Service Design · UX Research · Product Strategy · Costa Rica

Museo
JADE
Digital Redesign

The world's largest pre-Columbian jade collection had an online experience that drove visitors away. Broken ticket flows, an unusable mobile experience, no digital strategy. We changed that.

My Role
Product Designer
Type
Service Design
Institution
INS / Central Bank
Output
Strategy Deck + Blueprint
UX Research Service Blueprint User Segmentation Journey Mapping Mobile-First Proof of Concept
Sala Jade Entrada Tienda Numismática Sala Principal Educación Auditorio · Salida BUILDING MAP Mapa del Edificio ACTION Visit site Browse Buy tickets Visit museum Return FEELING 😐 🤔 😟 😊 😁 IMPROVEMENT Faster discovery Better content Native checkout Audio guides Membership SERVICE BLUEPRINT SOM SAM TAM PRODUCT DRIVERS Curiosity Education Convenience Entertainment MUSEO JADE Central Bank Museums · Costa Rica
Team
1 Strategist · 3 UX Designers · 1 PM · Client Lead
Core obstacle
Conflicting priorities → resolved with user data
Market
643 active users · 28.2% of museum segment (SOM)
Output
Product Strategy Deck · Service Blueprint · Roadmap
01 — The Problem

A World-Class Collection.
A Broken Digital Door.

Museo del Jade holds the world's largest collection of pre-Columbian American jade. 7,000 objects. Five floors. Award-winning architecture. And a website that lost visitors before they ever bought a ticket.

The project focuses on improving the digital experience for the Central Bank Museums of Costa Rica — addressing current issues with the online ticket purchasing process and enhancing user engagement by adopting modern digital trends seen in international museums.

"The museum's physical experience is world-class. Its digital experience was actively costing them visitors."
Ticket purchase redirected to boleteria.grupoins.com — a completely separate domain with different branding. Every purchase required users to context-switch twice, triggering abandonment.
80%+ of Costa Ricans browse via mobile. The museum's desktop-first website had image-heavy pages, slow load times, and touch targets too small to use reliably.
2x1 discounts (Sundays for nationals, Wednesdays for internationals), free entry for under-12, seniors, and indigenous peoples — none of this was visible before users reached checkout.
No membership program, no virtual tours, no Google Arts & Culture presence, no audio guide integration — all features international museums had made standard.
No analytics instrumentation: no conversion data, no drop-off data, no way to measure the cost of the current problems in numbers.
CURRENT TICKET FLOW — 7 STEPS TO ABANDONMENT View file →
1
Finds museum via Google Maps / TripAdvisor → lands on museodeljade.grupoins.com
2
Clicks "Comprar entradas" → redirected to boleteria.grupoins.com
⚠ TRUST BREAK
3
Different UI, branding, URL. User questions if this is legitimate.
⚠ ABANDON
4
Must re-select museum, date, ticket type — no context carried from previous site
5
Pricing appears for first time — no mention of 2x1 discounts or free days
⚠ DROP-OFF
6
Payment form — mobile-hostile, no autofill, small touch targets
⚠ MOBILE FAIL
7
Email confirmation — no path back to museum site, no follow-up, no relationship built
This flow existed because of a technical integration decision — not a UX decision.
CURRENT WEBSITE — museodeljade.grupoins.com Visit site →
museodeljade.grupoins.com
MJ
Museo del Jade
VISITAREXHIBICIONESCOLECCIONESACTIVIDADESTIENDA
Comprar entradas →
CONOCÉ LA MÁS GRANDE COLECCIÓN DE
Jade Precolombino
⚠ No pricing visible
⚠ "Comprar" → external redirect
⚠ No mobile optimization
⚠ No discount discovery
Current site at museodeljade.grupoins.com — ticket purchase immediately redirects to a separate domain, breaking user trust at the most critical conversion moment.
KEY CONTEXT

Museo del Jade is operated by INS (Instituto Nacional de Seguros), not the Central Bank directly. Unlike a purely public institution, INS has commercial operations — giving us a stronger business case for UX investment than a pure cultural mission argument would have provided.

