The Problem: Match Types Don't Work the Way They Used To

Most advertisers still choose match types based on a mental model that Google quietly retired years ago. If you think exact match means "only searches exactly like my keyword," your campaigns are already behaving in ways you didn't sign up for.

Here's the uncomfortable reality: Google has loosened both match types dramatically, and it never sent a memo. According to a former Google insider, "the way Exact Match works in 2024 is basically the way Broad Match worked a decade ago." Your exact match keyword [furniture store] can now serve on a search for "home furnishings shop" — a query that shares no words with your keyword at all.

Phrase match has stretched just as far. The same expert analysis shows a phrase match keyword like "Google Ads Course" matching searches such as "best live AdWords class near me." Search Engine Journal confirms the pattern: Google's 2024 updates have reshaped phrase match, expanding reach while reducing the precise control advertisers once counted on.

Why would Google do this? The answer is scale. Roughly 15% of searches every day are completely novel — queries Google has never seen before, per the same former Google employee. Loose matching lets Google monetize that long tail instead of letting searches go unmatched. That's good for Google's revenue; it's ambiguous for your budget.

This creates real confusion when you're deciding between exact and phrase match for control and cost efficiency:

• "Exact" no longer guarantees precision — close variants and implied meanings slip in, so you still need to mine the search terms report.
• Phrase match reaches every search exact match does, plus more, per Google's official documentation — so the two overlap far more than most advertisers realize.
• The loosening shows up in results: one brand campaign's ROAS dropped from a reliable 3x–4x to 2x after exact match keywords started matching "decidedly not Exact Match terms."

The practical consequence is that match type selection has become less about reach and more about steering. When Worqd builds lead generation campaigns, we treat match types as a starting hypothesis — then let search term data, not assumptions, decide where budget goes.

The old playbook is broken, but that doesn't mean match types are useless. It means the question has changed: not "which match type is safest?" but "how do I keep control when Google keeps widening the gates?" The rest of this article answers exactly that.

The Real Difference: Same Meaning vs. Includes the Meaning

Strip away the acronyms and Google's match types come down to one question: how tightly do you want to hold the steering wheel? According to Google's official documentation, exact match shows your ads on searches with the same meaning or intent as your keyword, while phrase match shows them on searches that include the meaning of your keyword — where the meaning can be implied and the search can be a more specific form of it.

That single word difference — "same" versus "includes" — has a big consequence. Phrase match covers every search that exact match does, plus more, while reaching fewer searches than broad match. A former Google insider puts it plainly: "a Phrase Match keyword can serve on all of the same searches that Exact can" (Inside Google Ads). The syntax follows suit: exact match uses square brackets like [red shoe], and phrase match uses quotes like "tennis shoes" (Google Ads Help).

In practice, both match types have loosened far beyond their names. A former Google employee argues that "the way Exact Match works in 2024 is basically the way Broad Match worked a decade ago" — an exact match keyword like "furniture store" can now serve on "home furnishings shop," and a phrase match keyword like "Google Ads Course" can match "best live AdWords class near me." Her cited reason: 15% of Google searches every day are novel, so Google widened matching to keep those queries monetized.

The numbers show what that extra reach costs and earns. An Optmyzr study of 992,028 keywords found exact match delivered the highest ROAS (415.33%) and CTR (21.66%) at the lowest CPC ($1.40), while phrase match won on conversion rate (9.31% vs. 7.98%) but posted lower ROAS (313.17%).

Contrast all of this with negative keywords, where Google flips its philosophy:

• Positive keywords get loose close-variant matching; negatives do not — Google "will not get loosey goosey in the negatives" (Inside Google Ads)
• Negative phrase match blocks searches containing the exact phrase, with no close variants (Inside Google Ads)
• Negative exact match blocks only the precise query itself

That strictness is why phrase match remains a top practitioner choice for negative keywords, even as experts move away from it on the positive side. When we build lead generation campaigns at Worqd, that asymmetry matters: you can trust a negative to block exactly what it says, but you can't assume a positive keyword stays as tight as it reads.

What the Data Says About Performance

Theory is nice, but ad budgets live and die on real numbers. Two large-scale studies — one covering nearly a million keywords, another tracking more than 16,000 campaigns — give the clearest picture yet of how exact and phrase match actually perform.

The headline finding comes from an Optmyzr study of 992,028 keywords across 15,491 ad accounts. Exact match leads on nearly every efficiency metric that matters:

• ROAS: exact match delivered 415.33% vs. phrase match at 313.17%
• CTR: exact match nearly doubled phrase match — 21.66% vs. 11.36%
• CPC: exact match was cheaper at $1.40 vs. phrase match's $1.71
• Conversion rate: the one metric phrase match won, at 9.31% vs. exact match's 7.98%

That last point matters. Phrase match converts the people it reaches at a higher rate, but it reaches more broadly and pays more per click — so the efficiency advantage still lands with exact match. If your goal is tight control over spend, the data points one direction.

