--:-- EST
OpenWrap SDK

Mobile Ad Revenue with a Self‑Serve Platform

How I created a zero-to-one, mobile-first platform that tripled what our support team could cover.

Role
Lead Designer
Year
2024 – ongoing
Type
B2B SaaS, shipped
+9 ptsMobile revenue share, year one
6 → 17Publishers per support rep
2.1×Account growth
Angry Birds 2
Android
•••
MonetizingGames
Ad units6Revenue, 7 days$20,321
Context

Web was the company. Mobile was the bet.

PubMatic helps app and website owners make money from ads. As third-party cookies went away browser by browser, audiences were already watching and playing in apps.

Problem

Mobile was a radio button.

The mobile tool was the web tool with mobile screens bolted on. Same one-page form. Same vocabulary built for websites.

The real ceiling

Support did the work for developers.

Every launch, price change, and network swap became a ticket, done by a rep logged in as the developer. Mobile could only grow as fast as we could hire support.

What I did

The pitch nobody asked for.

I helped the team pitch some vision work and create momentum for users that were quietly suffering.

Information architecture

From buried to its own product.

Card sortsTree tests

OpenWrap sat four levels deep in a web suite. Starting from a blank slate, we tested groupings and names until they made sense to developers.

Sitemap before: OpenWrap nested under Inventory, Header Bidding, alongside web tools
Before: buried under Inventory › Header Bidding
Shipped sitemap: Set up, Optimize, Troubleshoot, Reporting, Settings
Shipped: Set up, Optimize, Troubleshoot
Vocabulary

Their words, not ours.

Reusing the suite's terms would have felt consistent. It tested badly.

Competitor auditsSurveys
Profiles→Apps
Mediation→Optimize
Delivery→Troubleshoot

+ dozens more terms across the platform, reworked and aligned the same way.

User flow

Built for the return visit.

Developers didn't set up once and leave. They came back weekly to swap ad networks, adjust pricing, and run experiments.

  1. Sign upContract and kick-off
  2. IntegrateAdd app, ad units, bidders, SDK
  3. TroubleshootReal-time stats
  4. OptimizeFloors and breakdowns, every week ↻
The product

Everything starts with the app.

  1. AppsMost developers had 1 to 15 apps. Cards made them skimmable.
  2. Add appThe preview fills in as you go. No 40-field form.
  3. Ad unitsPick a format, not a form.
  4. BiddersMap partners and tune them weekly, without a ticket.
Design system

The new system's first real user.

A brand refresh was underway and a full platform redesign wasn't viable. I pitched the SDK as the long, low-stakes project the design system could build against. Both got an easier path to yes.

DecisionUse the project as the design system's proving ground.
New card component: app cards with status chips and an overflow menu
New drawer component: app properties and activity log
Outcome

Developers did their own work.

Same changes, done in the SDK in minutes. The support queue went quiet.

Competitive

The market leader said yes.

Our biggest competitor had declined to integrate with us. During the launch run-up, they agreed to a partnership. It wouldn't have launched successfully without this platform.

Read the announcement
↗Mobile got its own product, roadmap, and a metric leadership tracked
●Post-launch developer demos in Korea with our country manager
“It's game over. I can finally use you to improve our revenue.”
Prospective client · APAC market
What I'd measure next time

Design the product and the measurement together.

The project moved the number we committed to. What I can't fully answer is what drove the lift: developers doing the work themselves, or making smarter calls about which ad networks to use. I didn't have per-developer attribution wired in. That's the instinct I've carried into every project since.

Shipped work at PubMatic. Screens rebuilt with sample data.