Rethinking funding | Early Tomorrow

How might we help assessors prioritise billions in funding?

Services

Interaction design · UX research · Prototyping in code · Cross-functional collaboration (research, content design, engineering) · Coaching a junior designer

Description

The Department for Levelling Up, Housing and Communities (DLUHC) runs funds that allocate billions of pounds to regenerate local areas, including the Community Ownership Fund. I was brought in as an interaction designer to co-design an MVP tool with assessors and funding teams, helping them allocate work, prioritise applications and keep oversight of progress. I worked closely with research, content design and engineering to design, iterate and prototype the MVP, including prototyping in code, and I coached a junior designer along the way.

Impact

Designed an MVP assessment tool for the Community Ownership Fund
One simple tool for three user types with no permissions or role types needed
Prioritised with engineers, cutting what the data could not support
Prototyped in code and coached a junior designer.

Designed an MVP assessment tool for the Community Ownership Fund, helping teams allocate work and keep oversight of billions of pounds in funding
One simple tool for three user types, giving lead assessors, general assessors and policy teams what each needed without permissions or role types
Cut what the data could not support by testing with assessors and prioritising with engineers, so the MVP launched with features that were reliable
Prototyped in code and coached a junior designer, and the tool landed well, with one assessor saying they were “crying with joy because this is easier”

Designed an MVP assessment tool for the Community Ownership Fund, helping teams allocate work and keep oversight of billions of pounds in funding
One simple tool for three user types, giving lead assessors, general assessors and policy teams what each needed without permissions or role types
Cut what the data could not support by testing with assessors and prioritising with engineers, so the MVP launched with features that were reliable
Prototyped in code and coached a junior designer, and the tool landed well, with one assessor saying they were “crying with joy because this is easier”

Year

2022

2021

Timeline

6+ months

Location

London, UK

PROBLEM SPACE

The problems this work had to solve

This was a high profile central government service with large amounts of money at stake, a lot of unknowns, and an MVP that could not rely on permissions or unique role types.

Three Types of Users, One Simple Tool

Lead assessors, general assessors and policy teams all needed different things. Lead assessors needed enough information to prioritise work and keep oversight, general assessors needed to find work, and policy teams needed oversight. The MVP had no permissions or unique role types, so it had to be a low-complexity tool that worked for everyone.

Sparse Knowledge of the Process

We started with very little knowledge of how work was allocated and overseen. That meant forming hypotheses first, then running experiments and asking questions to learn what assessors actually needed, from stats and constituency to asset type and ways of showing progress.

Ambition Versus What the Data Could Support

Some of the most useful ideas did not survive contact with the data model. Constituency was too risky as free text, and score differences were not supported at all, so we prioritised with the tech team before testing again and landed on what could realistically ship in an MVP.

CONTRIBUTION

How I created value

I worked across research, design and code, testing with policy designers and assessors and then prioritising with engineering so the MVP was both useful and buildable.

Testing & Iterating Designs With Real Assessors

Researched the designs with policy designers, assessors and lead assessors. This showed that knowing how many applications were left against a deadline was helpful, funding requested was a good signal of complexity and profile, asset type was a useful way to understand complexity, and search was needed to find projects.

Prioritising With Engineering

Spoke with the tech team to prioritise before testing again. Constituency was dropped as free text with a risk of incorrect data, score differences were not supported by the data model, and asset type was available, which shaped the MVP and its handover.

Coaching and Prototyping in Code

Prototyped in code to get closer to what would be built, and coached a junior designer through the work. This is a different context from my earlier project: central government, high profile, and large amounts of money.