Skip to main content

We use necessary cookies that allow our site to work. We also set optional cookies that help us improve our website. For more information about the types of cookies we use, visit our Cookies policy.

Cookie settings

Spark Beyond is an Artificial Intelligence (A.I.) powered problem-solving platform that can generate insights from large, complex datasets.

We are collaborating with these leaders in the field of data science and A.I. to leverage the latest state-of-the-art technology and expertise to gain insights from our data.

We are incredibly grateful to the team at Spark Beyond who generously provided their A.I. expertise on a pro bono basis.

The problem:

We asked SparkBeyond to help us understand the factors that might predict when a Shout 85258 volunteer was losing confidence and might subsequently stop volunteering.

Qualitative interviews that our coaching team held with volunteers had already revealed several reasons why volunteers stopped volunteering:

  • For example, a change in life circumstances often made it hard for someone to continue volunteering.
  • In addition, volunteering with Shout can be challenging because of the nature of some of the conversations with texters - it was therefore not surprising to learn that many volunteers chose to take a break for this reason.
  • One further factor that emerged was that maintaining confidence was important for volunteers, especially in the early stages of their time as a volunteer.

"Using Spark Beyond's A.I.-powered platform, we were able to examine a very large number of possibilities that would have been unfeasible using more traditional analytics approaches."

Mark Ungless

Director of Data Insights, Mental Health Innovations

Our solution

We hypothesised that maintaining a regular cadence of volunteering activity would be important for maintaining confidence and continued volunteering. If we could understand what level of activity was important to maintain confidence, then we could better tailor our coaching activities to support the volunteers.

Working with SparkBeyond's data scientists we set about trying to unpick this puzzle. Using their A.I.-powered platform, we were able to examine a very large number of possibilities that would have been unfeasible using more traditional analytics approaches.

We found a number of fascinating factors that appeared to predict volunteering activity - taking 3 or more conversations per month appeared to be particularly important. This makes intuitive sense - it is easy to imagine that retaining confidence when doing something challenging like volunteering for Shout may well require weekly practice.

Crucially, though, SparkBeyond's approach allowed us to identify a particular activity level and the magnitude of its importance that we would otherwise have been unaware of. Having gained this insight, our coaching team now works with volunteers to maintain this level of activity as far as possible, so as to help volunteers build and maintain confidence in their abilities.

You might also be interested in