“Working with DataKind really helped us to understand our own data and our own processes better... It really did exceed all our expectations.”

Nathan Palmer
Head of Programmes, St John Ambulance

St John Ambulance responds to health emergencies, supports communities, and saves lives. The organisation’s clinical expertise and the skills of St John people make it unique; a volunteer-led health and first aid charity, with national presence, reach and scale. Volunteers have relieved people from illness, injury, distress and suffering for over 140 years. They are also playing an essential role in the effort to get England safely vaccinated against Covid-19.

Volunteers give their free time to support to the public, making St John Ambulance the nation’s leading first aid charity. In 2020, they spent 125,000 hours on the road in ambulance crews, and 125,000 hours volunteering in hospital emergency wards. In a specialist training programme developed with the NHS, they trained over 4,000 volunteers in Covid care and logistics. This was on top of their usual winter support for the NHS, training 60,000 people in first aid and doing 14,000 hours of support to ease winter pressure on the NHS before the UK lockdown in March 2020.

With a five-year plan to improve their services on the way, St John wanted to get to grips with their volunteer data. To know its volunteers better, understand what drives them, and ensure they are getting the support and resources they need, so that we all continue to benefit.

This project, run in October and November 2020, used historical data reflecting a volunteering cohort and programmes up to 2020. St John Ambulance has since recruited almost twice as many volunteers again to support the NHS Covid-19 vaccination programme, and the charity continues to adapt how they will manage and recruit volunteers during and after the pandemic. While COVID has delayed the implementation of some of the key learnings, and changed the context of some of the questions, the work is still key to support the understanding, and growth, of St John’s volunteer network, post-pandemic.

What did they find out?

St John Ambulance were able to look more holistically at their volunteers in order to build ‘pen portraits’.
They came away with multiple ways to cluster their volunteers that will help them assess their engagement.
Volunteers’ demographics were generally very representative of their regions. SJA were able to identify diversity and representation issues to improve on in future.
They were pleased to find that their retention rate was much higher than they anticipated, and can focus on improving engagement with specific roles that see higher churn.
They have a way to measure and compare resourcing needs, helping them to be more efficient when staffing events and other activities.
They found that their data provided lots of useful segmentation, and can make the case for better data collection and use of external data sources to improve this.

What are St John Ambulance’s data challenges?

St John worked with DataKind UK on a two-month DataDive project to look more closely at their volunteer engagement, map volunteering activity, and explore how cost-effective different activities are. They were supported throughout by Data Ambassadors. Additional volunteers took part in a final DataDive weekend to produce the analyses you can see below. It was an opportunity for SJA to tackle several questions about their data that they were struggling to explore:

Understand engagement. What does good/high volunteer engagement look like? Who is more likely to be engaged, and can we tell why?
Map activity and volunteers. How representative are the volunteers of the communities they serve? Where are they, and where are they most active?
Explore cost-effectiveness. What is the cost of the volunteering model, and how does it relate to the effectiveness of the volunteers? How does the cost-effectiveness of the volunteer programme change for different volunteer segments?

They hope to use this information to invest in becoming more sustainable, and increase their impact by recruiting volunteers in ways that provide more engagement, retention, and diversity. They also want to use the outcomes from the DataDive project to inspire other teams in their organisation that a data-driven approach can truly make a difference.

Understanding St John Ambulance volunteers

Engagement across demographics and roles

To get to grips with volunteer engagement, the project team created measures based on the hours and the type of volunteering people did. DataDive weekend participants used clustering to see whether volunteers could be grouped together based on the type of activities they participated in, and whether they were recent or frequent volunteers. The main finding was that there are three or four distinct clusters, based on their engagement patterns.

Highly engaged volunteers were often skilled workers who took on multiple volunteering roles. Volunteers who mostly worked at events, in roles such as staffing a first aid station at a concert or sporting event, tended to be less engaged. Higher-engagement groups had significantly more involvement with roles like ambulance operations or supporting the NHS over winter crunches, and the most engaged group spent more time on administration.

By using variables including Length of Service, Age, or Training, St John had the beginnings of a much fuller portrait of different types of volunteers, and ways to challenge existing views they had about what might predict high engagement. They were particularly surprised to learn that people were less involved over time if they had originally joined the charity’s youth programmes.

These findings all help St John understand which roles have high turnover versus those that lead to long-term commitment, and why this might be the case, in order to encourage more volunteers to stay engaged for longer.

Who and where are St John Ambulance’s volunteers?

Alongside a deep exploration of volunteer engagement patterns, other DataDive volunteers looked at where St John volunteers are located within the UK, and how representative they are of the communities they serve. They created a Tableau dashboard to compare demographic categories across regions. Volunteers were generally very representative of the areas they were in.

Other breakdowns based on age, gender, and volunteering patterns provided clear trends and gave St John a selection of benchmarking outputs that they can work towards. They were able to spot disparities that need improvement, such as the lower percentage of women in leadership roles.

They also assessed retention and churn rates to spot which areas might need more staffing support in the future. As a result of this churn analysis, St John were particularly pleased to discover that over 75% of volunteers are still with them nearly three years later, and 65% were still volunteering after four years!

Comparing scenarios for cost and resources

Another useful dashboard from the weekend let the team hypothetically raise or lower costs against different spending options, showing how changing resources would impact other roles. It’s a great evidential tool for checking and supporting difficult conversations about costs. For example, events can be analysed for the types and severity of injuries that tend to happen, so that they can deploy volunteers with the right skills to them.

This tool is a boon in terms of allocating resources and assessing which trained volunteers, volunteering hours, and types of equipment are needed in any given situation. This could be a great boost in their overall effectiveness at events and other activities.

What’s next?

Nathan Palmer, St John Ambulance’s Head of Programmes, commented that the project has been “Turbo-charging our fluency with our own data. Working with DataKind really helped us to understand our own data and our own processes better. Then to have a priceless session with real experts analysing this and bringing exciting findings to life was so valuable to us. It really did exceed all our expectations.”

The St John team hopes to re-scope their volunteering programme and are really excited to be able to use lots of the tools and methods demonstrated to them. The DataDive project’s insight will reinforce their next steps. There are still parts of the organisation that are wary of data, so they plan to package the results for these different audiences. They have since received funding for a data audit, and conducted an in-house data assessment.

Finally, their Insight Lead, Kinga Salisbury, said that the project was a “mind-blowing experience!” We’ll take that!

Thank you so much to St John Ambulance’s charity representatives Kinga, Nathan, Lauren, and James for their time and incredible enthusiasm. Congratulations to the project’s Data Ambassadors Tahir, Irene, Daisy, and their troubleshooter Sukhil for completing this amazing project. And a final big thank you to all of the weekend’s volunteers for your spectacular work!

Engaging 15,000 volunteers with St John Ambulance