We cut marketing spend by a third, growth still went up

Last year, we cut our India marketing spend by roughly a third.

On paper, that sounds like the sort of decision companies make when they’re under pressure. And usually, it comes with a familiar tradeoff. You spend less, people hear about you less, growth slows down and a few quarters later the business slows.

That isn’t what happened to us.

In FY26, adjusted net revenue grew 37% to ₹760 crore. Retail GMV grew 35%. In Q4, for the first time in Cars24’s history, we turned profitable.

At the same time, CAC as a percentage of retail GMV dropped from 3.0% in January 2025 to 1.7% in January 2026.

And interestingly, brand awareness didn’t fall with the spend either. If anything, it strengthened.

Today, around 3.8 crore Indians interact with the Cars24 ecosystem every month across Cars24, CarInfo, VehicleInfo and Team-BHP.

I don’t think this happened because we became unusually good at marketing overnight. The shift was more structural than that.

Over the last few years, we slowly started questioning a belief that most internet companies, including us, operated with for a long time: that growth mostly belongs to whoever can keep spending the most consistently.

For a while, that model genuinely worked. You bought visibility. You bought distribution. You bought attention. And as long as the spending continued, the growth usually followed.

But consumer behaviour has changed quite dramatically over the last few years, even if many companies still market as though it hasn’t.

People don’t discover products in straight lines anymore. They move between YouTube videos, WhatsApp forwards, Reddit threads, ChatGPT conversations, creator recommendations and community forums. Trust gets built slowly and often long before the actual buying moment arrives.

And increasingly, the companies that stay useful between transactions end up spending far less when the transaction finally comes.

A lot of what worked for us came from accepting that shift earlier than most. Not immediately but early enough that some of those decisions are now starting to compound.

There were a few major changes underneath that.

1. We stopped thinking about attention as something you rent

For years, most internet marketing followed a fairly simple playbook.

You bought a TV slot. A digital campaign. A celebrity face. A YouTube pre-roll. The campaign ran, people saw you and then the spend stopped, the visibility fades away.

There’s nothing inherently wrong with that model. It still works. We still use parts of it ourselves. But over time, we became uncomfortable building long-term growth entirely on channels that reset every month.

So we started putting more energy into building audiences and communities that would continue existing long after a campaign ended.

Our YouTube channel today has around 870,000 subscribers and does roughly 30 million monthly views. In FY26 alone, it generated over 400 million organic views.

But the interesting part isn’t really the scale.

It’s that a useful piece of content keeps working quietly in the background.

Someone researching their first car today still lands on reviews we uploaded two years ago. Someone trying to understand ownership costs still ends up watching an older explainer. The work continues to travel.

The same thing happened with Team-BHP.

When we brought Team-BHP into the Cars24 family, one of the easiest things we could have done was aggressively monetise it. Add conversion funnels everywhere. Push listings harder. Turn the community into a lead engine.

We deliberately chose not to.

People trust Team-BHP because it is independent, thoughtful and genuinely useful. That trust took years to build and we felt our responsibility was to preserve it.

Today, almost anyone seriously researching a car in India eventually ends up there. A thread written years ago still influences buying decisions. A detailed ownership review still ranks on Google and still helps someone choose.

That kind of trust behaves very differently from paid reach. It compounds slowly.

2. We started paying attention to where discovery is actually moving

One thing large companies often get wrong is waiting for behaviour to become obvious before adapting to it.

You can already see this happening with AI.

People aren’t only searching “best used car under 8 lakh” on Google anymore. Increasingly, they’re asking ChatGPT. Or Gemini. Or Perplexity.

And the nature of discovery changes quite meaningfully when that happens.

Traditional SEO was largely about ranking links. AI discovery is increasingly about becoming part of the answer itself.

We spent a lot of time over the last year restructuring how our inventory, data and content surfaces to these systems. Today, on a wide set of non-branded used-car prompts tested across major LLMs, Cars24 appears more consistently than anyone else in the category.

We also partnered directly with ChatGPT for inventory discovery. So if someone asks what used car they should buy in Mumbai, Cars24 inventory can appear directly inside the answer itself, along with pricing, financing information and links.

The interesting part here isn’t novelty. It’s the intent.

A traditional ad interrupts someone. An AI recommendation often appears when someone is actively trying to make a decision. That changes the economics of distribution.

And unlike performance marketing, where acquisition usually becomes more expensive every year, systems like these increasingly reward good data quality, consistency and usefulness over time.

The work is harder upfront. But it compounds.

3. We tried to make the product easier to trust

At some point, we realised a lot of marketing in the consumer internet is really just companies repeatedly trying to signal trust.

We started asking ourselves a simpler question: What if the product itself could do more of that work?

So we simplified the proposition around our retail cars into things customers could explain easily to another person.

Every retail car now comes with:

• New tyres
• New battery
• 30-day return
• Lifetime warranty

Internally, the test was simple.

Could a customer explain the proposition to a sceptical parent or friend in one breath?

Because if they can, something important happens. Your buyers start carrying the story for you. The obvious concern with guarantees like these is that they become financially unsustainable.

In practice, we saw the opposite.

Returns stayed healthy because the real work happened upstream. Better inspections. Better refurbishment standards. More discipline around the inventory itself.

A warranty only becomes expensive when the underlying product quality is inconsistent. Which is really the broader point.

Good marketing can create attention. Operational quality reduces the amount of persuasion required.

4. We stayed close to customers even when they weren’t buying

Most automotive platforms interact with customers at one moment

Buy a car.

Sell a car.

Then disappear.

But car ownership in India isn’t just one moment. It’s years of mini interactions

Insurance renewals.

FASTag recharges.

Pollution checks.

Traffic challans.

Checking what your own car is worth.

That’s where products like CarInfo and VehicleInfo became strategically important for us. Not because they directly sell cars every day, but because they allow us to remain useful long before the buying or selling intent appears.

More than 30% of customers who booked a Cars24 inspection in the second half of FY26 had already interacted with CarInfo or VehicleInfo in the previous six months.

That means we didn’t need to reacquire them from scratch through expensive paid channels. They already knew us. I think this is the part of modern consumer businesses people still underestimate.

The real moat often isn’t acquiring the customer once. It’s remaining useful enough that when the high-intent moment eventually arrives, you’re already the first place they think of.

But The Other Side Of The Coin is 

None of these bets were fast…

Building a YouTube audience takes years. Community trust takes even longer. AI visibility required infrastructure and data work long before most people even considered it a marketing problem.

There were quarters where spending aggressively on media would absolutely have produced cleaner short-term numbers. There still are.

But over time, we became more interested in building systems that continue working after the spend stops.

A trusted community thread keeps ranking.

A useful video keeps getting discovered.

A customer who already has your app comes back without reacquisition.

A strong product proposition reduces the amount of convincing required.

Those things behave less like campaigns and more like infrastructure. We still use performance marketing. We still run campaigns. This isn’t a romantic argument against advertising.

It’s simply an acknowledgement that the economics of attention have changed. The loudest company in the category doesn’t automatically win anymore. Increasingly, the companies that compound trust, usefulness and distribution over time do.