Match Types in 2026: Why Exact Match Isn't Exact and What to Do About It

TL;DR: Google has quietly redefined every match type. Exact match now triggers for close variants and meaning matches. Phrase match behaves like the old broad match modifier. Broad match gives Google maximum freedom to interpret your keywords. The result: 20-40% of actual search queries are invisible to you, and broad match CPCs rose 29% between 2023-2025. The fix is a disciplined progression path — start exact, move to phrase after data, add broad only after 30-50 conversions.

If you set up Google Ads five years ago and have not revisited your match type strategy, your campaigns are not running the way you think they are. Google has fundamentally changed how match types work — without changing their names.

This guide explains what each match type actually does in 2026, when to use each one, and how to protect your budget from Google's expanding interpretations.

---

How Match Types Have Evolved

Match types used to be straightforward. Exact match meant your ad showed for that exact search. Broad match meant Google could get creative. The boundaries were clear.

Those boundaries no longer exist.

The Timeline of Erosion

Each change expanded Google's ability to show your ads to more searches — searches you did not explicitly target. More impressions and more clicks benefit Google's revenue. Whether those clicks benefit your business is a different question.

---

Match Types in 2026: What They Actually Do

Exact Match [keyword]

What Google says: Shows ads for searches that have the same meaning as your keyword.

What actually happens: Your ad can appear for searches Google considers semantically equivalent — even if the words are completely different.

Example: Keyword: [plumber near me]

The problem: The last two examples show searches that may have different intent. Someone searching for "pipe repair professional" might want a specialist, not a general plumber. "Handyman for plumbing" is explicitly a different service category. But Google's meaning-match algorithm may trigger your ad anyway.

Exact match in 2026 is the tightest control you have — but it is not truly exact.

Phrase Match "keyword"

What Google says: Shows ads for searches that include the meaning of your keyword.

What actually happens: Phrase match now behaves like the old Broad Match Modifier (which Google eliminated in 2021). It matches searches that contain the concepts in your keyword, plus additional terms.

Example: Keyword: "plumbing repair service"

Key difference from old phrase match: Old phrase match required the exact phrase in order. New phrase match requires the meaning, in any order, with additional context.

Broad Match keyword

What Google says: Shows ads for searches related to your keyword.

What actually happens: Google uses your keyword as a loose theme and matches based on AI interpretation of user intent. With broad match, Google considers:

• The keyword itself
• Other keywords in your ad group
• Your landing page content
• The user's recent search history
• The user's location and device

Example: Keyword: plumbing repair

The core risk: Broad match optimizes for volume, not precision. Google's own documentation admits broad match gives them "maximum flexibility" with keyword interpretation. That flexibility serves Google's revenue model — more impressions, more clicks, more ad spend.

The data confirms the risk: Broad match CPCs rose 29% between 2023-2025 while masking higher CPAs behind increased volume. It feels efficient because CPC drops, but cost-per-acquisition often climbs.

"Broad Match is a classic example of false efficiency. It feels like it's working because your cost-per-click drops, but your cost-per-acquisition skyrockets." — TG Digital

---

The Invisible Query Problem

One of the most significant issues with modern match types is query visibility. Google restricts which search queries you can see in your Search Terms Report.

20-40% of actual search queries remain invisible to advertisers due to "low-volume" privacy restrictions.

What this means in practice:

A $20 million ad spend analysis found that up to 85% of spend went to unseen queries in some accounts. You are paying for clicks you cannot audit.

The broader match type you use, the more invisible queries you generate. Exact match gives you the most visibility. Broad match gives you the least.

How to Combat Hidden Queries

1. Use exact and phrase match as your foundation. These generate fewer invisible queries.

2. Review your Search Terms Report weekly. The queries you can see are your best signal for negative keyword additions.

3. Run N-gram analysis monthly. Break search queries into 1-4 word phrases to identify patterns of waste:

   • 1-2 word n-grams: Finding negative keywords
   • 3-4 word n-grams: Finding new keywords to target as exact/phrase

4. Set up automated search query mining scripts. PPC specialists run an average of 3.8 scripts per account. The Search Query Mining script is one of the five scripts every account needs.

5. Track "Other search terms" category in your reports. When this category represents a growing percentage of spend, it signals that hidden queries are consuming your budget.

