Workable HR: Zero to One
Designed, shipped, and scaled from 2022 to now.
I was the founding designer of Workable's HR Suite. Over three years I took it from a research sprint to a multi-module HRIS at scale, then grew into leading the HR Apps & Design System team.
- Product
- Workable HR: a multi-module HRIS, built alongside Workable's established ATS
- Timeframe
- 2022 – Now
- Role
- Founding HR designer → Lead, HR Apps & Design System
- Team
- Solo → leading a 4-person HR Apps & Design System team
01Role & releases
From first designer to platform lead
The journey in Workable HR, from first designer to a multi-module HRIS at scale. Highlights, releases, and impact.
02Constraints
The tricky bits
Merge with ATS, simplify setups, evolve the design system, lead a bigger team.
Two products, one experience. Shared patterns, real limitations.
ApproachReused where viable, forked where necessary; wrote cross-product rules.
Dense configurations, many edge cases.
ApproachProgressive disclosure + presets; task-first flows, not page-first forms.
Parallel applications, quick scaling, rising coordination cost.
ApproachWeekly crits, async RFCs, definition-of-done; player-coach execution.
03The decision
Reuse, fork, or write the rules
The hardest call: how two products share one system without flattening each other.
One design system. Two products with different jobs. Every shared component was a question of fit.
Reuse everything
✕ Breaks fitFast and consistent, but forces HR into pipeline shapes.
Fork everything
✕ Breaks scalePerfect fit, but two systems, double maintenance.
Rule-based
✓ ChosenReuse where jobs match, fork where they differ.
Tradeoff accepted
Slower first mile.
- Design-system governance written before screens.
- Cross-product RFCs added review time upfront.
- Short-term inconsistency during migration, accepted on purpose.
Result
One system scaled to two products: HR flows kept their shape.
- Shared primitives held across HR and ATS.
- HR-specific surfaces diverged only where the job required it.
- Cross-product consistency stayed legible, not accidental.
04Deep dives
Four problems, four systems
Each surface started from a real workflow breakdown. The problem, the design response, and the numbers that followed.
Hassle-free transition from candidate to employee
My role: Owned end-to-end: research, IA, flows, and shipped UI.
- Problem
- Scattered tasks. Manual docs.
- Design
- Single portal. Templates. Progress.
- Discover
- Research exposed drop-offs at documents and provisioning.
- Develop
- One portal with templates, assignees, and progress.
+100%
WAU of the onboarding portal in the past year
4.71
avg. custom workflows per account
Employee profiles that maintain themselves
My role: Owned end-to-end: data-model UX, approvals, audit trail, org chart.
- Problem
- Data lives in many places, high admin load, inconsistent fields.
- Design
- Self-service edits with approvals, full audit trail, org chart & document vault.
53%
of paying customers use HRIS (ATS + HRIS), as of Aug 2025
12+ mo
HRIS accounts have outnumbered ATS-only among active accounts since mid-2024
Clear time, clean payroll
My role: Set direction and the core patterns; reviewed and shaped delivery with the team.
- Problem
- Fragmented tracking. Complex policies. Manual approvals.
- Design
- Schedules & policies. Simple clock-in/out. Clear approvals. Payroll-ready exports.
77.6%
of time entries are clock-based (vs 22.4% manual)
98.7%
of time entries go unedited after submission
Reviews people actually finish
My role: Player-coach: set direction, designed the key flows, delegated the rest.
- Problem
- Low completion. Inconsistent templates. Scattered feedback.
- Design
- Template builder & cycles. Lightweight manager/self reviews.
2.2×
growth in paying accounts running a real review cycle: 24.6% in the last 12 months, up from 11.3% in Jan 2025
05The spine
One system, two products
The design system was the project's backbone, the reason HR could move fast without drifting from the ATS.
Shared primitives
Type, color, spacing, and base components stayed common across HR and ATS: one visual language, one maintenance surface.
Intentional forks
HR-specific surfaces (configuration, approvals, lifecycle) diverged only against a written cross-product rule, so every fork stayed legible.
Patterns over screens
Task-first flow patterns and presets, not one-off layouts, so each new module assembled from known parts.
Governed, not frozen
RFCs and a definition-of-done kept the system evolving with the product instead of calcifying or sprawling.
06Operating model
The design operating cadence
Lightweight rituals that cut rework, boosted consistency, and let multiple modules ship in parallel.
- 01HR squad
In-team syncs
Ad hoc · multiple/week
Design reviewFlowsDirection - 02Cross-product · Foundations
Design collab
2×/week
Design reviewFlowsDirectionHRIS/ATS consistency - 03Org level
Lead collab
Weekly
RitualsProductHiring - 04Design org
Design Hive
Monthly
Design reviewDirectionAlignment
07Reflection
What scaled, what didn't
What scaled
- Rule-based reuse: the cross-product ruleset outlived every individual screen.
- Task-first flows over page-first forms; presets absorbed the configuration density.
- The operating cadence (weekly crits, async RFCs, definition-of-done) let modules ship in parallel.
What didn't
- Coordination cost rose faster than headcount as parallel delivery scaled.
- Some configuration surfaces stayed denser than I wanted, even with progressive disclosure.
- The cross-product ruleset was written reactively; it should have existed before the second module, not after.
Next time
- Invest in the shared ruleset and presets on day one, not at the first conflict.
- Make coordination a designed system, not a meeting that grows with the team.
- Treat the design system as the product's spine from zero, not a layer added at scale.
Next case study
Building the first design assistant