Making benefit fraud reporting work with incomplete information.

DWP wanted more people to report suspected benefit fraud online, reducing reliance on paper submissions and the call centre while improving the information received by investigation teams.

The existing service made that difficult.

People reporting fraud often had useful information, but they rarely knew every detail about the person they were reporting. They might know a nickname, an approximate address, where someone worked or a pattern of behaviour, without knowing a full legal name or date of birth.

The form treated those gaps as errors. Mandatory questions prevented people from continuing, even when the information they could provide might still support an investigation.

The challenge was to make reporting quicker and easier without reducing the quality of information collected.

Working closely with user researchers, we built a picture of how the existing service performed using several sources of evidence.

We reviewed previous online submissions, including complete reports, partially completed forms and reports containing very little usable information.

We spent time with the call-centre team to understand the questions people asked, the information they struggled to provide and the operational pressure created by handling reports manually.

We also ran usability sessions with people who could imagine someone they might report, including participants who had previously used the service.

Four clear needs emerged:

• People wanted to complete the report quickly, often using a mobile phone.
• They needed to submit a report without knowing every personal detail.
• They wanted confidence that the process was anonymous.
• They needed reassurance that the information had been successfully handed over.

The research showed that the service had been structured around the information the organisation wanted, rather than the information people could realistically provide.

Instead of forcing everyone through one long, fixed form, we redesigned the journey around what the person knew.

Common types of benefit fraud were introduced earlier, allowing the service to ask questions that were more relevant to the selected situation.

We reviewed which questions genuinely needed an answer and removed unnecessary blockers where people might not know exact details.

The redesigned journey allowed people to:

• Provide a name, nickname or other identifying information.
• Continue when exact details such as a date of birth were unknown.
• Explain what they had observed in their own words.
• See questions relevant to the type of report being made.
• Understand that they could report anonymously.
• Review their answers before submitting.

Clearer GOV.UK patterns and content also made the service feel more trustworthy and familiar, which was particularly important for a sensitive and anonymous reporting journey.

The goal was not simply to make the form shorter.

The service needed to balance two competing needs: reduce friction for the person reporting while collecting enough information for investigators to understand and act on the report.

We used evidence from previous submissions and call-centre conversations to identify the information that was most useful, then shaped questions around the way people naturally described what they knew.

This helped the service capture partial but potentially valuable information without presenting every missing detail as a failure.

I worked closely with the wider squad throughout the process, sharing user journeys, prototypes and research findings so that design decisions were understood across design, research, delivery and development.