Background

As a Category 1 responder under the Civil Contingencies Act, the Environment Agency has a legal obligation to prepare for emergencies including warning and informing others of the location, timing and magnitude of flooding from rivers and the sea in England.

The changing needs of users, the back-end service contract expiration and response to extreme weather events

There is an existing flood warning service (XWS) but it only caters for single users, needing warnings for a single location.

The challenge

A targeted flood warning system (TFWS) allows multiple users to sign up for flood warnings for multiple locations. There is an existing service but there are few registrations and a low take-up. Organisations are usually invited to use the service and are are hand-held through the registration process.

We initiated a discovery to understand how users interact with the service and establish the end-to-end service from the user perspective. We would answer how much of the low take up was due to lack of visibility, what the issues with the current service are and identify areas of improvement.

The team

The team consisted of a Service Owner, Delivery Manager, User Researcher and Business Analyst from both the Environment agency and DEFRA.

I joined the team as a Senior Interaction Designer during the discovery phase of the service. My role was to work with the researcher to create research outputs from discovery and prepare the ground for the alpha phase.

Discovery phase

The discovery phase sought to answer the following questions:

• who are the current and likely users and what they’re trying to do?
• what current services or channels do organisations use to stay across the threat of flood?
• what problems or frustrations do organisations experience and how are they currently supported?
• what do users need from a targeted flood warning service to achieve their goals?

I worked with the user researcher to clearly communicate the outputs from discovery to aid insightful decisions.

The current service

Targeted flood warning system (TFWS) as it was known was a digital service in use by just over 100 businesses. It was set up to inform organisations with multiple land and property assets when a risk is likely to occur so that they can proactively manage and mitigate any impact as soon as possible.

Since the service was created by an external agency, there was little documentation. I created a map of the user journey architecture of the current service.

I created a user journey map to:

• demonstrate the hand holding nature of the registration process
• show how many end users and internal staff are involved in the service
• highlight the pain points for users in the service
• demonstrate the user experience in each stage
• show how the service aligns with the current flood warning service

I created a user story map to:

• help encourage a collaborative way of working
• show how TFWS unites and diverges with the current flood warning service
• provide an way of discussion opportunities to the challenges discovered
• show how we might define a possible minimum viable service.

Since this was an as-is map, this exercise would be repeated in early alpha with the assistance of the whole team.

Data analysis

I conducted analysis of the service data to understand more about the user needs. For instance, utility companies have 1000s of locations at risk of flooding with potentially hundreds of users needing access to the warning system notifications. There are likely to be a handful of administrators for each organisation which would require servicing.

The main difference we identified between the targeted flood warning service and the current single user flood warning service was that users need to add multiple locations - up to the 1000s. This prompted us to question the viability of an entirely separate service for organisations rather than a branch (or feature) of the current service.

Adding multiple locations

I created a map showing the possible solutions for adding locations based on technical capability and the complexity of the data (number of locations, users and notification settings). This highlighted the needs for a multi-faceted approach to file upload.

I created a very simple prototype for version 1 which was deliberately light on information in order to tease out the user needs from the participants.

Finally, I created a prototype which for participants could use in a research setting. A research plan has been delivered and testing date is dependent on the results of a reoganisation of the flood teams.