But match type doesn't operate in a vacuum. A separate Adalysis study of 16,825 campaigns found that phrase match performance swings heavily depending on your bid method. Under Max Conversion Value bidding, phrase match had the highest CPA and lowest ROAS of any match type — despite converting better than broad match. Under Target ROAS bidding, the picture flips: phrase match drove 35% of revenue from just 24% of conversions, a strong revenue contribution.

Adalysis describes this as a paradox: phrase match converts well but delivers worse CPAs and ROAS, likely because bids are set too high for it. The study's takeaway — "exact match for acquisition, phrase match for revenue" — suggests the highest-value orders often come through phrase match keywords, even when efficiency metrics lag.

There's a practical lesson here for anyone running lead generation. Match type performance isn't a fixed property; it's a function of how your bidding strategy, budget, and goals interact. At Worqd, this is why we treat match type decisions as inseparable from bid method and lead-handling — a keyword choice that looks weak in one setup can be a revenue driver in another.

One caveat worth knowing: no study has compared exact and phrase match head-to-head in a single controlled experiment. Seer Interactive's tests grouped both together as a control against broad match, which underperformed by 36–74% on CPA. So treat these numbers as strong aggregate evidence, not a guarantee for your specific account. Test, watch your search terms report, and let your own data confirm the pattern.

How to Choose: A Practical Decision Framework

By now you know the definitions and the data — but match type decisions happen in real accounts, with real budgets and real bidding strategies. Here is how to turn the research into a working framework.

Choose exact match when control and budget efficiency matter most. In a large Optmyzr study of nearly one million keywords, exact match delivered the highest ROAS (415.33%) and CTR (21.66%) at the lowest CPC ($1.40). A former Google insider puts it simply: use exact match when you need tighter control or have less budget.

Choose phrase match for reach and revenue — but only under the right bidding. An Adalysis study of 16,825 campaigns found phrase match struggled under Max Conversion Value bidding, yet under Target ROAS it drove 35% of revenue from just 24% of conversions. The lesson: match type and bid method must work together, not in isolation.

Here is the practical checklist for your account:

• Default to exact match for high-precision campaigns, tight budgets, and lead generation, where it also showed the lowest CPA.
• Reach for phrase match when you want volume and revenue contribution — and pair it with Target ROAS bidding.
• Review the Search Terms Report over one to two months, then expand or contract deliberately rather than guessing upfront.
• Lean on negative keywords — where phrase match remains a top practitioner choice, since Google stays strict on negative matching.

One mistake to avoid at all costs: never run the same keyword in both exact and phrase match. Phrase match already covers every search exact match does, so duplication forces your keywords to compete for the same auction and splits your data and learning unnecessarily. One keyword, one match type, one clean signal.

Perhaps the sharpest takeaway from expert practitioners: phrase match's best current role may be in your negative keyword lists, not your positive ones. The same former Google insider who once called phrase match the "Goldilocks solution" now recommends exact or broad for positives, reserving phrase match for blocking unwanted queries — because Google has loosened positive match types so much that "the way Exact Match works in 2024 is basically the way Broad Match worked a decade ago."

At Worqd, this is the kind of decision we make when building paid search plans for lead generation clients: start with the match type your budget and bid strategy can actually support, then let search term data — not guesswork — drive expansion. If you want a partner to run that whole path from first click to booked call, book a growth call and we will find where your campaigns are stuck first.

Putting It Into Action: Testing and Refining Your Match Types

Knowing the difference between exact and phrase match matters most when you put it to work in a live account. Here's how to test, measure, and refine without guessing.

Start by watching your Search Terms Report for one to two months. Practitioner guidance suggests beginning with a single match type, reviewing what queries actually trigger your ads, then expanding deliberately — moving from exact to phrase if volume is too thin, or contracting from phrase to exact if you're pulling irrelevant searches. June 2024's enhanced reporting helps here: it now surfaces misspelled queries, revealing roughly 9% more search terms that were previously hidden in the "Other" category.

One rule before you test: never run the same keyword in both exact and phrase match. Because phrase match already covers every search that exact match does, duplication splits your data and slows learning — only one keyword per account can enter any given auction.

For a clean head-to-head test, testing best practices call for separate ad groups per match type with identical ads and landing pages, so match type is the only variable. Track these metrics over several weeks:

• Quality Score — how relevant Google judges your keyword-to-query fit
• CPC and CTR — exact match led a 992,028-keyword study with a 21.66% CTR and $1.40 average CPC
• Conversion rate — where phrase match actually won, at 9.31%
• CPA and ROAS — exact match topped the same study at 415.33% ROAS

Keep your bid method in mind as you interpret results. A study of 16,825 campaigns found phrase match struggled under Max Conversion Value bidding but performed well under Target ROAS, where it drove 35% of revenue from just 24% of conversions.

This is exactly how Worqd runs paid search: test what matters, drop what doesn't, and scale only the winners. If you'd rather have one partner handle the whole path — from first click to booked call — book a growth call and we'll find where your funnel is stuck first.