For the structural framework to organize these keywords, see The Optimal Google Ads Account Structure for Service Businesses.

---

When to Use Each Match Type

The Decision Framework

Exact Match: Your Starting Point

Use when:

• Launching new campaigns with no conversion history
• Targeting your highest-intent, highest-value keywords
• Budget is limited and every click must count
• You need maximum control over which queries trigger your ads

Best for:

• Emergency service keywords ("emergency plumber near me")
• Branded terms ("your company name")
• High-CPC industries where one irrelevant click costs $50+
• Service businesses under $3,000/month ad spend

How many keywords: 10-30 exact match keywords is a strong starting foundation. Focus on keywords with commercial or transactional intent.

Phrase Match: Your Growth Phase

Use when:

• You have 15-30 conversions per month and understand which queries convert
• You want to discover new search variations without losing control
• Your negative keyword list is robust (100+ terms minimum)
• You review Search Terms Report at least weekly

Best for:

• Service-specific keywords ("plumbing repair service")
• Location-modified searches ("dentist in [city]")
• Service + modifier combinations ("affordable house cleaning")

Transition from exact to phrase:

1. Identify exact match keywords that consistently convert
2. Add phrase match versions alongside the exact match
3. Monitor for 2-3 weeks
4. If phrase match generates quality traffic at acceptable CPA, gradually shift budget

Broad Match: Only With Smart Bidding + Data

Use when (ALL conditions must be met):

• You have 30-50+ conversions per month per campaign
• Smart Bidding is active (Target CPA or Target ROAS)
• Your negative keyword list exceeds 200 terms
• You review Search Terms Report weekly
• You accept that 20-40% of queries will be invisible

Why broad match needs Smart Bidding:

Without Smart Bidding, broad match shows your ads to the widest possible audience at whatever CPC wins the auction. Smart Bidding adds a filter — it adjusts bids in real-time based on conversion probability. The combination means Google can interpret queries broadly while (theoretically) lowering bids on low-intent queries.

The catch: Smart Bidding needs conversion data to work. Without 30+ monthly conversions, Smart Bidding is guessing — and broad match amplifies those guesses across a huge query surface.

If Google is pushing you toward broad match before you have this data, they are optimizing for their revenue, not your results.

---

The Progression Path

Here is the step-by-step match type progression for service businesses:

Stage 1: Data Collection (Months 1-2)

Match types: Exact only
Bidding: Manual CPC
Keywords: 10-30 high-intent exact match
Negatives: Build initial list of 200+ terms
Goal: Establish baseline CPA and conversion data

Actions:

• Review Search Terms Report daily for the first two weeks, then every 2-3 days
• Add negative keywords aggressively
• Track which keywords drive actual conversions (not just clicks)
• Do not change bidding during learning phase (7-14 days)

Stage 2: Controlled Expansion (Months 2-3)

Match types: Exact + Phrase
Bidding: Manual CPC → Maximize Conversions (when 15+ conversions/month)
Keywords: Add phrase match for proven converters
Negatives: Growing weekly based on Search Terms Report
Goal: Discover additional converting queries

Actions:

• Add phrase match versions of your top 10 exact match keywords
• Create new ad groups if phrase match reveals high-volume themes
• Begin transitioning to Maximize Conversions when data supports it
• Set Maximize Conversions budget at current level — do not increase simultaneously

Stage 3: Algorithm Partnership (Months 3-6)

Match types: Exact + Phrase + Selective Broad
Bidding: Target CPA or Target ROAS
Keywords: Add broad match for your top 5 converting themes
Negatives: 200+ terms, reviewed weekly, N-gram analysis monthly
Goal: Let Smart Bidding find additional converting queries

Actions:

• Add broad match only for keyword themes with 30-50+ monthly conversions
• Set Target CPA 5-10% above your current average CPA
• Monitor broad match performance separately — create a segment or label
• If broad match CPA exceeds your target by more than 20% for 2+ weeks, pause it

Stage 4: Optimization (Month 6+)

Match types: Full spectrum based on data
Bidding: Portfolio strategies across campaigns
Keywords: Continuously refined based on performance
Negatives: Automated scripts + manual review
Goal: Maximum conversions at target CPA/ROAS

Actions:

• Run automated search query mining scripts
• Implement N-gram analysis for negative keyword discovery
• Use ad variations experiments to test copy changes
• Restructure quarterly based on performance data

---

Match Type Strategy by Service Industry

Different industries have different keyword dynamics. Here are match type recommendations by vertical:

High-CPC Industries (Legal, Medical, HVAC)

CPCs in these industries range from $5 to $250+ per click. Match type mistakes are proportionally expensive.

