Skip to content

Case study · Senior IC

Untangling Municipal Complexity: Self-Service Tariff Authoring, Rate Simulation & Geospatial Zone Management in PM360

Role
Lead Senior Product Designer (Individual Contributor)
Timeline
2021 – August 2025
Partners
Product Manager, front-end & back-end engineers (Core UI / React migration), QA/SDET, Account Management
Scope
PM360 / Policy & Operator Tools, North American municipal and university operators

The problem

Moving rate setup out of spreadsheets and support tickets, where one mistake reaches drivers

Configuring policies across hundreds of zones (recurring rates, event rates, max durations, rate rollovers, multi-segment sessions) relied on spreadsheets (ZAR files), legacy Angular screens and internal tickets.

A misconfigured tariff could mean drivers overcharged or a city losing revenue, and operators only found out after it was live.

Before: spreadsheet & ticket loop

  1. Operator drafts rates in a spreadsheet (ZAR file)
  2. Emails the file or opens a support ticket
  3. Internal team re-keys it in legacy Angular screens
  4. Operator waits, then finds out how rates behave in live sessions

A mistake sends the operator back to step 1.

After: self-service PM360

  1. Operator authors the policy and rates in PM360
  2. Rate Tester simulates real sessions before publishing
  3. Guardrails block conflicting rules and flag drafts
  4. Operator publishes with confidence, no ticket needed

Discovery

Operators weren’t afraid of complex forms. They were afraid of publishing blind.

I ran contextual inquiries, built an ARO discovery journey and operator empathy maps across major city and university accounts, and usability-tested the zone filter.

Operators needed to know which rate would “win out” in a live driver session before they pressed publish.

The Rate Tester

Giving operators confidence to publish by letting them test real sessions first

I designed the React Rate Tester into the Create/Edit Rate funnel and Policy Detail views. It separates live rates from “(Draft)” and “(Draft in progress)” ones, wraps long policy names, and shows Free Parking and No Parking fallbacks.

Try it: change the date and time, include the draft rate, and turn Rate Rollover on and off to see how the price changes.

Start time

Policy rates in this test

  • Downtown Weekday Metered: $2.00/h, 8:00 AM–6:00 PM
  • Stadium Event Rate (Draft in progress): $8.00/h, 5:00 PM–11:00 PM
  • Street Sweeping: No Parking, 8:00 AM–10:00 AM

Rate in effect at start

Downtown Weekday Metered

Session price breakdown
FromRate appliedMinutesCost
5:30 PMDowntown Weekday Metered180$6.00
Estimated total$6.00
  • Rate Rollover is off: the whole session is priced at the rate in effect when it started.

Illustrative rates and rules built for this portfolio, not a live tariff.

Guardrails

Preventing conflicting rules from ever reaching a live city

Working closely with engineers, I designed guardrails for rules that can’t coexist and for staged rollouts.

Mutually exclusive toggles

Rate Rollover vs. conflicting options

Multi-Segment Sessions can’t be used with Session Grouping, Expiration Times or Metered Prepayment. Try turning them on.

Feature-flagged rollout

A banner operators can’t dismiss

During city-wide Rate Rollover migrations, a persistent system banner explains the temporary state. It is behind a feature flag, so each city moves when ready.

Rate Rollover is being enabled for this city. Some rate settings are limited until the migration finishes.

UI vs. API validation

One message, whichever layer catches it

Edge cases the UI can’t see, like overlapping time windows, come back from the API. I mapped them to field-level messages with the same wording and tone as UI checks.

End time overlaps another rate

“Weekday Metered” already covers 4:00 PM–6:00 PM on Fridays. Adjust the window or its priority.

I also laid the geospatial foundation: connecting abstract zone IDs to real map polygons, so self-service mapping could follow.

Outcomes

From ticket bottleneck to confident operator self-service

Self-service

Rate setup moved from internal ticket queues to operators.

One component system

Legacy Angular views unified into reusable Core UI React components.

A foundation for Pulse

The domain architecture directly informed Arrive Pulse.

Product screens

The real screens behind the demos