← Back to blog Agencies

Google Ads vs Facebook Ads: How to Decide Where a Client's Budget Goes

Your client spends $40,000 a month on paid media. Meta gets $28,000 and reports a 3.1x ROAS. Google gets $12,000 and reports 5.4x.

On the monthly call, the founder asks the question you could see coming from the first slide: why don't we just move the Meta money to Google?

Most of us have watched a client win that argument, move the budget, and then spend the next quarter wondering why total revenue fell while Google's ROAS stayed high.

The usual answer to Google Ads vs Facebook Ads ("it depends") is true but not much help on a client call. Here is how I think about it when it is a client's money and I have to defend the split.

What Google Ads and Facebook Ads are each good at

Start with what a person is doing at the moment your ad reaches them.

On Google, they are searching. They typed "waterproof hiking boots women's" into a box. The want existed before your ad showed up, and your job is to be the best answer when they ask.

On Meta, they are scrolling. Nobody opened Instagram to buy hiking boots. Your ad has to interrupt them and make them want something they were not thinking about.

So Google mostly captures demand and Meta mostly creates it. Performance Max and YouTube push Google into discovery, and Meta retargeting is pure capture, but as a starting model for a client conversation it explains most of what you see in the numbers. (Most clients still say Facebook Ads for what is now Meta Ads. I use both names because they do.)

Is Google Ads better than Facebook Ads?

For a given client, one of them usually is. Which one depends on a question you can answer before spending a dollar: do people already search for what this client sells?

A plumber in Phoenix has search demand. People type "emergency plumber near me" at 11pm with water on the kitchen floor. Google wins that easily.

A DTC brand selling a new kind of standing desk mat has very little search demand. People do not search for a product they have never heard of. Meta wins there.

Most ecommerce clients sit in between. A home goods store spending $40k a month has search demand for "linen duvet cover queen," but it also needs Meta to reach people who have never seen the brand. The answer there is both, and the hard part is the ratio.

Why comparing Google and Meta ROAS side by side misleads clients

Back to the 5.4x versus 3.1x.

Each number comes from a different system counting its own conversions with its own rules. Google Ads usually counts conversions up to 30 days after a click. Meta defaults to 7 days after a click plus 1 day after a view. When the same customer sees a Meta ad on Tuesday and searches the brand on Friday, both platforms can count that purchase.

I covered this in why Meta and Google both claim the same sale. The short version for this argument: Google's ROAS is often higher partly because Meta did the persuading and Google caught the customer on the way to checkout.

Brand campaigns widen the gap, because their ROAS comes from people who already knew the client's name. That is the trap I covered in why branded search ROAS looks so good.

So when a client says Google is better and the numbers prove it, the numbers prove something narrower. Google is better at claiming the last click. That says very little about where the next $10,000 should go.

Getting the "why not all Google?" question every month? Book a demo and we will show you how Coretas reads a client's Google, Meta and GA4 data together and builds a platform split you can defend line by line.

Book a demo

Google Ads vs Meta Ads compared

Google AdsMeta Ads (Facebook and Instagram)
What the person is doingSearching for something specificScrolling, not shopping
Main jobCapturing demand that already existsCreating demand that does not exist yet
Strongest forKnown products, services, local intent, high-ticket researchNew or visual products, impulse buys, broad audiences
How it tends to reportHigh ROAS, lifted by brand terms and last-click creditLower click ROAS, plus view-through conversions clients distrust
What limits scaleSearch volume in the categoryFresh creative and audience fatigue

When Google Ads should get more of the budget

I lean toward Google when the client has real non-brand search volume. If the core category terms show a few thousand searches a month in the client's market and the cost per click leaves room for margin, Google is where the efficient money starts.

Other signs that point to Google:

  • Buyers compare options before they purchase (B2B, high-ticket items, most services).
  • The product is a replacement or repeat purchase that people already know they need.
  • Local intent ("near me" searches) drives most of the business.
  • The client has a clean Shopping feed with competitive prices.

The ceiling to watch is search volume. Once you capture most non-brand searches at a sensible cost, more Google budget buys worse clicks at higher prices, and clients see ROAS slide and blame the agency when the category has simply run out of searches.

When Facebook Ads should get more of the budget

Meta earns a bigger share when the client has to create the want before anyone will search for it.

  • The product is new, unusual, or easier to understand on video.
  • The audience is defined by interests or life stage more than by search terms.
  • Average order value is low enough that impulse purchases make sense.
  • Search volume for the category is small, or marketplaces own the results page.

