Every D2C founder in India eventually asks it. "Should I move budget from Meta to Google?"

The question sounds sensible. It's the wrong question. Performance Max vs Meta isn't a contest, because the two platforms do different jobs. Judging them against each other is like asking whether your shop needs a signboard or a salesperson.

Here's what each one actually does, and how to split the money.

Meta finds people who weren't looking for you. Google finds people who already are. You need both, in the right order.

The one difference that matters

Meta creates demand. Someone is scrolling Instagram with no intention of buying anything. Your ad interrupts them, and if the creative is good, they want the thing. That's demand creation.

Google captures demand. Someone types "best copper water bottle" into search. They already want it. Google's job is to make sure you're the one they find. That's demand capture.

This changes how you read the numbers. Google will almost always show a better ROAS than Meta. Of course it does — it's serving people who already decided to buy. That doesn't make Google better. It makes Google later in the journey.

If you shift all your money to Google because the ROAS looks better, here's what happens over the next few months. Fewer people discover you. Fewer people search your category. Search volume dries up. Your beautiful Google ROAS slowly falls, because Meta was feeding Google the whole time.

What Performance Max actually is

Performance Max, or PMax, is a Google campaign type that runs across everything Google owns at once — Search, Shopping, YouTube, Gmail, Display and Discover. You give Google your product feed, some images, some text and a goal. Google decides where to show what.

The appeal is obvious: one campaign, everywhere. The catch is that you see much less about what happened inside it. Google decides the placement mix and reports back in summary.

For most Indian D2C brands, 60–80% of PMax spend ends up in Shopping placements — the product listings with a picture and price. That single fact tells you where to focus your effort, and it isn't the ad copy. It's the feed.

Which one to start with

Your situation Start with Why
Nobody searches your category yetMetaThere's no demand to capture. You have to create it first.
Established category, people search for itGoogleDemand already exists. Capture it before paying to create more.
New brand, known categoryBoth — Meta heavierCapture what search you can, build recognition with Meta.
Strong brand, weak search presenceGoogle, urgentlyCompetitors are bidding on demand you created.
Under ₹1L/month total spendPick oneSplitting a small budget starves both. Neither ever learns.

That last row matters more than founders think. Both platforms need enough conversions to learn who your buyer is. Split ₹60,000 across two platforms and neither gets enough data to work properly. Pick one, get it working, then add the second.

The budget split

For an Indian D2C brand spending meaningfully on both, a reasonable starting point looks like this:

  1. Meta — roughly 50 to 60%. Creative-led prospecting plus retargeting. This is your discovery engine and, in India, usually your volume engine too.
  2. Google Shopping and PMax — roughly 25 to 35%. Catching people already searching for your product or category. Higher return, smaller ceiling.
  3. Brand search — a small, protected slice. People searching your brand name by name. It looks like a waste until a competitor bids on your name and you lose customers you already paid to create.
  4. Testing — about 5 to 10%. Always running. YouTube Shorts, new formats, new angles. The budget you don't judge on this month's return.

Move the ratio based on one question: is my problem that people don't know me, or that people who know me aren't finding me? The first is a Meta problem. The second is a Google problem.

Shifting all your budget to the higher-ROAS platform is how brands slowly starve themselves.

The feed is the campaign

Here's what almost nobody in India does properly. Since most PMax spend goes to Shopping, and Shopping runs entirely on your product feed — the file listing every product with its title, image, price and details — the feed is the campaign.

Google matches searches to your products using that file. A weak feed means Google can't tell what you sell, so it shows your products to the wrong people.

Four things fix most Indian feeds:

Spending a day on your feed usually beats a month of tweaking bids.

Four PMax traps

  1. It eats your brand searches. Left alone, PMax happily takes credit for people searching your brand name — customers who were coming anyway. Your reported ROAS looks great and your real growth doesn't move. Run a separate brand search campaign and exclude brand terms from PMax where you can.
  2. You can't see where the money went. PMax reports far less than standard campaigns. Many advertisers run a standard Shopping campaign alongside it to keep visibility on what's actually converting.
  3. It needs conversion data to work. PMax is an automation layer. With too few conversions it has nothing to learn from and spends badly. If you're under roughly 30 conversions a month, fix that first — often with better conversion tracking.
  4. "Set and forget" is a myth. PMax needs fresh creative, a clean feed, working tracking, and regular review of what it's actually spending on. Automation moves the work; it doesn't remove it.

Common mistakes

  1. Comparing ROAS directly. Google captures existing demand, so its ROAS is naturally higher. Comparing the two numbers side by side tells you nothing about which deserves the next rupee.
  2. Splitting a small budget across both. Under about ₹1L a month, run one platform properly instead of two badly.
  3. Ignoring the feed. Founders spend weeks on ad creative and never open the file that controls most of their PMax spend.
  4. Cutting Meta when Google looks better. Meta is what creates the searches Google captures. Cut it and the Google numbers follow it down a few months later.
  5. Judging PMax in week one. It needs a learning period. Panicking in the first fortnight and restarting the campaign resets that learning every time.

Frequently asked questions

Neither is better, because they do different jobs. Meta creates demand by putting your product in front of people who were not looking for it. Google Performance Max mostly captures demand from people already searching. Google will usually show a higher return on ad spend simply because it serves people who already decided to buy. That does not make it better, it makes it later in the customer journey. Most Indian D2C brands need both.
A common starting point is 50 to 60 percent on Meta for prospecting and retargeting, 25 to 35 percent on Google Shopping and Performance Max, a small protected slice on brand search, and 5 to 10 percent on testing. Then move the ratio based on your actual problem. If people do not know you, that is a Meta problem. If people who know you cannot find you, that is a Google problem.
Because Google is serving people who typed what they want into a search box. They had already decided to buy something in your category. Meta is interrupting someone who was not thinking about buying at all. The higher Google number reflects warmer intent, not better advertising. If you move all your budget to Google on that basis, discovery stops, search volume for your category falls, and your Google returns drift down a few months later.
A product feed is the file that lists every product you sell with its title, image, price, brand, size and other details. For most Indian D2C accounts, 60 to 80 percent of Performance Max spend goes to Shopping placements, which run entirely off that feed. Google matches searches to your products using it. A weak feed means Google cannot tell what you sell, so it shows your products to the wrong people. Fixing the feed usually beats a month of bid tweaking.
Yes, if you let it. Left unmanaged, Performance Max will serve ads to people searching your brand name, who were coming to you anyway. Your reported return looks excellent while real growth stays flat. The fix is to run a separate brand search campaign and exclude brand terms from Performance Max where the account allows it, so you can see what new demand is actually costing you.
Below roughly one lakh rupees a month in total, it is usually better to run one platform properly. Both platforms need a steady flow of conversions to learn who your buyer is. Split a small budget across two and neither gets enough data, so both perform badly and you conclude that neither works. Pick the one that fits your situation, get it profitable, then add the second.
Give it a few weeks before judging it, and avoid restarting the campaign in that period. Performance Max is an automation layer that needs conversion data to learn from. If you are getting fewer than about 30 conversions a month, it has too little to work with, and the first fix is usually your conversion tracking rather than your bids. Editing or relaunching the campaign early resets the learning each time.

In closing

Performance Max vs Meta is the wrong frame. Meta creates the demand, Google captures it, and cutting one to feed the other quietly shrinks the whole machine.

Start with the platform that matches your situation. Protect your brand searches. Fix the feed before you touch the bids. And never judge the two on ROAS alone.

Want us to look at your split, your feed, and what PMax is really spending on? Book the 45-minute private audit (free). See how this fits our ecommerce marketing work.