F
Feasibility
Engineering & the Hopper partnership
Can we build it on the consumer platform we already run?
Fails when ignored: A business feature set that forks the consumer product.
Capital One Travel · Small business portal · 2023
Scoping business travel for the person arranging it.
Capital One was building a business travel portal for people booking their own trips. I showed the team that in small businesses, someone else usually does the booking, and reframed the roadmap around them.
01 · The strategic uncertainty
A consumer portal, a new segment, and a user the product frame had not named.
Capital One launched its consumer travel portal in 2021 and decided to expand into small business. The early frame centered on employees booking for themselves.
In many small businesses, someone else coordinates the trip while also handling HR, accounting or operations. That work was invisible to the roadmap, and the team had no evidence either way about how common it was.
50%
of surveyed small businesses had someone booking travel for other employees.
The existing product model centered the individual traveler and policy-management features. It did not yet account for the people arranging travel for others. This survey finding was separate from the 12 qualitative interviews I conducted to understand that work.
02 · The system I was designing inside
Four claims on the same roadmap, each of them reasonable.
Leadership wanted differentiation for the business vertical. Product wanted activation in the open beta. Engineering had a consumer platform and a Hopper partnership to build on. Nobody was arguing badly. The question was which of these got to decide when they disagreed.
A
Arranger
Owners, managers, HR and admin roles
Can someone booking for other people finish the job quickly?
Fails when ignored: A portal that assumes the booker is the traveler.
B
Business
Product & leadership
Does it move activation and adoption in the open beta?
Fails when ignored: Controls nobody asked for, built instead of the basics.
T
Traveler
Employees on the trip
Does the employee get a trip that actually works for them?
Fails when ignored: Policy enforcement that overrides the person travelling.
03 · The pivotal move
A workshop on what we did not know about arrangers unlocked the research.
15 cross-functional participants. Capital One and Hopper. No concept pitches.
We made assumptions about travel arrangers explicit, grouped the open questions by journey stage, and prioritized what the roadmap actually needed to resolve. The session gave the research a clear job and gave the team a reason to wait for the answer.
Research question
Who takes responsibility for arranging travel, and what other work surrounds that task?
Why it matters
The product frame assumed the traveler and booker were the same person. Understanding the role meant seeing the HR, accounting, and administrative work around it.
How we would learnSmall business interviews12 businesses
Research question
Walk us through the last trip you booked for someone else. Where did traveler context come from?
Why it matters
A recent example could reveal how preferences were gathered, remembered, and carried into the booking.
How we would learnContextual interviewsArrangers across roles
Research question
How are traveler preferences, budget, and business constraints balanced when comparing options?
Why it matters
The tradeoffs arrangers made would determine whether the product should emphasize policy enforcement or reusable traveler context.
How we would learnJourney mappingArranger journeys across roles
Research question
What happens after checkout, and what information carries into payment and reconciliation?
Why it matters
The booking experience does not end at purchase when the same person may also be responsible for accounting and operations.
How we would learnWorkflow interviewsOwners, admin and finance roles
“The workshop was not there to pick a direction. It was there to make the team agree that we did not yet know who we were building for.”
04 · Building empathy with small businesses
Twelve small businesses revealed how arrangers balanced employee needs with business constraints.
Business goals were connected to research objectives so the study could shape the roadmap. Jobs-to-Be-Done mapping then showed why arrangers chose their current solutions and where the product could reduce effort.
01
Arrangers prioritized employee preferences and kept a rough budget in mind, without necessarily adhering to a strict policy.
Across all 12 small business interviewsArrangers enable people, they do not police them
The role exists to take the burden off a travelling employee, not to enforce compliance. Trust in employees was high.
ArrangerTraveler
So what
Build profile and preference support, not a policy console.
02
Arranging travel overlapped with HR, accounting, administrative, and operational responsibilities.
Synthesis across arranger rolesArranging sits inside a wider operational role
The experience needed to support the surrounding work rather than assume a dedicated travel manager.
ArrangerBusiness
So what
Make traveler context reusable so booking for someone else does not require rebuilding it each time.
05 · The reframe
Arrangers were enabling people, not policing them.
Arrangers book travel to ease the burden on travelling employees because that responsibility overlaps with HR and accounting, not because the business distrusts its people. That distinction changed what was worth building.
