wide was defining "customer" to mean internal as well as external. Every team, wherever it sat, answered two questions in order: who do we serve? and then how much effort does it take them to get what they need from us? A finance team, an HR team, a repairs scheduling team - none of them talk to residents daily, and all of them have customers.

That did something a customer experience programme aimed only at frontline teams can't. It gave the whole housing association one measure of whether the work was any good: low effort, easy to deal with, the customer gets what they need. Back-office teams stopped being the audience for customer experience and became participants in it. That is what made customer experience an organisation-wide measure rather than a frontline one. And because internal and external effort are connected - staff losing time to bureaucracy is a cost customers eventually pay - improving one tended to improve the other.

Working together across departments. When people are anxious, they retreat to what's familiar. The "us and them" that forms after a merger isn't a personal failure, it's a predictable response to disruption - and treating it that way makes it easier to design for.

Cross-departmental sessions were opt-in. That was a choice, not a compromise: mandating attendance at culture change sessions during a housing association merger would have contradicted the message, and it broke the third of our five principles. Opt-in also filtered for people who were genuinely motivated rather than compliant. Around 20% of the housing association engaged through those early sessions, well above most tipping-point thresholds, and those early adopters became the champions who later ran sessions with their own teams alongside the Learning and Development team.

That's standard adoption-curve thinking, applied deliberately: put the energy into the early adopters rather than spending it trying to drag along the disengaged. The majority in any organisation are waiting to see evidence, and the only thing that produces it is other teams' results.

Human and kind communication. This started with something concrete: letters and notices that came across as hostile, confusing and unnecessarily formal, damaging relationships with residents and creating tension where none needed to exist.

Here's what came out of one session. A team rewrote the letter that offers residents contents insurance - the kind of letter that normally opens by announcing a scheme:

Three things happened there. It starts inside the reader's flat rather than inside our processes. It uses the words a person would use. And it volunteers something nobody asked about - that we take no commission - because the team realised the offer wasn't believable without it.

Nobody could have written that from a values statement. They wrote it from a question.

"Human and kind" also turned out to travel further than communication. Teams began using it as a test for how they behaved and how they made decisions, not only how they wrote - which makes sense, because communication doesn't just reflect a relationship, it creates one.

Closing the capability gap for managers

Asking "is our service designed around the needs of our customers?" is simple. Doing it needs skills most managers don't have - service mapping, facilitation, customer research, running sessions with their own team. That gap is a common reason culture change programmes stall: people buy the principle and can't act on it.

A core toolkit closed it - customer experience, effort and service design (including a Customer Experience Scorecard teams used to assess their own service across effort, working together, human and kind, and customer-led); cross-departmental problem solving; and how to write in a simpler, more human way. Specific teams got tailored support on their own challenges, including care and support, complaints, service charges, rent collection and HR.

For leaders, we ran "Leading the Way" workshops that brought Heads of Service and Directors together with people they didn't normally work with, and built a leadership programme with the internal L&D team.

Performance management, rebuilt on the same three questions

What an organisation measures tells people what it actually values, regardless of the poster. If performance management doesn't align with the culture you're building, it actively undermines it.

The existing framework used complex assessment criteria - the kind that become a box-ticking exercise. We replaced them with something simpler: people demonstrated how they worked to the three principles. The focus moved from scores to the conditions for good conversations - honest dialogue between peers about how they were working, rather than whether they'd hit a number. For leaders, a simplified 360 looked at the environment they created and whether it supported or blocked the three areas.

A culture built gradually through practical, visible changes

The system running without us

The clearest sign something had shifted was that the loops became visible and were turning without external intervention. Teams took the three questions and made them their own. They fed structural blockers upward. Leadership changed things. Those changes showed up in how customers experienced the organisation.

That circular pattern - teams owning their part, leadership responding, improvements reaching customers - is what embedded actually looks like. Not a programme still running, but a way of working that sustains itself. Three years after the work began, the three questions were part of the customer voice at both organisational and team level.

What changed for residents

Customer experience measurement shifted to effort and real-time feedback rather than retrospective satisfaction scores, which gave teams something they could act on directly. Cross-departmental problem solving became ordinary rather than something requiring an initiative. A merger that could have been defined by disruption and lost identity became the occasion for building something more intentional than either organisation had been before.

Designed to make us unnecessary

The move from externally supported to internally owned wasn't a handover at the end. It was designed in from the start: champions ran sessions with their own teams supported by L&D, managers used the toolkit day to day, and leadership modelled the approach and responded visibly.

A monthly Pulse Check gave teams a simple diagnostic across eight areas, from enabling environment and innovation to cross-team collaboration and customer focus, with a traffic light system to prioritise where to act. It wasn't measuring success or failure - it was reading the conditions, on the basis that culture isn't something you do once.