02 — Challenge

What Made This
Genuinely Hard

01
The ticketing system was owned by a separate IT team
Boleteria.grupoins.com was maintained by INS IT — not the museum. Proposing native checkout required cross-departmental buy-in, technical feasibility analysis, and executive sponsorship. We had to argue that the UX cost of the redirect outweighed the technical cost of replacing it.
02
No baseline data — building a case without metrics
No conversion rate. No session recordings. No drop-off data. We couldn't say "we're converting at X%, targeting Y%." Every insight had to come from qualitative research, competitive benchmarking, and proxy metrics from the ticketing system's basic logs.
03
Serving radically different audiences on one website
Tourist pricing (₡8,400), national pricing (₡3,000), free entry for indigenous peoples, museum professionals, seniors, and under-12. Group bookings for schools. Researchers needing collection depth. These aren't variants of the same user — they're completely different products.
04
Heritage institution risk aversion
Stakeholders were proud of what they'd built physically — and rightfully so. Suggesting the digital experience was failing felt like institutional criticism. Framing the conversation around opportunity, not deficiency, was a facilitation challenge as much as a design one.
The hardest constraint

Critical operational documentation was only available in Spanish — including internal process guides, staff interview notes, and ticketing system specs. This created an asymmetry: we could see the surface clearly, but backstage operations required translation cycles that slowed service blueprint work. This directly influenced our decision to default to visual artifacts (blueprint, journey maps, behavioral matrices) over text-heavy documentation — visual tools are language-agnostic and easier to validate across language boundaries.

THE CENTRAL STAKEHOLDER DISAGREEMENT

The business lead wanted virtual tours as the headline deliverable — inspired by MOMA and the Louvre. The UX team's research showed broken ticketing was losing revenue every month. Different people. Same goal. Completely different models of what "improve digital" meant.

How we resolved it: We ran a structured prioritization session using the drop-off analysis of the current ticket flow alongside the virtual tour proposal, then asked: "Which generates more revenue in 12 months?" The data resolved what opinion couldn't. Virtual tours became Phase 2. Ticketing became Phase 1.
REAL PRICING — Discovered during research (NOT visible during purchase flow)
Nationals / Residents
Regular: ₡3,000
Students: ₡1,000
International visitors
Regular: ₡8,400
Students: ₡6,300
+ Free: Under 12 · Seniors (CR) · Indigenous peoples · ICOM members · 2×1 Sundays (nationals) · 2×1 Wednesdays (internationals)
⚠ None of this was shown before users reached the payment screen
03 — Research & Audience

Six Methods,
One Clear Picture

Through user segmentation research, the team identified behavioral dimensions, mindsets, and extreme user profiles. This approach fostered stakeholder alignment and gave every design decision an evidence base traceable to specific user needs.

RESEARCH METHODOLOGY — Tools used across all phases View file →
Behavior
Dimensions
Mindsets
Extreme
Users
Personas
User
Journey
Service
Blueprint
Applied sequentially — each method built on the previous, creating a progressively richer picture of who the users are and what the service needs to do for them

User Segmentation Matrix — 4 Personas

We considered designing for the "average" visitor. The behavioral dimension mapping killed this approach: the distance between a price-sensitive first-timer and a trust-oriented returning member is too wide to serve with a single interaction model.

BEHAVIORAL DIMENSION MAP View file →
The "First Time
Visitor"
Cautious · Price-driven
The "Returning
Visitors"
Trusted · Price-driven
The "Educational
Groups & Researchers"
Cautious · Measured by Price
The "Casual Explorers
(Mobile-First)"
Trusted · Measured by Price
DRIVEN BY PRICE
MEASURED BY PRICE
CAUTIOUS
TRUSTED
The Returning Visitors (highlighted in teal) represent the highest-value segment — trusted relationship + price awareness = ideal membership program target.
Persona 01
The First Time Visitor
Cautious, often international. Found via TripAdvisor or Google Maps. Needs enough preview to justify the visit, and a purchase flow that doesn't feel like a phishing redirect when it jumps domains.
Price transparencyTrust signalsMobile checkout
Persona 02
The Returning Visitor
National, values the experience, visits multiple times per year. Has no digital relationship with the museum — no membership, no personalization, no reason to engage between visits.
Membership program2×1 Sunday discoveryLoyalty
Persona 03
The Casual Explorer (Mobile-First)
Discovers the museum spontaneously — walking by, Instagram, Google Maps proximity. Makes same-day decisions. Has a 3-minute window to go from curiosity to ticket on mobile. Current flow loses them in 60 seconds.
Sub-3min mobile flowSame-day bookingFast load
Persona 04
Educational Groups & Researchers
Teachers, archaeologists, cultural journalists. Currently handled entirely via email — buzonmuseodeljade@grupoins.com — with no digital self-service, no group pricing visibility, no scheduling tool.
Group bookingAcademic resourcesCollection depth
What the segmentation revealed that personas alone wouldn't: Personas 01 and 03 share a need (mobile-first, fast checkout) but for entirely different reasons. Persona 01 is cautious — they need trust signals before committing. Persona 03 is spontaneous — they need speed above all. Designing one checkout for both required a flow that built trust fast and moved fast simultaneously. That tension shaped the entire proposed checkout architecture.
04 — Service Blueprint