Previous design target
A centralized travel manager enforcing policy
- Businesses assumed to focus on cost minimization and strict expense control
- Bookings assumed to run through a dedicated travel department
- Goals framed around compliance, negotiated deals and streamlined expenses
Updated finding from research
An arranger balancing employee needs against business constraints
- Cost matters, but getting the employee there for the work matters more
- Arranging is spread across owners, managers, approvers, and HR and admin roles
- Goals framed around balanced options, employee trust, and expenses tracked in one place
The evidence made the shift from policy management to profile support legible as a product decision rather than a design preference.
06 · From jobs to opportunities
We organized the work around what arrangers were trying to accomplish.
Jobs-to-Be-Done synthesis compared near term improvements with longer term possibilities. Teal marks what the open beta could carry; everything else remained as direction.
Journey stage
Get employees onto the portal
ActivationFind options that work for the traveler
SelectionMake checkout and reconciliation easier
CompletionRoadmap horizon
Prioritized for open betaDirection beyond beta
Value unlocked
Activation · adoptionAdoption · experience qualityEfficiency · experience quality
Near term
Import employee information
Capture preferences in profiles
Reuse information already collected
Longer term
Connect directly with HR systems
Rank and filter using traveler preferences
Automate employee and expense context
Workflow breakdown
The arranger cannot book for someone who is not enrolled.
Traveler context otherwise lives in the arranger's memory or a spreadsheet.
Travel work spills into accounting when context must be rebuilt after checkout.
Opportunity
Reduce the setup work required before the first booking.
Reuse preferences so each booking becomes faster and more trusted.
Carry employee and expense context through the end of the trip.
Evidence + decision
Supporting research
Booking on behalf of another employee was common across the small business research sample.
Synthesized research observationRoadmap decision
Made enrollment and employee context a first-class part of the open beta rather than an edge case.
Supporting research
Arrangers balanced employee preferences with a rough budget rather than enforcing a strict policy.
Synthesized research observationRoadmap decision
Shifted the near-term roadmap from policy management toward profile management.
Supporting research
Arranging frequently overlapped with accounting and operations responsibilities.
Synthesized research observationRoadmap decision
Connected immediate efficiency improvements to a longer-term experience-quality opportunity.
Scroll horizontally to follow the full journey on smaller screens.
The map connected immediate roadmap choices to a longer product direction without losing the underlying jobs.
07 · The moments that mattered
A second workshop turned the two near term jobs into touchpoints.
Existing capabilities, direct competitors, and comparative solutions helped break the near term opportunities into the moments that carried the most weight.
Getting employees onto the portal
The arranger cannot book for anyone who is not there yet, so enrollment is the first real barrier.
PrioritizedHolding preferences somewhere reusable
Seat, airline and hotel context lives in the arranger's head or a spreadsheet until a profile can hold it.
PrioritizedFinding an option that suits the traveler
Balancing an employee's needs against a rough budget, which is the judgment the arranger is actually making.
Longer termClosing out the trip
Checkout and reconciliation, where travel work spills into the accounting side of the same person's job.
Longer term08 · Impact
OPEN BETA · FIRST HALF 2024The research changed both the product and the way success was measured.
Stage 2 launched to 200,000 businesses with the roadmap shifted toward profile management. This describes the launch audience, not verified adoption. A senior product lead added qualitative KPIs so the team could track whether the experience was becoming genuinely useful, not only whether people completed transactions.
Measured12small business interviewsOwners, managers, and administrative and HR roles
Measured15workshop participants alignedCapital One and Hopper, across Product, Design and Engineering
Outcome200,000businesses included in the Stage 2 launchStage 2 shifted the roadmap from policy management to profile management
Recognition1 of 3teams given the inaugural Summit AwardAcross the whole Capital One Experience Design organization
Roadmap
Stage 2 launched to 200,000 businesses with the roadmap shifted from policy management to profile management.
Measurement
A senior product lead redefined the KPIs to include qualitative measures, not only transactions.
Recognition
The Summit Award recognized connecting business and customer needs and challenging business as usual.
Practice
The unknowns workshop became the way this team opened an ambiguous problem.
A user the roadmap had missed became the basis for moving a launch stage onto the work that mattered to the people doing the booking.