Research and discovery for a subscription service

Findmypast is a technology-driven genealogy subscription service. As lead designer, I led a multi-phase discovery project, to understand what community features customers wanted.

Design lead • Product design • User research • User testing • UX/UI

The brief

Find out what community features Findmypast customers want to use, so we can increase their engagement.

Approach

We had some evidence that our audience wanted to connect and communicate with each other (a growing FMP Facebook group). But we didn’t know what features they’d want to use on Findmypast. So, I led a discovery project with the aim of finding this out, and then ran a design sprint, so we could create and test new community features.

I led two concurrent research phases to find out what FMP customers wanted from their community.
Using research evidence, I setup and led a design sprint, to generate a testable prototype, for a new community feature.

Research

Understand what makes successful communities

Understand what our customers want from an FMP community.
Generate insights that will tell us where to focus our efforts (what features to develop) and then test new feature ideas with customers.

Research phases

1: Interviews
We ran a series of interviews to find out why people join family history communities, what do they think of the FMP Facebook group, and what does a good community look like.

2: Survey
We then conducted a customer-wide survey, to find out what is the actual demand for community features on FMP, and what needs do they serve.

Design sprint

Once we had clear evidence about what FMP customers wanted from a community experience, I could start planning a design sprint, so we could turn our research insights into testable community features.

I recruited subject matter experts and key stakeholders to be part of the design sprint, to maximise our ability to create valuable feature concepts.

We had a concept to test.
Member Directory.

Our research and design sprint led to us mapping out an experience we could then test with customers in a prototype, as a simple proof of concept.

Every FMP customer with a tree would have a searchable profile.

FMP users would be able to find researchers who could help them with their research.

We would build profiles directly from member’s tree data. We’d then use that data, along with user-inputted data, to suggest relevant matches for users.

FMP customers would be able to find other researchers with matching research interests, and connect with them to get help with blockers to their family history research.

Feedback was positive

They thought this was a potentially useful tool. They saw the value in connecting with other people, through common research areas, interests and skills. They thought it could help break down research brick walls.

The concept was clear and the prototype easy to use. They compared us to Ancestry and preferred our concept of connecting people and the way we executed the private messaging aspect.

Other things we learned

They like the concept, but may need to be convinced of the value, before adopting it. A lot of FMP users keep their main trees on Ancestry (about 30%). They’d need to see clear value before uploading their trees with us. Also, more experienced researchers may try resolving research problems on their own.

They want to vet other people’s trees and experience before trusting them for help. They expect to be able to see other people’s trees. They’d look for similar or higher experience.

Outcomes

We successfully met all of our project goals. We produced statistically significant evidence that our customers wanted community features, in a very targeted way, and also tested a new community feature, that demonstrated it could generate engagement.

We had positive results from our prototype testing. It showed us that customers would engage if they could find help with their research road-blocks, and also would be open to helping other researchers.
This was the first time Findmypast had run a discovery project, and the project was seen as a successful example of how to generate actionable value from long-form research.