"We needed to really think about how we can turn our research into something tangible that people could easily use.”
The challenge
Unions 21 had produced substantial research and insight on organisational culture in the union movement as part of their Union Operating Models work. But research sitting in reports only goes so far. The challenge was making that insight feel real and actionable for the people who needed it most – union staff, senior leadership teams, and boards.
Reflection on organisational culture is not something unions always have time for. Day-to-day pressures of representing members, running campaigns, and managing complex organisations often mean that stepping back to think about how the organisation itself works gets pushed down the priority list. Becky Wright, Executive Director, wanted to offer something that could help that process – something accessible to staff at all levels, from frontline officers to General Secretaries.
The question was: how do you take detailed research findings and turn them into something people will actually use?
The solution
Becky chose Loveable, an AI-powered no-code platform that allows users to create full web applications by describing what they want in plain language. The platform is already used in our AI Fluency training programme, where union staff are encouraged to experiment with it to build an AI assessment tool for colleagues to use. So Becky decided to experiment too – and build an interactive culture diagnostic tool.
The tool asks users 32 questions drawn directly from Unions 21's existing research. Based on the responses, it determines:
The primary and secondary culture type of the user's union
Whether that culture is productive, unproductive, or agile
Users receive two outputs: a spider diagram showing the distribution of their results across culture types, and a written diagnosis explaining what their results mean in practice.
A key design decision was to make all responses anonymous. No personal data is collected, which removed the need for complex data protection approvals and made it straightforward for unions to share the tool widely with their teams.
"In our AI training we’re always encouraging mindful experimentation, so I thought – why not give this a go?"
How it was built
The tool was built in about two weeks by someone – Becky – who has no coding skills and expertise. The process was iterative – she tested, adapted, and learnt the platform's capabilities along the way.
Before launching, the Unions 21 team tested the tool with people from different unions and different roles, making adjustments based on their feedback. This cross-role testing was important for ensuring the questions and outputs made sense to people in very different positions within the movement.
The most important lesson from the build process was the value of thinking carefully about what you want to achieve before you start building. The technology can move fast, but being clear on your goal makes the iterative process much smoother.
"While I need no technical expertise, it's really useful to properly think through what you want to build and what the outputs are."
Key benefits
People are interrogating research in a way that feels accessible to them. Within a few months of launch, the tool had been used over 160 times. Those who have used the tool have said that the results either a) felt exactly right and confirmed what they suspected or b) were unexpected but, on reading the diagnosis, felt accurate.
The tool’s success is helping generate more data, improving our insight about the union movement. This volume of responses is now generating concrete, aggregated insight for Unions 21 about culture across the wider union movement – an unintended but valuable benefit.
When unions see an easy-to-use tool, they want to use it on themselves. Individual unions have since asked to run the diagnostic internally with their own teams – a use case Unions 21 hadn't originally anticipated. This demonstrates how a well-designed tool can find applications beyond its original purpose.
Low cost and no technical maintenance. The tool costs less than £50 per month to run and requires no technical expertise to maintain, making it ver sustainable.
"I wasn't anticipating unions to come forward and ask to use the tool internally. I thought it would just be a broader movement tool."
Common questions
What were the main challenges in creating the tool?
The early challenges were mostly user error rather than platform limitations. One key lesson: remember to link your app to a database from the start. If you're collecting responses, thinking about data storage and what you'll do with the information needs to happen upfront, not after launch.
Beyond that, the main investment is the time spent thinking through requirements before building. The technology itself was not the bottleneck –clarity of purpose was.
Do you need technical skills to build something like this?
No. The tool was built with no prior technical expertise using an AI-powered no-code platform. However, you do need to think clearly about what you want to build and what the outputs should be.
How long does it take and how much does it cost?
The tool was built in approximately two weeks, working iteratively. The tool costs less than £50 per month – for the subscription to Loveable.
What about data protection?
By designing the tool to collect no personal data and keeping all responses anonymous, Unions 21 avoided the need for complex data protection approvals. This is worth considering from the start of any similar project.
If you want to try this
If you're thinking about building something similar, here's what Becky suggests:
Start with a clear purpose. Ask yourself: what do you really want to achieve? Don't build something just because the technology is exciting.
Think about data from the start. If you're collecting responses, how will you store that information and what will you do with it?
Know what you want to build before you start. Clarity upfront makes the iterative building process much smoother.
Test with real users before launching. Different roles and contexts will surface issues you wouldn't catch alone.
Be creative, but responsible. AI-powered tools open up new possibilities, but think through the ethical implications of what you're building and how it will be used.
AI Transparency statement
This case study was generated using AI based on conversation with Becky Wright.