Medium-CPC Industries (Cleaning, Tutoring, Pest Control)

Low-CPC Industries (Car Rental, Restaurants)

---

Negative Keywords: Your Most Important Match Type Tool

If match types are the accelerator, negative keywords are the brake. Without them, every match type bleeds money.

The data:

• 25% of accounts have zero negative keywords
• Accounts with negative keywords convert at 13% vs 4.6% without them — a 3x improvement
• A well-maintained negative keyword list of 200+ terms can save 20-30% of monthly spend

Negative Keyword Match Types

Negative keywords have their own match types, and they work differently from positive keywords:

Important: Negative keywords do NOT block close variants or synonyms. If you add "free" as a negative, searches for "complimentary" will still trigger your ads. You must add both.

Building Your Negative Keyword List

Start with universal negatives for service businesses:

Then build from data:

1. Week 1: Review Search Terms Report daily. Add obvious waste.
2. Weeks 2-4: Review Search Terms Report every 2-3 days.
3. Monthly: Run N-gram analysis to catch patterns.
4. Ongoing: Automated scripts flag high-spend, zero-conversion queries.

Negative Keyword Levels

---

The Real Cost of Wrong Match Types

Let us quantify the impact with a real scenario:

Service business: HVAC repair in Dallas. Monthly budget: $5,000. Average CPC: $15.

Scenario 1: All Broad Match, No Negatives

Scenario 2: Exact + Phrase Match, 200+ Negatives

The difference: Scenario 2 generates 56-69% more conversions at 36-38% lower CPA — from the same budget.

The savings compound over time. At $1,000-$1,300 saved per month, that is $12,000-$15,600 per year redirected from waste to profitable clicks.

---

Match Type Interaction With Ad Groups

Your match type strategy and ad group structure must work together. Here is how they interact:

Single Theme Ad Groups (STAGs) + Match Types

Ad Group: Water Heater Repair
├── [water heater repair] — Exact match
├── "water heater repair service" — Phrase match
└── water heater repair — Broad match (add later when data supports)

Key rule: If you run the same keyword in multiple match types within one ad group, exact match takes priority when the query matches exactly. Phrase match captures variants. Broad match catches everything else.

Cross-Campaign Match Type Strategy

For remarketing campaigns, you can afford to use broader match types because the audience has already visited your website. They are pre-qualified, so even loosely matched queries are more likely to convert.

---

AI Max for Search: Google's Latest Match Type Push

In 2025-2026, Google introduced AI Max for Search campaigns, which promises automated keyword expansion powered by AI. Google claims a +24.9% improvement in optimization score when enabled.

What AI Max Actually Does

AI Max analyzes your landing page text, existing keywords, and ad copy to automatically find new search queries it believes are relevant. It essentially applies broad match logic across your entire campaign — without you explicitly choosing broad match.

Why You Should Be Cautious

The optimization score improvement (24.9%) measures compliance with Google's recommendations, not actual campaign performance. As we covered in the invisible query section, Google's metric alignment does not always match advertiser ROI.

The risk: AI Max targets extremely low-intent searches based on website text, which can waste budgets on irrelevant queries. If your plumbing website mentions "kitchen renovation tips" in a blog post, AI Max might trigger your ads for "kitchen renovation ideas" — an informational query with zero conversion intent for a plumbing service.

When AI Max Might Work

• You have 100+ monthly conversions and robust negative keyword lists
• Your landing pages are focused exclusively on service pages (no blog content)
• You can dedicate time to weekly query monitoring
• Your budget can absorb a testing period

When to Avoid AI Max

• You have limited conversion data (under 50/month)
• Your website has diverse content (blog posts, educational resources)
• You are budget-constrained and cannot afford waste
• You prefer control over volume

Recommendation: Treat AI Max as an experiment, not a default. Run it in a separate campaign or use Google's experiment feature to split-test against your standard campaigns. Monitor for 30 days with aggressive negative keyword management.