Mapping Every Layer
of the Experience

The service blueprint was the most valuable deliverable of the engagement — not because it was the most visible, but because it revealed operational constraints that would have broken well-designed frontend features.

MUSEOS DEL BANCO CENTRAL — Service Blueprint View full blueprint →
PHASE
ORIENTACIÓN
USO
RETENCIÓN
FIDELIZACIÓN
😊 VISITOR EMOTION
😕
"demasiado sencillo"
😊
"aprendiendo mucho"
😟
"aburrido"
❤️
"amo el museo"
🙁
"aquí no vuelvo"
FRONTSTAGE (visible)
Website, RRSS, Google listing
⚠ Tickets → external redirect
Staff, signage, exhibits, audioguías
⚠ No digital guides integrated
No follow-up communication
⚠ No membership offer
No loyalty program
⚠ Zero digital relationship
No re-engagement
BACKSTAGE (hidden)
SEO (minimal), social content, INS IT manages boleteria system
Staff scheduling, exhibit management, curator workflows
Group requests via email only (buzonmuseodeljade@grupoins.com)
No mailing list, no CRM, no membership infrastructure
No analytics, no feedback loop
CRITICAL PATH
The blueprint highlighted "Generación de tiquete de entrada" as the moment where the most user abandonment occurs — and where backstage decisions (IT team owns the external system) had the highest frontstage cost (trust loss, drop-off).
Built on the Museos del Banco Central service blueprint. The blueprint revealed that several proposed features — like real-time availability or self-service group booking — assumed backstage capabilities that didn't exist, allowing us to phase the roadmap against operational reality rather than design aspiration.
FRONTSTAGE FINDING
The redirect is a frontstage scar from a backstage decision
The external ticketing redirect wasn't designed as UX — it was a technical integration choice made by IT. The blueprint made this visible, reframing the conversation from "fix the UX" to "change the underlying architecture."
BACKSTAGE FINDING
Real-time features require infrastructure that doesn't exist yet
We almost designed a real-time availability calendar. The blueprint showed the ticketing system couldn't expose inventory data to the museum website. That feature was removed from Phase 1 before any wireframes were drawn.
RETENTION GAP
Post-visit is a complete void — no digital follow-up exists
After a visitor leaves, there is zero digital touchpoint. No email, no membership offer, no content follow-up. The blueprint made this structural gap visible as a deliberate product opportunity, not just a nice-to-have.
05 — Key Design Decisions

The Hard Calls
and Why

These were the decisions that shaped the final direction — where we disagreed with initial assumptions, stakeholder requests, or conventional approaches. This is where the thinking happened.

