Role
Business Context
BookMyShow saw roughly 50 million monthly page views, but not every session converted — even regular bookers browsed without transacting sometimes. I led design of an AdTech platform to monetize that non-transacting attention, built around a distinction between a "Booker" who transacts regularly and a "Browser" who rarely does. The hardest constraint was protecting the booking experience the business depends on while building a genuinely new revenue line on top of it. We split ad inventory into internal "Promoted" content and third-party "Ads," built custom formats like a transparent hero banner for blockbuster releases and reusable "crossover" ads for multi-product brands, and published creative guidelines so the platform could scale without a designer reviewing every asset.
The obvious move was to add ad inventory wherever there was open screen space. But BookMyShow's two user types needed different treatment: a Booker — someone who transacts regularly — has zero tolerance for anything that slows down booking, while a Browser — someone who hasn't transacted in over a year — has attention that's genuinely available to monetize. Treating both the same would have meant either under-monetizing Browsers or degrading the experience for Bookers, the users the business depends on most. The real problem was building a system that could tell the difference and behave accordingly.
That gap between total traffic and completed transactions represented monetization the business wasn't capturing. The mandate was to build an AdTech platform that could turn non-transacting attention into revenue, without compromising the booking experience that drives the business's core transactions.
Strategic Decision
Protect the Booker, Monetize the Browser
Context: ad revenue and booking conversion could work against each other if ads were one undifferentiated inventory type.
Options: (1) one ad strategy uniformly across all users, or (2) split strategy by intent — protect the booking flow for Bookers while maximizing ad exposure for Browsers.
Trade-off: differentiation cost more to design and build (two clusters, guiding principles for each), but a uniform approach risked leaving revenue on the table or damaging the transaction flow.
Decision: split ad offerings into two clusters — Promoted (internal, sellable) and Ad (third-party, sponsored) — governed by rules that prioritized never interrupting a Booker's flow.
Why: never get in the way of a Booker, never compete with internal messaging like offers and rewards, and maximize revenue from Browsers specifically. Revenue growth couldn't erode the experience that made the platform valuable.
Design Principles
1. Never Block a Booker — a user actively booking should never have an ad interrupt that flow.
2. Internal Messaging Comes First — offers and rewards from BookMyShow take priority over third-party ad inventory.
3. Monetize Attention, Not Screen Space — ad placement followed where non-transacting attention already existed, not empty space.
4. Always Have a Fallback — if the ad platform failed to serve, the space defaulted to internal content, never a blank placeholder.
Key Product Decisions
1. Hero banner ads
Problem: production houses wanted high-impact placements for major releases, but a standard banner wouldn't carry the buzz of a blockbuster or feel native.
Options explored: a standard rectangular ad unit vs. a custom, transparent-background unit built around the movie's lead character and a famous line of dialogue.
Decision: build the custom hero format.
Solution: a transparent-background ad breaking widget monotony, leading to a hero image, popular videos, and a primary "Interested?" CTA feeding directly into ticket booking.
Business impact: converted ad engagement directly into ticket sales rather than treating advertising and booking as separate funnels.
2. Crossover ads
Problem: third-party brands selling multiple offerings (a credit card's tiers, a coffee chain's seasonal menu) needed a format that could showcase a range, not one product.
Options explored: a single static ad image per brand vs. reusing existing UI components (the same horizontally-scrolling card pattern used for movies) to showcase multiple products.
Decision: reuse existing UI components for multi-product ad units.
Solution: crossover ads let brands like Starbucks and Amex show a scrollable set of products within a format users already knew how to browse.
Business impact: created a sellable ad format for brands with multiple offerings, widening the pool of advertisers.
3. Creative guidelines for scalable ad quality
Problem: as the ad platform grew, creative needed to stay consistent across advertisers without a designer reviewing every asset.
Options explored: review each ad creative manually vs. publish guidelines specifying image ratios, resolution, safe zones, and pass/fail examples.
Decision: publish creative guidelines.
Solution: a Creative Graphic Guidelines document covering anatomy, technical specs, and side-by-side examples of acceptable and unacceptable creative.
Business impact: let the platform scale to more advertisers without a proportional increase in manual review.
Outcomes
Business: created a new, dedicated revenue stream from ~50M monthly page views' worth of previously unmonetized, non-transacting attention.
Customer: Bookers kept an uninterrupted booking flow, while ads shown to Browsers were tied to content and brands relevant to what they were already looking at.
Operations: fallback logic meant ad-serving failures degraded to internal content instead of a blank or broken experience.
Engineering: creative guidelines reduced the need for manual, ad-hoc review as the number of advertisers grew.
Support: no support-specific data was available for this feature.
Reflection
The biggest surprise was how much the Booker/Browser split simplified every downstream decision. Once we had that distinction, questions like "should this ad show here" stopped being subjective and became a matter of checking which user we were looking at. I'd also build the creative guidelines earlier next time, alongside the first ad formats rather than after — we ended up writing standards for a system that had already shipped, retrofitting consistency instead of designing for it from the start.
The principle I still use: revenue features should be designed around protecting the behavior that makes the product valuable in the first place, not layered on top of it. Ads that compete with your core transaction are a tax on the business, not a new revenue line.