---

The Match Type Audit: Evaluating Your Current Setup

If you already have a running Google Ads account, here is how to audit your match type strategy:

Step 1: Export Your Search Terms Report (90 Days)

Pull every search query that triggered your ads over the past 90 days. Sort by:

1. Cost (highest first) — identify your biggest spend queries
2. Conversions (highest first) — identify your best performers
3. Cost with zero conversions (highest first) — identify your biggest waste

Step 2: Classify Query Intent

For your top 50 queries by spend, classify each as:

Step 3: Calculate Match Type Performance

Create a pivot table showing performance by match type:

What to look for:

• If broad match CPA is more than 30% higher than exact match, broad match is likely underperforming
• If phrase match CVR is close to exact match CVR, you are grouping keywords well
• If broad match generates high volume but low conversions, it is wasting budget

Step 4: Identify "Promotion Candidates"

Look for phrase or broad match queries that:

• Convert consistently (3+ conversions in 90 days)
• Have a CPA at or below your target
• Are not already targeted as exact match keywords

These queries should be promoted to exact match keywords with their own ad copy. This captures the converting traffic at maximum Quality Score efficiency.

Step 5: Build Your Negative Keyword Action List

From the audit, you should have a list of:

• Informational queries to negate
• Irrelevant queries to negate
• High-cost, zero-conversion queries to negate
• Pattern-based negatives (entire categories like "jobs," "free," "DIY")

Target: Add 50-100 negative keywords from this single audit. Repeat quarterly.

---

The "Google Says vs Reality" Match Type Edition

The pattern is consistent: Google recommends options that increase the volume of searches your ads appear for. More impressions, more clicks, more revenue for Google. Whether those additional clicks convert for your business is secondary to Google's revenue model.

"You lose a lot of control over your ads and account and give it over to Google, who definitely does not have you saving money as a goal." — Ryder Meehan, 15 years Google Ads experience, Upgrow

---

Monitoring and Optimization

Weekly Checklist

• Review Search Terms Report for all campaigns
• Add at least 5-10 new negative keywords
• Check broad match performance vs exact/phrase
• Monitor "Other search terms" spend percentage
• Flag any query with 50+ clicks and zero conversions

Monthly Checklist

• Run N-gram analysis (1-4 word phrase breakdown)
• Evaluate match type performance by CPA and ROAS
• Consider promoting high-performing phrase match queries to exact match
• Review broad match experiments — pause if CPA exceeds threshold
• Update negative keyword lists across all levels

Quarterly Checklist

• Full Search Terms Report audit (export all data, analyze trends)
• Evaluate match type progression readiness
• Restructure ad groups if match type data reveals new keyword themes
• Test new match type configurations in experiments (not in live campaigns)

---

Key Takeaways

Match Type Reality in 2026

1. Exact match is not exact. It matches close variants and meaning matches. It is still your tightest control.
2. Phrase match is the old BMM. It captures meaning in any order with additional context.
3. Broad match gives Google maximum flexibility. Only use it with Smart Bidding and 30-50+ conversions.
4. 20-40% of queries are invisible. The broader you go, the more queries you cannot see or audit.
5. Broad match CPCs rose 29% (2023-2025). Volume increases but so does cost-per-acquisition.

The Progression Path

Non-Negotiable Rules

• Never start with broad match. Build data with exact and phrase first.
• Never run broad match without Smart Bidding. It needs algorithmic bid adjustments.
• Never skip negative keywords. They are the difference between 4.6% and 13% conversion rates.
• Never accept Google's default recommendation to "upgrade" to broad match before you have sufficient conversion data. Google benefits from broader matching. You benefit from precision.

"Google's recommendations are sometimes in your best interest. They are always in Google's best interest." — Brad Geddes, industry veteran

---

Series Navigation

This article is part of the Google Ads Efficiency Playbook 2026 — a data-driven series helping service businesses capture demand without overpaying.

Related articles:

• The Optimal Google Ads Account Structure for Service Businesses
• Google Local Service Ads: The Pay-Per-Lead Alternative to Click-Based Ads
• Call Campaigns & Extensions: Why Phone Calls Convert 10x Better for Service Businesses

---

This guide draws on data from WordStream (16,000 campaigns), PPC Land (15,000 accounts), North Country Consulting ($20M query analysis), and Google's own documentation to present the reality of match types in 2026.