vishu.design/ Work / Property Analytics
All work
B2B · Lead designer02 · Property Analytics

One view of performance: aligning Booking.com's partner teams around what matters most

The ranking dashboard at Booking.com is where property owners check their visibility: search and hotel page impressions, booking trends and reservations over time. It was basic, and as we discovered, it didn't meet partners' needs. I led the redesign, and brought the programme teams and the teams behind partner scores together around one vision for it.

CompanyBooking.com
My roleLead designer · cross-team direction
Team2 product teams, data science, research, UX writing, eng
Timeline3 months
Teams alignedGenius, Preferred, score owners
PlatformsWeb · Mobile
Hero · Final designs across web and mobileDesktop · Mobile
Quick glimpse of some of the final designsDesktop
At a glance · Impact
↓Support burden reduced: partner help-centre visits down within the first quarter.
↑Compset creation increased after launch, showing higher feature adoption.
↑Property page scores improved, reflecting better content quality and completeness.
→Premium sign-ups didn't materially lift, which informed the follow-up strategy.
At a glance · Scope & leadership
01Drove discovery, defined UX principles and success metrics with product leadership.
02Brought the Genius and Preferred programme teams and the teams that own partner scores behind one vision for the dashboard.
03Owned end-to-end design across web and mobile; partnered on instrumentation.
04Settled the order of metrics with score owners, led decisions on metric meaning, timelines and upsell trade-offs.
05Shaped the roadmap for later phases with my product manager.
01 · Problem & goals

Partners were confused by the very data meant to guide them

  • 01High support ticket volume: partners were confused by the data the dashboard surfaced, confirmed by local account managers.
  • 02Metrics weren't enough for partners to make strategic decisions about discounts and visibility.
  • 03An opportunity to cross-sell premium programme memberships and reduce support burden.
02 · Bringing teams together

Every team wanted a place on the same page

The dashboard wasn't just a design problem. It was the one place every partner checks, so a lot of teams wanted to be on it. The Genius and Preferred programme teams wanted their programmes built in. The teams that own the review, property page and reply scores wanted their scores seen and acted on.

Without a shared direction, it would have turned into a pile of widgets competing for attention. I worked across all of these teams to agree on one vision for the product, and kept partners at the centre of every call.

Programme teams
Genius
Preferred
Wanted: their programmes built into the dashboard, not bolted on
Partners · property owners
Hotels, apartments, homes
Wanted: a clear, reliable picture of performance they can act on
Shared visionOne view of performance3 principles · one order of importance · one roadmap
Core team
Product · data science · research · UX writing · eng
Wanted: reliable metrics and a plan we could ship
Score owners
Review score
Property page score
Reply score
Wanted: their scores seen first, so partners act on them
Who I brought together, and what each group wanted from the same pageFig. 01
03 · Product & UX principles

Three principles to settle disagreements

Before we got into solutions, I got the core team and the partner teams on board with three product and UX principles. When two teams wanted different things, these gave us a shared way to decide: which option is more reliable, transparent and consistent for partners?

ReliabilityPartners make key decisions with this, so the data has to be right.
TransparencyA clear narrative, so partners always understand what they're seeing.
ConsistencyMetrics represented the same way across partner-facing products.
04 · Research

What partners actually struggled with

Old dashboard annotated with research highlightsFig. 02
  • 01Partners mistook the popularity score for their position in search results.
  • 02Crucial data like conversion was missing; cancellations and content/review scores were expected too.
  • 0330 days of metrics wasn't enough. Partners needed to compare with the past quarter and year.
  • 04It was unclear how the competitive set was calculated, or why partners couldn't set it up themselves.
Problem statement

How might we give property owners a clear picture of their performance and actionable insights, while driving premium programme sign-ups?

Success metrics I defined
Graph interactionsScore improvementsCompsets createdProgramme sign-upsSupport ticketsHelp-centre visits
05 · Key decisions & trade-offs

Clarity over density, trust over upsell

Ordered metrics by impact, not by teamThe teams behind the reply and property page scores wanted those scores at the top, to get partners acting on them. I pushed for conversion, average daily rate and cancellations first: they have the biggest effect on a property's performance, and research showed they're what partners watch most. The scores follow right after, so partners start with what matters most to their business.
Reframed the popularity scoreAdded contextual copy and benchmarks so it wasn't misread as a ranking.
Extended timelines with compare modeCapped mobile at 6 months to keep charts readable and tap targets usable.
Progressive disclosureAdvanced metrics on demand, to reduce cognitive load.
Programmes in context, not as adsGenius and Preferred show up where they're relevant, like when performance dips, instead of as banners everywhere. Both programme teams got a clear place on the page, and partners kept their trust in the data.
06 · Explorations

Going wide before going deep

I explored reframing the popularity score as a dynamic metric, a "performance at a glance" funnel, extended timelines with compare mode, and upsell banners that only appear on performance dips.

Wireframe explorationsFig. 03
07 · Final designs

Every pain point, answered across web and mobile

Performance at a glance, with a clearer search results scoreFig. 04 · Desktop
Compare mode across bookings, search and property viewsFig. 05 · Desktop
Competitive set and contextual programme cardsFig. 06 · Desktop
Graph interaction prototypeFig. 07 · Prototype
Mobile, limited to 6 months to keep charts readable and tap targets usableFig. 08 · iPhone
08 · Impact & learnings

What changed after launch

↓Help-centre visitsWithin the first quarter
↑Compset creationPartners actively using the competition widget
↑Property page scoresBetter content quality and completeness
→Premium sign-upsNo material lift, which informed what came next
Next steps · roadmap for later phases, shaped with my PM
  • →Experiment with upsell timing linked to specific performance dips.
  • →Lightweight goal-setting and progress nudges for key metrics.
  • →Expand compare mode to cohort benchmarks by market and star class.
  • →Tighter copy and in-graph annotations to prevent misinterpretation.
What I learned

The hardest part wasn't the charts. It was getting teams who each owned a piece of the partner experience to agree on one order of importance. Agreeing on principles early is what made that possible.

Next case study · 03 / B2B · Sole design leadFrom pitch to proof: setting the direction for Booking.com's premium programmeBooking.com · Preferred programme
Preferred programme