AvailableFrom January 2027 for product, service, design systems and leadership contract roles.Get in touch
Luke James Taylor Logo

— bpFleet. A design system for bp's biggest B2B portfolio.

  • Fleet management, moving businesses towards carbon neutrality.
  • A portfolio with revenue of over £11 billion a year.
  • From one person to a team of 11.
Client
bp

The short version

A 30-second read
  • The problem

    bp's largest B2B portfolio needed to deliver at pace, but the central design system wasn't ready to support it.

  • What I did

    I led a sister design system to bpCore, grew it from just me to a team of 11, and rebuilt the component library.

  • What changed

    9 teams and 38 journeys built on it, over 40 components, and a plan to bring the two systems together.

  • 9teams
  • 38journeys
  • 40+components
Read the full story

01/ Summary

What was delivered and how
  • — Services
    • Design system leadership
    • Hiring and team building
    • Design system strategy
    • Heuristic evaluation
    • Governance and contribution
    • Evangelism
  • — Deliverables
    • A team charter, vision and roadmap
    • Design tokens, a design kit and a rebuilt component library
    • Documentation and onboarding for designers and developers
    • A contribution model and request process
    • Service metrics for the design system
    • A plan to align with bpCore
  • — Outcomes
    • 9 teams building on the system
    • 38 journeys designed and built with it
    • Over 40 components in a rebuilt, tested component library
    • A team of up to 11, run like a product team

02/ At a glance

Design System Lead for bpFleet, a sister system to bpCore
  • 9teams building on it
  • 38journeys designed and built
  • 40+components
  • 11people, at its largest

03/ Understanding the problem

The portfolio needed pace. The central system couldn't give it yet.
  • Untested components

    Nobody could be sure a component would behave the same way twice.

  • An outdated tech stack

    Bloated libraries that slowed teams down rather than speeding them up.

  • Inconsistent design assets

    Design files and code that didn't agree with each other.

  • No contribution model

    Teams couldn't add what they needed, so they built it themselves.

  • An unclear roadmap

    Nobody could plan around what the central system would do next.

  • Uneven journeys

    A heuristic review of 9 fleet portal journeys, scored by 6 reviewers, showed how much they varied.

Why fleet needed its own system

  • Deliver components at pace, with autonomy
  • Keep teams unblocked, with no dependencies
  • Full control over complex fleet patterns
  • bpCore solves for every bp application, so it can't be tailored to fleet
  • bpCore's own team would soon be overwhelmed by fleet's requests

Built on bpCore's foundations

Not starting again: the fleet system used bpCore's colours, grids, layouts, typefaces, spacing and type hierarchy, and built what fleet needed on top. Consistency without uniformity.

Plug the gap with a sister system, always aligned with bpCore. Not a rival: the end goal was always to bring the two together.

04/ Building the team

Hiring for each stage the system reached
  1. Just me. Could bpCore, bp's central design system, serve the whole fleet portfolio, and SuperFleet in particular?
  2. Three designers. Onboarding product designers to the pattern library, creating components and supporting new journeys.
  3. Three software engineers. Including a lead, to help teams use the component library, then rebuild it on a modern stack.
  4. A Product Owner and a Scrum Master. To run the system like a product, with a backlog and a cadence.
  5. A Design System Evangelist. Who took adoption to the next level. A team of up to 11.

05/ Running it like a product

A charter, a roadmap, and the rituals to deliver it

A team charter

Who we are, why we exist, how we work, how we're funded, how we measure ourselves and how we communicate. Funding was split across fleet projects in proportion to the support each needed. Kanban, with a daily stand-up, weekly refinement, retros every 2 weeks and a drop-in design clinic every Friday.

A roadmap in 8 streams

Each with a 2022 goal. For example: teams consuming basic tokens, then dark mode and accessibility tokens; a design kit that always mirrors the code; and any developer able to contribute with ease.

13 areas of work

Team charterVisionRoadmapBacklogGovernanceService developmentContributionDesign languageDesign kitComponent libraryDocumentationEvangelismAlignment with bpCore

06/ What we built

Every gap, answered
  1. Quality

    On a modern stack, chosen with the developers who'd use it.

    FromUntested components
    ToA rebuilt, tested component library
  2. Consistency

    Typography, colour, spacing and sizing stored as code.

    FromDesign files and code that disagreed
    ToTokens, a design kit and code always in sync
  3. Contribution

    And hotfixes for contributions that fixed an immediate problem.

    FromNo way to contribute
    ToContributions at any fidelity, from idea to live
  4. Direction

    Open to everyone it affected.

    FromAn unclear roadmap
    ToA roadmap in 8 streams, each with a 2022 goal
  5. Adoption

    Plus a drop-in design clinic every Friday.

    FromTeams left to work it out
    ToOnboarding journeys for designers and developers

The design system shouldn't build the product. The product builds the design system.

07/ Never a bottleneck

Alone we can go far. Together, we can go further, faster.

Decentralised by design

The go-to design system team for fleet, serving every team across the portfolio, starting with SuperFleet and total cost of ownership. Decentralised, so we were never a bottleneck: teams built the bespoke UI they needed, and we made sure it stayed consistent, reusable and scalable.

Inherit, then give back

The fleet system inherited from bpCore, and contributed back to it, so the work for fleet improved the experience across bp. It supported SuperFleet, total cost of ownership and in-truck payments, with more fleet products lined up next.

The value for bp

  • Faster time to market

    Reusing elements cuts redundancy, so new and improved components ship faster.

  • Capacity

    No duplicated effort: teams across the portfolio design and build each component once.

  • Onboarding

    New colleagues and teams get up to speed quicker, and deliver value sooner.

08/ Measuring the service

Six measures of whether it was working
  • Contributions

    The share of components contributed by other teams, and how many arrive each month.

  • Adoption

    The share of fleet initiatives using the system, with ladders of adoption.

  • Retention

    Teams that stop using the system in favour of something else.

  • Quality

    Bugs reported by the people using the system, and how fast they're fixed.

  • Satisfaction

    Surveys and interviews with designers and developers, including onboarding.

  • Cycle time

    Split into design and engineering, and reviewed every 8 weeks.

Putting a number on reuse

The team's formula for what reuse saves: hours saved a week, times 52 weeks, times an hourly rate, times the number of people. Try it.

2.5 hrs
$70
20

Saved a year

$182,000

2.5 hrs × 52 weeks × $70 × 20

09/ Outcomes

From plugging a gap to bringing two systems together

What it delivered

  • 9 teams

    Building on the system.

  • 38 journeys

    Designed and built with it.

  • Over 40 components

    In a rebuilt, tested component library.

  • A team of up to 11

    Run like a product team.

  • Part of something bigger

    By June 2022, design ran across 22+ fleet initiatives in 5 sub-portfolios, with the design system as one of its core capabilities.

  • Then alignment

    Taking the best of both systems, with a transition plan towards bpCore.

Available from January 2027

Looking for a design leader?

I take on contract roles as a design consultant, head of design or fractional design lead, across product, service and design systems. Remote anywhere, or hybrid across the UK.