In most B2B accounts, the larger share of paid budget goes to Google to capture demand that already exists, a smaller share goes to Meta to create demand, and LinkedIn joins only when the deal size can absorb its higher cost. There is no universal ratio. A frequent starting point is the 70/30 rule, roughly 70% on demand capture and 30% on demand creation, but the split that actually works is decided by three things: how mature demand is in your category, how long your sales cycle runs, and how much a customer is worth.
This article treats “Google or Meta” and “Google or LinkedIn” as one decision, not three separate ones, because the answer to all of them is the same frame. We keep the platform comparisons here as sections rather than sending you elsewhere, so you can size one budget across every channel in a single read.
What actually decides the Google, Meta and LinkedIn split?
Your split follows from three variables: the maturity of demand in your market, the length of your sales cycle, and the value of a won deal. Together they set how much budget should harvest intent that already exists, and how much should manufacture it.
The first variable is the most decisive. Ask a plain question: is someone already typing a query into Google that describes your solution? If yes, there is volume to capture, and it would be odd not to fund it first. If no, no search budget will create that demand, and the money has to move upstream to discovery.
This matters because of the 95:5 rule. Research popularised by Professor John Dawes at the Ehrenberg-Bass Institute holds that only about 5% of your potential buyers are in-market at any given time, while the remaining 95% will buy eventually but not now. Google Search speaks to the 5% who are looking today. Meta and LinkedIn are how you reach the 95% who are not, so that they recognise you when their window opens.
The second variable, sales cycle length, changes how patient the budget must be. On a short cycle, a dollar spent today reads inside the month. On a nine-month cycle, the buyer you reach today will not sign this year, and demand creation has to be funded as an investment that runs ahead of the sale.
The third variable is deal value. A low-commitment purchase converts off a decent landing page. A structuring purchase, with several stakeholders and a multi-year contract, needs repetition, content and proof before anyone engages. The higher the average contract value, the more it pays to have been seen before you are searched for.
Is there an ideal Google and Meta split for B2B?
There is a useful default, and there is the honest answer that it does not survive contact with your own data. The useful default is the 70/30 rule: put around 70% of budget on Google to capture demand and 30% on Meta to create it. It is a reasonable first allocation for a services business where commercial search volume already exists, and it appears across most current budget allocation guides as the industry starting point for B2B.
The honest answer is that the ideal split is a range you replace as soon as you have numbers. Reason by scenario, not by a single percentage. Three configurations cover most B2B cases.
| Situation | Google (capture) | Meta and LinkedIn (create and retarget) | Logic |
|---|---|---|---|
| Strong existing demand, short cycle | 70 to 80% | 20 to 30% | Harvest first, social recovers non-converting visitors |
| Category still needs educating | 30 to 40% | 60 to 70% | Fund awareness, Google covers brand and problem queries |
| Long cycle, high ACV, multiple buyers | 45 to 55% | 45 to 55% | Balance immediate harvest with sustained exposure |
These bands are entry points, not goals. They exist to frame a first allocation when you have no clean data of your own. The moment your own figures arrive, they decide.
There is a strategic reason the balanced row leans further toward demand creation than instinct suggests. The LinkedIn B2B Institute research by Les Binet and Peter Field found that the strongest B2B growth comes from a near even balance between long-term brand building and short-term activation, a modest tilt from the classic 60/40 brand-led split used in consumer marketing. On a long cycle with a high ticket, under-funding demand creation is the most expensive false economy in the plan.
One line before the platform detail: whatever the split, your landing page must be ready before the first dollar is spent. Budget sent to a page that does not convert is budget lost, which is why we always treat the site and acquisition together as part of B2B paid acquisition.
Meta Ads vs Google Ads in B2B: which does what?
Google and Meta are not competitors for the same job, they are two halves of the funnel. Google captures intent: it shows your ad to someone who has already decided they have a problem and typed it into a search box. Meta interrupts: it shows your ad to a qualified person who was not looking for you, in the hope of planting the problem before a competitor does.
That difference sets what each is good at in B2B. Google Search answers three questions faster than any other channel: does my market look for me, at what price, and does my page convert. Nothing else gives you those answers in a few weeks. It is almost always where a B2B account should begin.
Meta earns its share on the other side of the funnel. It reaches decision-makers who use Facebook and Instagram daily but would never search for your category, and many B2B teams generate their cheapest qualified leads there using lead-form ads with qualifying questions built in. It is also where retargeting lives: the visitor Google sent you but who did not convert is recovered on Meta, at a fraction of a fresh acquisition cost, even as the media itself keeps getting more expensive, as we cover in how much Facebook ads cost.