Alongside it, a maturity model gave the organisation a way to locate itself and see what came next. Five stages: tactical, where good practice happens in pockets and solves specific problems; planned, where it becomes deliberate rather than reactive; evaluated, where it's measured on a schedule against objective criteria; embedded, where it's business as usual; and cultural, where it's a core part of the operating model and of how the organisation plans. The value of naming the stages isn't the scoring. It's that "we're at planned, and evaluated is next" is a conversation an organisation can act on, where "our culture needs to improve" isn't.

What was hard about culture change at this scale

Culture change in social housing of this kind takes 18 to 24 months to take hold, and we said so at the start.

It wasn't frictionless. There were people who'd watched programmes come and go and had good reason to be sceptical. There was constant tension between wanting visible results and knowing this needed time. Some teams engaged immediately and others were much harder to reach, which required patience rather than pressure. The feedback loop surfaced uncomfortable truths that leadership then had to respond to. And all of it happened while the merger itself created continuous disruption and competing demands.

None of that is unique to this project. It's what working with culture as a living system looks like, rather than a programme with a start and an end date.

Why culture change in social housing needs distillation, not another framework

The instinct, faced with a culture change problem, is to design something: a bespoke framework, a tailored competency model, a programme architecture. It feels like the serious response. It demonstrates effort proportionate to the problem.

But complexity in the framework is usually a sign that the analysis stopped early. A model with sixteen behaviours is what a mountain of evidence looks like when nobody reduced it. And every layer of that model is a layer someone has to interpret, explain and champion before anything changes - so by the time it reaches the people doing the work, the energy has gone into understanding the concept rather than changing the practice.

The alternative is harder, not easier. Staying with contradictory evidence long enough to find the pattern underneath is slow, and it produces something that looks slight. Three questions don't look like six months of work. That's exactly why most organisations don't do it - not because they don't know they should, but because distillation is invisible while framework- building is demonstrable.

That's the transferable lesson. The value wasn't in what we found. It was in what we were willing to throw away.

Why it works: artefacts first

Edgar Schein's three levels of culture explain the mechanism. Schein describes culture as operating at three levels: artefacts (the visible things - how services are delivered, how people communicate, how decisions are made), espoused values (what an organisation says it believes), and underlying assumptions (the deep, often invisible beliefs that drive behaviour).

Most culture programmes work on the middle layer. They define values and hope those values change the artefacts above and the assumptions below. This project bypassed that layer. It worked directly on artefacts - service design, communication, team practice, performance management - and let the assumptions shift as a consequence.

That's the significant move. Rather than telling people what to value and hoping behaviour follows, we changed what people did and let the meaning follow. Nobody needed to be told customer experience mattered. They experienced it mattering, through redesigned services, through effort data that showed where things were broken, through leadership responding to what teams raised. The values became real because the practice made them real.

It's also why the three questions worked as leverage points for culture change. They're places where a small shift in thinking ripples across a whole system, which is what let hundreds of teams making small changes add up to system-wide shift without a system-wide mandate.

Culture change and the Tenant Satisfaction Measures

The Regulator of Social Housing's Tenant Satisfaction Measures now require every housing association to report on exactly the outcomes this approach was built around - satisfaction with services, communication, and how landlords listen and act. The TSMs measure things that are shaped by culture: how residents experience an organisation, not what it says about itself. Organisations still relying on values statements to meet them are working against the grain.

The Competence and Conduct Standard asks the same question

From October 2026, the Competence and Conduct Standard goes further. Alongside qualification requirements for senior housing staff, it sets expectations about conduct - how people actually behave toward residents. That is a meaningful shift in what a regulator is willing to look at, and it lands squarely on the territory this project worked in.

Conduct is an artefact, in Schein's sense. It is visible, it happens in specific interactions, and it can be worked on directly. It is not an espoused value, and it cannot be reached by declaring one. A housing association that meets the standard by writing a behaviour framework and running training will produce people who can describe the expected conduct. One that changes what its letters say, how its handovers work, and what its performance conversations reward will produce the conduct itself.

Read that way, the three questions are a conduct test that happened to arrive years before the standard did. "Are we being human and kind in the way we communicate?" isn't a value. It's a question a housing officer can answer about a letter sitting in front of them, right now, and act on before the end of the day. That is the difference the standard is going to expose.

We've since built a practical guide for housing leaders preparing for it - the Competence and Conduct Standard toolkit.

Where to start

For any housing association facing significant change - a merger, a restructuring or the demands of a new regulatory standard - the question worth sitting with isn't what your values should be. It's whether you've done the work of finding out what's actually true about how you operate, and whether you'd have the nerve to reduce it to something small enough for everyone to hold.

If the merger is the pressing part, our article on culture after a merger covers why structural integration so often completes while the organisation still feels like two. For the wider practice, see developing organisational culture.

About this work

James Freeman-Gray was principal consultant on this programme. He developed the approach, worked with the executive team, and ran the workshops it spread through.

See more of our work across social housing, including another housing association merger integration.