DECISION 01 — Most Impactful
Native ticketing over improving the external flow
The initial brief asked us to "improve the ticket purchase experience." The team initially interpreted this as: make the redirect smoother. After research, we reframed the problem: the external redirect isn't fixable by improving its UX — it's a structural trust problem. When users land on boleteria.grupoins.com after being on museodeljade.grupoins.com, they see a different interface, different branding, and a different URL. For first-time visitors — particularly international tourists — this pattern triggers abandonment regardless of how well-designed the target page is. It reads as a phishing redirect.
The uncomfortable truth we had to deliver: Fixing the current flow would require less work but achieve less result than building native checkout. We documented exactly why the UX cost of the redirect outweighed the technical cost of replacement — and this became the central argument of the Product Strategy Deck.
DECISION 02 — Most Contested
Service Blueprint before any wireframes
We insisted on completing the service blueprint before designing any screens. Stakeholders wanted visual output quickly — the blueprint felt like overhead. But it revealed something critical: several proposed features assumed backstage capabilities the museum didn't have. A real-time availability calendar requires live inventory data. A virtual tour booking system requires content moderation workflows. A group booking self-service requires capacity to respond within 24 hours. The blueprint showed us which features were design problems and which were operations problems dressed as design problems.
What we almost designed that would have been wrong: A virtual tour booking feature with a staff review step. The blueprint showed 2 digital staff managing all requests. That feature would have created a 5-day response bottleneck we'd have had to redesign post-launch.
DECISION 03 — Most Surprising
Price transparency over content depth
Our initial hypothesis was that users dropped off because the content wasn't engaging enough. So the first prototype emphasized richer exhibit previews, 360-degree images, and deeper storytelling. Testing invalidated this. Users who saw richer content and then hit the pricing wall — prices only visible at the payment step — dropped off at the same rate. The engagement problem wasn't content. It was information asymmetry. Users felt asked to commit before being told what they were committing to.
What changed: We moved pricing, discount eligibility, and free day information to the discovery stage — the moment a user first considers visiting. This reduced the psychological commitment required at payment because users had already mentally accepted the price earlier in the journey.
DECISION 04 — Scope Discipline
We deliberately didn't redesign the physical experience
The service blueprint surfaced significant in-museum opportunities — wayfinding, staff communication, group management, exit experience. Some team members wanted to scope these in. We said no. The engagement's value came from depth in digital, not breadth across physical and digital. A comprehensive service redesign would require different team composition, different timelines, and different client implementation capacity. Doing half of it well was less valuable than completing the digital piece fully.
How we handled it: All physical experience opportunities were documented in a "Phase 3" section of the Strategy Deck — acknowledged, researched, handed off — without inflating the current scope.
WHAT WE CONSIDERED AND REJECTED
Designing for a single "average" museum visitor
Early in the project, the team debated optimizing for the most common visitor type and deprioritizing edge cases. The behavioral dimension mapping ended this debate. There is no majority segment. The tourist, the returning national, the spontaneous explorer, and the educational group each represent meaningful audience shares — and serve different business goals (revenue, loyalty, volume, institutional credibility). A design that optimized for one would have actively excluded the others.
06 — Proposed Solution

A Coherent Strategy,
Not Just a Redesign

We propose simplifying the ticket purchase process, improving mobile usability, and introducing virtual tour capability — all scoped against the backstage capacity the service blueprint revealed.

"The goal is not to build a better website. It's to build a digital system that matches the ambition of what the museum does physically."
01
Native Checkout
End-to-end purchase within museodeljade.grupoins.com. No external redirect, no trust loss. Pricing and discounts visible at discovery. Completable on mobile in under 3 minutes.
02
Mobile-First Architecture
Rebuild for touch-first usage. Performance budget: under 3 seconds to interactive on 4G. Every critical CTA above the fold on 375px viewport. Image optimization throughout.
03
Transparent Pricing at Discovery
Pricing, 2×1 eligibility, free days (May 18, Sept 27), and audience-specific rates surfaced at the awareness stage — before users even consider purchasing.
04
Virtual Tours + Audio Guides
Integrate existing audioguide content (already at /audioguias) with virtual tour capability. Uses existing museum assets — no new content production required in Phase 1.
05
Membership & Loyalty
Digital membership that rewards returning visitors. Benefits: priority booking, exclusive content, early notification of Mañanas Culturales and conference cycles.
06
Blueprint-Aligned Phasing
Each feature phased against its backstage requirement. No feature launches before the operational capacity to support it exists. No beautiful solutions to problems the institution can't yet sustain.
07 — Product Drivers

Six Values,
One Decision Filter

These six attributes — validated with users and stakeholders — became the filter for every feature in the roadmap. If a proposed feature didn't serve at least one driver meaningfully, it was deprioritized.