The practical rule that follows: Meta needs a traffic history to perform. Retargeting, lookalikes and similar audiences all draw on a record of visitors and conversions. Launching Meta against an empty account throws away half its capability, which is the main reason to start with Google and let it build the audience Meta will later use.
LinkedIn Ads vs Google Ads B2B: when is LinkedIn worth it?
LinkedIn sits at the far end of the demand-creation spectrum, and it competes with Google only in the sense that both want the same budget. Google captures demand cheaply and broadly. LinkedIn creates demand expensively and precisely, by targeting on job title, seniority, company and industry in a way no search campaign can match.
The cost gap is real and worth naming. Google Search runs an average cost per click of $5.26 across all industries, according to the 2025 WordStream benchmarks built on 16,446 US search campaigns between April 2024 and March 2025. LinkedIn, by contrast, ran a cost per click between $10.48 and $15.72 across quarters in the 2025 HockeyStack benchmark of more than seventy B2B SaaS companies and 28 million dollars of spend. LinkedIn costs roughly two to three times as much per click.
So LinkedIn is worth it when the maths of your deal absorbs that premium. A cost per lead of a few hundred dollars is painful if your contract is worth a few thousand and easy if it is worth six figures. The test is your average contract value: the higher it is, and the more your buying committee is defined by exact job titles, the more a share of budget belongs on LinkedIn. Below a certain deal size, the same money buys more qualified pipeline on Meta or Google, and LinkedIn stays off the plan.
How much should you spend to test each channel credibly?
A credible test is defined by a volume of conversions and a duration, not by a headline budget. The question is not “how much to spend” but “does this budget reach enough conversions, for long enough, for the result to mean anything”.
Two reference points frame it. The first comes from Google: an automated bid strategy can take up to three weeks or one to two conversion cycles to calibrate. Any reading taken before then judges a system that is still learning. The second is your own sales cycle. If deals sign in four months, a six-week test tells you your cost per lead, not your cost per acquired customer, and only the second one pays the bills.
To size the spend, start from a cost per lead and work back. The WordStream data puts the median B2B cost per lead at $70.11 across all industries, and at $103.54 for Business Services. As an illustration, not a published figure: reaching thirty qualified leads in Business Services at that median would take on the order of $3,100 in media over the test. These are US medians, useful for sizing, not a forecast for your account. If the budget will not cover the volume over the duration, do not cut the duration. Cut the scope, and test one channel rather than two.
How do you reallocate budget by performance?
Reallocate on cost per qualified lead, the cost of a prospect your sales team agrees to work. It is the only metric that ties a media spend to a business reality. Everything else, cost per click, raw cost per lead, impressions and engagement, helps you diagnose but should never decide.
The mechanic is simple to state and demanding to hold. Each month, compare cost per qualified lead across channels, move a share of budget toward the stronger one, and watch whether the advantage holds. Move in steps, never all at once: a channel abruptly starved of budget loses its learning history, and you pay to rebuild it on the way back. A few traps recur:
- Splitting the budget in half on day one. Two halves too small for either to leave the learning phase gives you two inconclusive tests instead of one usable answer.
- Reallocating every week. A budget that moves every Monday never lets a channel stabilise. Set a monthly rhythm and hold it.
- Judging Meta or LinkedIn by Google’s cost per lead. Expecting a demand-creation campaign to match a search cost per lead is like blaming a seed for not being a harvest.
- Forgetting brand search in the maths. Queries on your own name cost little and convert well. Counted alongside non-branded acquisition, they flatter the average and hide the true cost of winning new demand.
For any of this to be possible, you have to connect leads to their source all the way to signature. Without that chain between ad, CRM and revenue you will always reallocate on approximations, and the two metrics that matter most, cost per qualified lead and its relationship to lifetime value, are exactly the ones we untangle in ROAS, MER, CAC and LTV.
In short
- The split is set by three variables: demand maturity, sales cycle length and average contract value, not by a fixed ratio.
- Start with Google in most cases: it gives the fastest data and builds the audience Meta needs. Add Meta for demand creation and retargeting, and LinkedIn only when your ACV absorbs its two to three times higher cost per click.
- The 70/30 rule is a fair first allocation, not a target. Replace it with your own cost per qualified lead as soon as you have data.
- A credible test is measured in conversions and duration, covering at least the algorithm’s learning period and a sales cycle.
- Reallocate on cost per qualified lead, in monthly steps, never on cost per click or volume.
The right split is not the one an article prescribes, it is the one your own numbers end up imposing, provided you measure them cleanly from day one. If you want to size one budget across Google, Meta and LinkedIn against your real deal economics, book a diagnostic and we will map it to your pipeline together.