My short answer: the question is framed wrong. Google Ads and Meta Ads are not rivals — they sit at different points of the funnel. Google captures existing demand; Meta creates it. Asking which is better is like asking whether a hammer beats a screwdriver: it depends on what you are trying to drive.
Where I start: is there demand at all?
The first thing I look at is whether the product sits in a category people actually search for. If they are typing "women's ring designs", demand already exists on Google and leaving it uncovered means leaving money on the table. If the product defines a new category that nobody searches for yet, there is no demand to capture — the weight shifts to Meta, because people first have to discover that such a thing exists.
In practice most e-commerce accounts sit somewhere between the two. So I set the opening split from search volume data, not from theory: meaningful category volume means starting Google-heavy, thin volume means starting Meta-heavy.
The brand-search trap
This is where most budget conversations derail. If the bulk of conversions in the Google account comes from brand search, that number measures how well known the brand is, not how well Google is working. And more often than not, it was the Meta side that built that awareness.
So before any budget decision I split the Google account into brand and non-brand. Non-brand performance is Google's real contribution. Every "Google converts better, let's move budget there" decision made without that split is really billing Meta's demand generation to Google — and as you squeeze Meta, brand search on Google starts falling too.
I look at marginal return, not average ROAS
Channel budget decisions are not made on average ROAS. What matters is what the next unit of spend in that channel will return. A channel can run at 6 ROAS and still be the wrong place for more money if doubling the budget drops the marginal return to 1.5.
The way to know is measurement, not guesswork: raise budget in steps and watch the incremental conversions the incremental spend buys. If impression share lost is still high there is room to grow on Google; if you are near saturation, more money there only pushes up cost per click.
No split discussion before measurement is sound
There is almost always a gap between what Meta reports in-platform and what GA4 shows, because attribution windows and models differ. Shifting budget before both channels are measured with the same ruler is building to the wrong tape measure.
So before the split conversation I want: Pixel and Conversions API working with proper deduplication on Meta, Enhanced Conversions on Google, the same conversion definition on both, and where possible real sales data from the CRM. For lead-gen businesses I make the channel call on leads that turned into sales, not form fills — the quality gap between channels is often larger than the cost gap.
How I start in practice
On a new account, if the demand signal is strong I weight search and keep Meta small with retargeting plus prospecting tests; if demand is weak I do the reverse. Then I open a four-week reading window, watch marginal return, and move budget in 20% steps rather than one jump. Large jumps reset the learning phase on both platforms — you hand back the time you gained.
And season matters. Raising the Google share while demand is already climbing, using Meta to generate demand while it is flat — a split that ignores the calendar is applied at the wrong time even when the ratio is right.