Google Ads shows ROAS 8, Meta only 4. Reducing Facebook and Instagram to invest more in Google seems obvious.
Short-term reallocation may improve visible figures, but gradually weaken new-customer acquisition, branded searches and eventually Google’s own results.
The platforms often act at different journey stages. The sales-closing channel may not have created initial interest.
Ask not ‘which platform has better ROAS?’, but what additional value does each euro on Meta or Google create?
Google Ads often captures expressed demand
On Search and Shopping, users express needs through searches:
- ‘invoicing software for freelancers’
- ‘sofa bed fast delivery’
- an exact brand or product name.
Intent is visible. Ads appear while people actively seek solutions, sometimes minutes before purchase.
This explains Google’s often high ROAS, especially branded. But branded searches do not prove Google created demand: discovery may come from Meta, content, recommendations, email or other Google campaigns.
Google also builds consideration through YouTube, Demand Gen, Display and Performance Max.
Meta Ads often creates and stimulates demand
Facebook and Instagram users are not necessarily searching for your product. Ads interrupt feeds to introduce problems, desires or solutions.
Meta is particularly strong at:
- brand discovery
- introducing products and benefits
- reaching people unaware of the solution
- building consideration and re-engaging visitors.
Meta is more than interest-based advertising. Delivery relies on objectives, signals, creative and account data. It also closes sales directly: measure each campaign’s real role.
Why last click often favours Google Ads
Imagine this journey:
- someone discovers your brand in an Instagram video
- visits your site without buying
- searches your brand on Google two days later
- clicks a Search ad and completes the order.
Last click gives Google all credit, but without the Meta exposure the branded search may never have happened.
Google Analytics defines attribution as credit allocation across pre-conversion touchpoints. Its data-driven model may distribute credit across interactions, while last click favours the final channel.
Platforms use their own windows, so Meta and Google may claim one conversion or differ from GA4 and payment tools.
Adding platform figures does not reliably measure total performance.
What happens if you abruptly cut Meta Ads?
Sharp budget cuts may cause delayed effects:
- fewer new prospects discover the brand
- retargeting audiences replenish more slowly
- direct traffic and branded searches may fall
- Google has less existing demand to capture
- marginal cost rises as profitable Search potential saturates.
Decline may not be immediate. Previously exposed people keep searching and buying for days or weeks. Meta seems unnecessary until overall slowdown emerges.
This does not mean never reducing Meta: campaigns may be non-incremental, poorly targeted or unprofitable. Base decisions on overall measurement or tests, not Ads Manager ROAS alone.
Why shifting all budget to Google also saturates
Profitable searches are finite. Once the best queries are covered, more budget often buys generic searches, lower-intent audiences or costlier auctions.
Historical average ROAS need not represent the next euro’s return.
Allocate budgets using marginal ROAS : what extra revenue or margin will another €10,000 on Google generate versus the same increase on Meta?
Highest average ROAS does not always mean greatest remaining growth capacity.
How should you split Meta Ads and Google Ads budgets?
1. Build a common source of truth
Compare platforms with GA4, CMS, CRM, payments and margin. Platforms alone cannot be your marketing accounts.
2. Separate branded and non-branded Google searches
Branded Search often has excellent ROAS by capturing existing demand. Analyse separately from generic searches to avoid overstating Google’s new-customer capacity.
3. Track overall indicators
Measure:
- new-customer acquisition cost
- contribution margin after advertising
- total revenue / media budget, sometimes called MER
- new-customer share
- branded search volume
- overall conversion rate
- revenue changes as budgets change.
4. Change budgets gradually
Move part of the budget, then wait the average conversion delay. Abrupt cuts disrupt delivery and analysis.
5. Measure incrementality
With enough volume, use geo-tests, controls, experiments or lift studies to estimate sales that would not happen without ads, rather than merely claimed sales.
Meta also offers incremental attribution optimising for conversions its models consider caused by ads.
Frequently asked questions
Is Google Ads always more profitable than Meta Ads?
No. Product, awareness, search volume, creative and buying cycle matter. Google may close existing demand better; Meta may have more capacity to create demand.
Should Meta and Google conversions be added together?
No. Both may attribute one sale. Use a common source and conversion paths to limit double counting.
Which metric best allocates budget?
None alone. New-customer CAC, margin, MER, marginal return and incrementality tests provide a stronger view than platform ROAS.
Manage acquisition as a system with Alyads
At Alyads, we examine each channel’s contribution to discovery, consideration and conversion rather than artificially pitting Meta against Google. Budget goes where it genuinely creates growth.
Want profitable acquisition in France or Israel? Let’s talk.