The ceiling on Meta is creative. An account spending $30,000 a month needs a steady supply of new ads, or it hits frequency fatigue long before it runs out of audience.

Cutting Meta also rarely shows up as a drop in Meta revenue alone. It shows up two to six weeks later as fewer branded searches and weaker Google numbers. Clients who cut Meta to fund the winner often end up hurting the winner.

How to split a client's budget between Google and Meta

This is the process I use for a new client or a quarterly review.

  1. Pull 90 days of spend and revenue from both platforms, plus revenue from the client's store or GA4. Treat the store's revenue as the source of truth.
  2. Calculate blended ROAS: total revenue divided by total ad spend across every platform. This is the number the business actually lives on.
  3. Separate brand search from everything else on Google and report it as its own line.
  4. Size non-brand search demand in Keyword Planner. That gives you a realistic ceiling for Google.
  5. Fund Google to capture that demand at a profitable cost, then put the remaining budget into Meta prospecting.
  6. Change one platform at a time, 15 to 20% per move, and watch blended ROAS for two to three weeks before the next change.

For a lot of ecommerce clients this lands somewhere between 60/40 and 70/30 in Meta's favor. For service and B2B clients it often flips. I would not treat any ratio as a rule, because the ratio is an output of the process.

Step six is where agencies usually lose discipline. Move both platforms in the same week and the blended number cannot tell you which change did what.

How to explain the Google vs Meta split to a client

The founder in the opening scenario does not want an attribution lecture. What works for me is changing the scoreboard. Put blended ROAS at the top of every report and show platform ROAS underneath as diagnostic detail. Once the client is watching total revenue against total spend, the move-everything-to-Google argument loses most of its pull, because they can see what happened the last time someone tried it.

If they still want to test it, let them. Cut Meta by 25% for three weeks, hold Google steady, and watch total revenue and branded search volume. A test the client chose convinces more than any slide.

And write the split down with the reasoning next to it. "Google gets $12,000 because profitable non-brand search in your category tops out around there." In my experience, clients argue far less with a number when the reasoning sits right next to it.

Doing this for every client, every month

For one client, all of this takes an afternoon. For twenty clients every month it is a full-time job, which is why most agencies drift back to whatever ratio they set at onboarding.

Coretas does this work for agencies. It reads each client's Google Ads, Meta Ads and GA4 history, shows spend, ROAS and CPA for both platforms next to blended ROAS in one dashboard, and builds one cross-platform media plan and budget with the platform mix explained, so you can defend the split to the client. Budget moves come in as daily ranked recommendations that your team approves, Coretas Copilot answers why it suggests each one, and every client sits in one workspace with its own access controls. You can see the plans and agency pricing on the Coretas pricing page, or start a free trial on one client account.

Want to see the right split for one of your clients? Start a free trial, connect their Google and Meta accounts, and Coretas shows blended ROAS and a recommended platform mix from their own data. Seven days, no card required. Prefer a walkthrough? Book a demo.

Start free trial

Frequently asked questions

Is Google Ads or Facebook Ads better for ecommerce?

It depends on whether people already search for what you sell. Stores with clear search demand for their categories usually get efficient sales from Google Shopping and Search, while new or visual products tend to grow faster on Meta. Most brands with a meaningful budget run both.

Why does Google Ads show a higher ROAS than Meta Ads?

Google reaches people who are already searching, often late in the buying process, and it frequently gets last-click credit for customers that Meta ads introduced to the brand. Brand campaigns inflate it further. A higher platform ROAS does not mean an extra dollar will return more on Google.

How should I split my budget between Google Ads and Meta Ads?

Size the non-brand search demand for your category first and fund Google to capture it at a profitable cost. Put the rest into Meta prospecting, then adjust one platform at a time in 15 to 20% steps while watching blended ROAS.

What is blended ROAS?

Blended ROAS is total revenue divided by total ad spend across every platform, measured from your store or analytics rather than from the ad platforms. It cannot double count a sale, which makes it the best single number for judging whether a Google and Meta mix is working.

Can I run Google Ads and Facebook Ads at the same time?

Yes, and most businesses with a meaningful budget should, since Meta creates interest and Google captures the searches it produces. Judge them together on blended numbers, because each platform will claim some of the same sales.

How long should I test a budget shift between Google and Meta?

Give each change at least two to three weeks before judging it. Cuts to Meta can take several weeks to show up as lower branded search volume. Change only one platform per test so you can tell which move caused the result.