← All work

Research & journey design

Redesigning the online donation form

A calmer, clearer way to give, including in memory of someone. Online completion rose from 38.2% to 45.6%.

Client
British Heart Foundation
Role
UX Designer
Duration
Aug 2024 – Nov 2024
Team
UX Lead, 2 UX Designers, UI Designer, Business Analyst (Nagarro)
Tools
Figma, Miro, Google Analytics, Mouseflow
Deliverables
Research, user journeys, form components, prototypes, user testing

In short

The BHF's online donation form was losing six in ten people who started it. I helped redesign it end to end: research, a new four-step journey, the Gift Aid and donation-summary components, and two rounds of user testing. After launch, completion rose from 38.2% to 45.6%.

About

Website Transformation stage 3 looked at every form on the BHF site, and the donation form came first. It is where supporters give money for research, including people donating in memory of someone they have lost. The work was part of a wider Online Donations × CRO programme.

Scope

The one-off and monthly online donation journey, including the in-memory route and Gift Aid. The redesign also produced reusable form components (stepper, donation summary, Gift Aid, Keep in Touch, address finder), which were handed over to the UI team and the design system. Item collection and house clearance forms reused the Gift Aid work but were outside this journey.

The problem

People were dropping out

Only 38.2% of the people who entered the funnel finished. Analytics and Mouseflow showed where they left, field by field.

The page structure worked against them

The heuristic evaluation kept coming back to page structure: no sense of how long the form was, where you were in it, or what came next.

Gift Aid was hard to understand

The declaration was legal text that people skimmed or skipped, so donors missed a free 25% boost.

What we wanted to achieve

  • Higher completion across the donation journey.
  • Fewer validation errors.
  • Shorter time to complete.
  • New agreed baselines to measure against.

My role

Research
Analytics and Mouseflow review, heuristic evaluation, competitor review of donation summaries.
Journey design
The new four-step flow and the in-memory questions.
Components
I owned Gift Aid and the Donation summary, and worked on the stepper with Antonio Dell'Omo, who added it to the design library at handover.
Testing
The stepper sessions in Sept 2024, the Gift Aid design critique, and the final testing and affinity mapping of the redesigned form before handover.

The approach: four phases

  1. 01 · Discover

    Find the drop-off

    Analytics, Mouseflow, heuristic evaluation, CRO insights.

  2. 02 · Define

    Agree the target

    Focus areas, KPIs, a traffic-light status for each form element.

  3. 03 · Design & test

    Build and compare

    Components, journeys, stepper prototypes, design critique.

  4. 04 · Validate

    Test the whole form

    Moderated testing of the full redesigned form, then affinity mapping.

DiscoverFinding the drop-off

We started with the numbers. Google Analytics showed where people entered and left each journey. Mouseflow showed what they did in each field. Then the UX team ran a heuristic evaluation, rating every issue by severity. The most severe issues clustered around page structure. A CRO insights board added motivations, impacts and recommendations.

The donation journey before the redesign, mapped screen by screen during discovery.

Design & testThe new journey: four steps

Your donation → Your details → Payment → Thank you. Best practice says 3–5 steps. Each step says how far along you are and what comes next ("2 of 4 · Next: Payment"), and Back appears from step two onwards.

The new four-step journey on mobile. The radial stepper shows where you are and names the next step; Gift Aid sits with the amount, and the summary returns before payment.

Design & testThe stepper, tested

Two navigation designs went into testing: a horizontal stepper and a radial "wheel". We ran four moderated sessions on 12–13 Sept 2024, at the BHF Camden shop, at head office and remotely. Participants made a £20 in-memory donation, added Gift Aid, then went back to change their details. The radial design shows one step at a time with the next step named beside it, and that was the version we took forward.

The two stepper designs we tested: horizontal and radial.

Design & testGift Aid, in practice

Gift Aid adds 25% to a donation at no cost to the donor, but only if they understand it and tick the box. The copy had to come from Fundraising Compliance and be signed off by Legal, so the design had to do the explaining:

  • A plain headline, "Boost your donation by 25%", with the declaration hidden until the box is ticked.
  • The Gift Aid logo kept, because people recognise it and it draws the eye.
  • A worked example: your £20 is worth £25.

We tested placement, background, logo and helper link in an internal design critique, and made separate versions for money and for donated goods.

Two of the Gift Aid variants from the design critique: with or without the logo, and with or without the £50 → £62.50 banner.

Design & testDonation summary, in practice

Three of the four charities we reviewed had no donation summary. The new form shows the amount as a banner under the stepper and again before payment, with Edit links back to each step, so donors always know what they're giving.

The Payment step: the donation summary first, then your details with an Edit link back to the step where you entered them.

ValidateWhat testing told us

Before handover to development, we tested the full redesigned form with five participants and grouped their feedback into six themes: Usability & flow, Tone & sensitivity, Gift Aid, Transparency & trust, Preferences & flexibility, and Clarity of copy.

Affinity map from testing the redesigned form: 5 participants, 6 themes, 4 key trends.

Four key trends

Warmth works, if it's optional. People liked the compassionate tone, but some hesitated when asked to share a personal story.

→ explain why we ask, and let people skip

Motivation questions need framing. The "how long ago" question felt intrusive this early in the journey.

→ mark optional, or move later

Gift Aid still needed simpler words. One participant skipped it every time because she didn't understand it.

→ bite-size copy and a tooltip

Labels need to say what they do. Some participants missed "Edit", and "details" and "donate" were overused.

→ clearer labels, less repetition

What testing decided

Soften the story request

Make it clearly optional and explain what Heart Stories are.

Simplify Gift Aid

Bite-size copy, a "How it works" tooltip and a visible £20 → £25.

Clearer labels

"Change amount" and "Review details" instead of "Edit"; motivation questions marked optional.

Key numbers

45.6%completion, up from 38.2%.
+19%relative increase in completion.
~5,300more completed donations in six months, compared with the old rate.
4.5 / 5overall rating from a test participant.

Outcomes & reflection

The redesigned form went live in 2025. Comparing Nov 2024 – Mar 2025 with Dec 2025 – May 2026, the share of people who completed a donation rose from 38.2% to 45.6%, and abandonment fell from 61.7% to 54.4%. With 71,746 people entering the funnel in the second period, that's roughly 5,300 more completed donations than the old rate would have produced.

These are before-and-after figures, not an A/B test. The two periods cover different months, and other CRO work ran alongside, so I treat the rise as the redesign's direction of travel rather than its exact effect.

If I did it again, I'd test the in-memory questions with bereaved supporters earlier. They turned out to be the most sensitive part of the form, and the hardest to get right from inside the team.

“It feels human, like you are talking to me.”

Test participant