Product design case study · giffgaff internal tools
Rebuilding the agent platform
Replacing the tool giffgaff's support agents use to manage a member's account, with a brief set by what agents told us and what they actually spent their time on.
Overview
The problem
Agents were helping members over chat using an ageing tool that logged them out, loaded slowly and buried what mattered under clutter. Every delay kept a member waiting on the other end.
The brief
Replace it with a platform that removes friction from agents' day, built from what agents need rather than everything the old tool did. Everything added had to add value, not noise.
Starting with the people who use it all day
We surveyed 27 agents about the old tool, visited the contact centre to watch them work, and counted how often each task actually happened.
| Frustration | In their words |
|---|---|
| Repeated logins, short timeouts | "Timing out too quickly disrupts my flow" |
| Slow loading and updates | "It takes ages for updates to appear, sometimes you're not sure if it's gone through" |
| Cluttered comments | "There's too much rubbish in comments that makes it hard to find what matters" |
| Dense pages, heavy pagination | "It's just too cluttered sometimes, I miss key info" |
| No history of what happened | "There needs to be a history log, so we don't waste time redoing things" |
Source: survey of 27 agents, in their own words
Counting the work
What agents actually spend their time on
| Agent task | Times a year |
|---|---|
| Looking up the identifier of a dispatched phone | 72,000 |
| Handling delivery problems, changes and timescales | 13,685 |
| Requesting a replacement device, at the time done by hand | 4,008 |
| Extending a return window | 360 |
| Issuing a VAT receipt | 279 |
The task at the top was being handled through a shared spreadsheet, and no agent had ever raised it. It had stopped feeling like a problem. It only surfaced because we counted.
Not one of them asked for more features
Almost every complaint was about friction, clutter or waiting. That set the bar for the new platform: everything on screen has to earn its place, and the most frequent tasks come first.
Deciding what the new platform would and wouldn't do
Scope was ranked by agent time consumed, not by how loudly something was asked for. That put the hidden, high-volume tasks first, and it changed how we answered the most requested feature.
A hard call
Meeting the need without bringing back the clutter
What they asked for
Bring the comments section across, redesigned, with visible history and a clearer hierarchy.
What the research said
The contact centre visit put clutter and non-essential information at the top of the list. Comments were a large part of that clutter.
The decision
We built a dedicated member history area, kept away from the core dashboard. Agents get the full history when they need it, without it crowding the view they work from every day.
A platform built around how agents work
Agents do two kinds of work: helping one member, and running the service. The new platform is organised around that split, from the navigation down to each page.
Structure
Find a member, or run the service
The old tool used a row of tabs along the top that had already run out of room. The new platform uses a side navigation, which can grow downwards and group related pages.
It's split in two: everything about a member, and the operational tools agents use to run the service.
Member view
One member, everything at a glance
Agents start with a person on the other end of a chat, not with a dashboard. So the member view leads with who they are, what they pay and what they have.
Focus
Space and status for the task in hand
Inside a member's account the main navigation collapses to icons, handing the space back to the work.
Status sits on the surface rather than behind tabs. Triaging a fault shouldn't mean clicking through five of them to find out whether anything is wrong.
How it was built
The giffgaff design system, made ready for code
The same system members see
The platform uses giffgaff's consumer design system rather than an off-the-shelf admin kit. That was challenged, fairly, because it added time. Agents settled it: the familiarity helps when they describe a member's screen while looking at their own.
Designed straight into code
Because the design system was machine-readable, screens could be generated from real components and tokens, then reviewed against the design intent. A navigation concept was running on a test environment within two hours.
How agents rate the new platform
9.4/10
Easy to use day to day
9.4/10
Easy to find things in the menus
9.1/10
The navigation, rated before agents had used it
One caution on the scores: daily users rate their tools generously. "How easy is this?" produced nines. "Which task takes you longest, and what makes it slow?" produced the roadmap.
Set against what agents originally raised
| What agents raised | Where it landed |
|---|---|
| Repeated logins, short timeouts | Single sign-on removed the manual logins entirely |
| Slow loading and updates | Rebuilt on a faster platform; one agent in ten still picks it as a hassle |
| Cluttered comments | Moved into a dedicated member history area, away from the core dashboard |
| Dense pages, heavy pagination | Simpler pages, organised around the member and the service |
| No history of what happened | A dedicated member history area; engineer install history is still flagged as unclear |
The biggest irritation now is error messages that don't explain what's wrong, chosen by half the agents. That's a content problem presenting as an interface one, so it went to the teams who own those messages.
What I learned
- 1
Count the work before designing it
The biggest hidden cost had never been reported, because it had stopped feeling like a problem.
- 2
Expertise hides friction
Ask which task takes longest, not whether the tool is easy.
- 3
Meet the need, not the literal request
Agents asked for comments back. What they needed was history they could find, without the clutter. A dedicated history area gave them both.