Google Ads: When It's Worth It (And When It Isn't)

Google is the only major network where users tell you what they want before you spend a dollar. That intent signal is the entire reason to be here — and the reason it's expensive. If you're not buying intent, you're probably overpaying.

Who crushes it on Google. Anything with explicit purchase intent: home services (plumbers, roofers, movers), B2B SaaS with a known category, legal, healthcare, locksmiths, e-commerce with branded demand, anything people Google in a panic. If your customer types a query that contains your solution, Search is the highest-converting channel that exists. Period.

Who shouldn't bother. New product categories nobody searches for yet. Impulse-buy DTC at low AOV (Meta and TikTok will eat Google for breakfast on a $35 candle). Pre-revenue brands trying to build awareness — Google is a harvester, not a demand creator. If your category has under ~1,000 monthly searches and no one knows your brand, you're starving.

2025 benchmarks worth memorizing. Search CPCs run $1–$4 for most B2C, $4–$15 for B2B and professional services, and $50+ for legal, insurance, and rehab. Performance Max delivers 3–6x ROAS for e-commerce with a clean feed and 30+ reviews on hero SKUs. YouTube CPVs sit at $0.02–$0.12. Display CPMs are $2–$10 but Display traffic converts roughly 1/10th as well as Search, which most calculators conveniently forget.

The campaign mix that actually works in 2025. Search on your money keywords (exact + phrase, not broad unless you trust Smart Bidding with at least 30 conv/week). Performance Max for e-commerce or for harvesting cross-network demand once you have conversion volume. YouTube for remarketing audiences only — prospecting on YouTube is a slow burn that needs $5K+/month and patience most teams don't have.

Google vs Microsoft Ads. Microsoft is 20–35% cheaper CPC on identical keywords and skews older, more affluent, more B2B (Office/LinkedIn integration). If you're already running Search profitably on Google, port the campaigns to Microsoft — it's free incremental volume. But Microsoft's total query volume is roughly 7–9% of Google's, so it's a supplement, never a primary channel.

Google vs Meta. Different jobs entirely. Google captures existing demand; Meta creates it. If your product solves a problem people know they have, Google. If you have to teach someone they want it, Meta. Most growing brands need both — Meta to fill the funnel, Google to close it. The mistake is running them on the same attribution window and pitting them against each other.

What's changed and what most teams are getting wrong. Manual CPC is essentially dead — Smart Bidding outperforms it once you have conversion data, but it needs that data. New accounts under 15 conv/week should run Maximize Conversions, not Target CPA. Performance Max is no longer optional for e-commerce, but it will cannibalize your branded Search if you don't add brand exclusions (this still isn't on by default for most accounts — check it). Broad match has gotten dramatically better with Smart Bidding but is still a money pit without it.

Where your budget actually goes. Expect 15–25% of spend on irrelevant queries even with tight match types — that's the cost of doing business on Google now. Negative keyword hygiene weekly for the first 90 days. After that, monthly is fine.

Example

A regional HVAC company spending $6,000/month: $4,000 on Search across 'ac repair [city]' and emergency-intent terms (CPC $8–$14, CPA $60–$95), $1,500 on Local Services Ads (pay-per-lead at $25–$50), $500 on YouTube remarketing to site visitors. Books 60–90 jobs/month at an average ticket of $450. Microsoft Ads added at month three contributed another 15 jobs for $800 spend.

Questions