How Accor Hotels India turned every rupee of search spend into thirty.
A search programme rebuilt around revenue, not bookings — capturing high-intent travel demand and routing budget toward the room nights that actually paid the brand back.
Cost-effective bookings are useful. Profitable ones are the point.
Accor came in asking for cost-per-booking targets. We came back with a different scoreboard — one that asked which bookings were worth chasing in the first place.
Accor Hotels India wanted to drive cost-effective hotel bookings, with two proposed KPIs: number of room-night bookings and cost per booking. Both reasonable. Both incomplete.
A booking at a city business hotel and a booking at a beach resort don't earn the brand the same revenue — but a CPL-style metric pretends they do. Optimizing toward a flat cost-per-booking floor would have funnelled money into the cheapest conversions, not the most profitable ones.
If you only count bookings, you reward the cheapest one. If you count revenue, you reward the right one.
Bookings & cost per booking.
- Number of Room Night Bookings
- Cost per Booking
- Treats every conversion as equal
- Risks bias toward cheap, low-revenue stays
Revenue & Return on Investment.
- Return on Investment per booking
- Revenue generated per campaign
- Lets us double down on hotels with higher revenue per booking
- Ties performance marketing back to business goals
Capture the intent in the category. Then rank the intent by what it pays.
SEM, set up to do two things at once — make Accor top-of-mind for travellers shopping for stays, and route spend toward the queries with the highest revenue per booking.
Show up where the buyer is already shopping.
We activated SEM with a single focus — capture the intent that already exists in the category. Brand queries protected the asset; generic queries — "hotels in Kerala", "5 star hotels Goa" — captured the buyer who was open to a new option.
- Brand defense on "Accor Hotels" and chain variants
- Generic capture on destination + category queries
- Ad copy tuned to revenue tier, not just clickability
- Landing pages aligned to the search query, not the homepage
Two trees, one budget.
The keyword set ran on two parallel trees — brand, where the cost is low and the revenue is loyal; and generic, where the cost is higher but the buyer is genuinely new to the brand. Each tier of the generic tree was sized by the revenue per booking it had historically delivered.
- Brand tree: chain, sub-brand, property names
- Generic tree: destination, star rating, traveller intent
- Bids weighted by realised revenue per booking, not generic CTR
Right search terms identified
Brand and generic both — "Accor Hotels" defended the loyal demand, "hotels in Kerala" and similar captured the buyer who was open to a new option.
Real-time optimization
Messaging, keyword coverage and budgets adjusted live — moving spend toward queries that were converting, away from those that weren't, every day.
Top-of-mind for travellers
Brand presence on category queries kept Accor in the consideration set even when the traveller didn't start the search by name.
30:1 ROAS, 22,000+ room nights — performance that paid for itself thirty times over.
Optimizing to ROI let Accor realise more benefit from performance marketing — by shifting spend toward the hotels that earned higher revenue per booking.
Every rupee Accor put into search came back as thirty in booked-room-night revenue.
Bookings delivered through paid search alone, across the brand's India properties.
Replaced cost-per-booking — and made every rupee of spend traceable to a business outcome.
Need search marketing measured against revenue, not clicks?
We rebuild SEM programmes around the metric that actually moves the business — and route budget to the queries that earn it back, in real time.
Trusted by marketing & growth leaders at