🔍
Curiosity
Content that invites exploration before commitment — making the museum irresistible before the user has paid.
⚡
Convenience
Zero friction from intent to ticket. Mobile-optimized, under 3 minutes, no redirects, no surprises.
📚
Education
Rich collection content and audio guides that serve researchers and school groups before and after the physical visit.
🎭
Entertainment
Virtual tours, interactive previews, shareable moments — making the museum feel current and worth returning to.
🏛️
Cultural Connection
Design that reflects Costa Rican cultural identity. The largest jade collection in the world should feel like that matters digitally.
✨
Inspiration
Visual storytelling that moves people. The architecture communicates "jade divided in half" — the digital experience should carry that intentionality.
HOW DRIVERS RESOLVED THE VIRTUAL TOURS VS. TICKETING DEBATE

We ran both proposals through the driver matrix. Native checkout scored high on Convenience and Cultural Connection (removing the trust-destroying redirect). Virtual tours scored high on Education and Inspiration — but required Convenience to work first. Users need a functional purchase flow before they'd invest in virtual content. The driver matrix made the sequencing decision data-driven rather than opinion-driven.

08 — Obstacles & Resolutions

What We Had to
Navigate

Obstacle
Conflicting stakeholder priorities
The business lead wanted virtual tours. The UX team's research showed ticketing was losing revenue every month. Two people. Same goal. Completely different interpretations of what "improve digital" meant.
Resolution
Structured prioritization session. Both proposals evaluated against the same criteria: which generates more measurable value in 12 months? The drop-off analysis won the argument. Virtual tours became Phase 2.
Obstacle
No baseline analytics to measure from
No conversion rate. No drop-off data. No session recordings. We had to argue for UX investment without being able to say "we're converting at X%."
Resolution
Supplemented with competitor benchmarking, heuristic severity scoring of each friction point, and proxy metrics from the ticketing system's basic logs. Not perfect — sufficient to make the case.
Obstacle
Operational docs only in Spanish
Critical process guides, staff interviews, and ticketing system specs were only available in Spanish — creating asymmetry in what the full team could analyze without translation overhead.
Resolution
Bilingual team members led document review. We defaulted to visual artifacts (blueprint, journey maps) over text-heavy documentation — visual tools are language-agnostic and easier to validate across language boundaries.
Obstacle
Proving ROI to a public heritage institution
INS runs Museo JADE as a cultural mission. Arguing for UX investment in terms of ticket conversion felt like reducing a heritage institution to an e-commerce problem.
Resolution
Built a dual business case: revenue (online ticket conversion) and mission (digital access for users who can't visit physically — researchers, educators, diaspora Costa Ricans). Both cases needed to land.
09 — Impact & Learnings

What This
Changed

As a Proof of Concept engagement, the measurable impact is in strategic clarity — not shipped code. The Product Strategy Deck gave the museum a research-backed roadmap and a prioritized direction for the first time in its digital history.

4
Validated personas with documented needs, journey maps, and feature implications
6
Product drivers validated with stakeholders — used as a live decision filter
1st
Service blueprint ever produced for Museo JADE — mapping all touchpoints across digital and physical

What I take forward

01
Map the service before designing the screen
The service blueprint revealed backstage constraints that would have broken well-designed frontend features. We almost designed a real-time availability feature that assumed inventory data the ticketing system couldn't expose. Now I treat service mapping as a prerequisite, not a parallel deliverable.
02
User data wins internal debates faster than anything
When stakeholders disagreed on feature priority, user data resolved it in 20 minutes. Arguments that had circled for days ended when we showed the drop-off analysis. Building the habit of making user data the deciding vote changed the entire dynamic of the engagement.
03
Heritage institutions need a different ROI language
Framing design value purely in revenue terms doesn't resonate with cultural institutions. Learning to articulate impact in terms of mission — access, education, cultural reach — alongside revenue made our recommendations land with the full stakeholder group, not just the commercial side.
04
Scope discipline is a design skill
Keeping the engagement focused on digital — when physical experience opportunities were clearly visible — required design judgment about where intervention would have the most leverage. A focused, well-evidenced digital strategy was more valuable than a sprawling engagement that exceeded implementation capacity.
PRODUCT STRATEGY DECK — Full research presentation Open in Google Slides ↗
Scroll through the full strategy deck — includes all research synthesis, the complete service blueprint, persona documentation, market analysis, and proposed product roadmap. Note: some slides are in Spanish.
Explore the Full Research
Product Strategy Deck · Personas · Service Blueprint · Journey Maps · Market Analysis
View Strategy Deck View Figma File ← Back to Portfolio