Google Ads Keyword Match Types Explained 2026

Google Ads keyword match types determine where every pound and dollar of your campaign budget actually goes. Imagine this scenario: you have set aside £3,000 per month for Google Ads. You have built your campaigns, written your ads, added your keywords. Two weeks in, you pull up the search terms report and discover something uncomfortable. You targeted “women’s running shoes” but your ads fired for queries like “shoe repair near me”, “sports shop job application” and “treadmill servicing”. Clicks rolled in, money was spent, conversions were nowhere.

This is not a rare edge case. It happens in the majority of Google Ads accounts we audit, whether the advertiser is spending $2,000 a month in the US or £10,000 a month in the UK. The root cause is almost always the same: keyword match types were either selected without thought or left on the default setting. When you add a keyword to Google Ads, the match type dictates which search queries will trigger your ad. Choose the wrong one and your budget evaporates on irrelevant clicks. Choose the right one and the same budget delivers significantly more conversions.

As of 2026, Google offers three keyword match types: broad match, phrase match and exact match. Alongside these three sits a fourth mechanism that works in the opposite direction: negative keywords. Each match type has its own notation, its own logic and its own best-use scenarios. Understanding when to deploy each one is not optional knowledge for anyone managing a Google Ads account. It is the difference between a profitable campaign and an expensive learning experience.

The Three Match Types: A Deep Dive

Every keyword you add to Google Ads carries an instruction about how loosely or tightly it should be matched against what people actually type into the search bar. The same keyword behaves completely differently depending on which match type you assign to it. Each match type also has a specific notation in the Google Ads interface, and knowing these symbols will save you time when reviewing account structures.

Broad Match

Notation: No symbols at all. The keyword is entered as plain text.

Broad match is the default setting in Google Ads. If you add a keyword without changing the match type, broad match is what you get. The system will show your ad for any search that it considers related to your keyword. The definition of “related” here is extremely generous.

Suppose you add “women’s running shoes” as a broad match keyword. Your ad could appear for searches including:

• women’s running shoes price (directly relevant)
• ladies trainers for jogging (synonym, different phrasing)
• best shoes for running 2026 (partial relevance)
• sportswear shop London (loose association)
• shoe cleaning kit (irrelevant, but Google found a connection)

The range is enormous. Those last two examples represent genuine budget waste because the search intent behind “sportswear shop London” or “shoe cleaning kit” is fundamentally different from someone looking to buy running shoes online.

Between 2024 and 2025, Google substantially upgraded the AI infrastructure behind broad match. In 2026, the algorithm no longer relies solely on keyword similarity. It factors in the user’s search history, their location, time of day and the broader context of their search session. The same broad match keyword can trigger different ads for different users. This makes broad match smarter than it used to be, but control still sits firmly with Google’s algorithm rather than with you.

Where broad match works well: Maximum reach. Discovery of search terms you did not anticipate but that convert. Entering new markets where you do not yet know which queries carry commercial value. Testing new product categories. Pairing with Smart Bidding strategies when you have sufficient conversion data (more on this later).

Where broad match causes problems: Without negative keywords, budget waste is almost guaranteed. Irrelevant clicks drag down your click-through rate (CTR), which in turn damages your Quality Score. Lower Quality Score means higher cost per click on the keywords that actually matter. It creates a negative spiral that is difficult to reverse once it takes hold.

Phrase Match

Notation: The keyword is wrapped in double quotation marks — “women’s running shoes”

Phrase match shows your ad for searches that preserve the meaning of your keyword. It is considerably narrower than broad match but more flexible than exact match. Back in 2021, Google redefined phrase match to focus on “searches that include the meaning of your keyword” rather than just preserving word order. Before that change, phrase match only triggered when the exact sequence of words appeared in the query. Now it matches on intent and meaning.

With “women’s running shoes” as a phrase match keyword:

• women’s running shoes price — meaning preserved, will match
• cheap women’s running shoes UK — meaning preserved with additions, will match
• best ladies running trainers — synonym, meaning preserved, will match
• women’s walking shoes — different activity, meaning shifts, will not match
• men’s running shoes — different gender, will not match
• treadmill prices — entirely different product, will not match

For most campaign types, phrase match is the most balanced option. It casts a wide enough net to avoid missing potential customers while keeping the scope tight enough to prevent serious budget waste. In our experience managing accounts across both UK and US markets, phrase match is where we typically anchor the core of a campaign. It is especially effective in service-based industries where intent clarity matters — searches like “accountant Manchester”, “personal injury solicitor London” or “HVAC repair Dallas” carry clear commercial intent that phrase match captures reliably.

Where phrase match works well: Intent preservation means the person searching is largely aligned with what you are offering. Budget efficiency is noticeably higher than broad match. Combined with negative keywords, phrase match creates a highly controlled structure that still has room to breathe. It works as the primary match type for most campaigns regardless of industry.

Where phrase match causes problems: Google’s interpretation of “meaning” does not always match yours. If you target “language course”, your ad might also show for “language school reviews”, which could be unwanted depending on your business model. Discovery capacity is lower than broad match, so you may miss emerging search trends or unconventional queries that still convert.

Exact Match

Notation: The keyword is enclosed in square brackets — [women’s running shoes]

Exact match is the tightest match type available. Your ad only shows for searches that carry the same meaning as your keyword. Note the emphasis on “same meaning” rather than “same words”. Even in exact match, Google captures close variants, typos, reordered words and synonyms. So [women’s running shoes] will also match queries like “running shoes women’s” or “ladies running shoes”. But the scope ends there.

With [women’s running shoes] as an exact match keyword:

• women’s running shoes — direct match, will trigger
• running shoes for women — word order changed, meaning identical, will trigger
• ladies running shoes — synonym, will trigger
• women’s running shoes price — additional word, may or may not trigger (Google decides)
• cheap running shoes women’s — different qualifier, likely will not trigger
• women’s sports shoes — different category, will not trigger

Exact match delivers the highest control. CPC tends to be lower because the ad-to-search alignment is very strong, which pushes Quality Score upward. But impression volume is limited. If you rely on exact match alone, you will miss a significant share of potential customers who phrase their searches differently from your keyword list.

Where exact match works well: Keywords that you already know convert, based on historical data. Brand terms. Campaigns with tight budgets where every penny counts. Long-tail keywords are particularly well suited to exact match because they already describe a specific search intent — there is less room for misinterpretation.

Where exact match causes problems: Low impression and click volume. You may need to add dozens of close variations manually to cover the full range of how people search. Zero discovery potential means you will never stumble on a profitable query you had not thought of.

Match Type Comparison Table

One point the table does not capture: these match types are not mutually exclusive. The strongest Google Ads accounts use all three simultaneously, in separate ad groups, with different bid strategies and negative keyword lists tailored to each. The table shows their individual characteristics, but the real power comes from combining them strategically.

Which Match Type to Use and When

Choosing the right match type is not a one-size-fits-all decision. Campaign objectives, budget size, industry dynamics and the maturity of your account all factor in. The scenarios below cover the most common situations we encounter when managing accounts for UK and US businesses.

Scenario 1: New Campaign, Limited Budget

Your monthly budget is between £1,000 and £2,500 (or $1,500-$3,000) and you are launching Google Ads for the first time. Every pound needs to work hard. Start with phrase match. Select 10-15 keywords, all phrase match. For the first two weeks, check your search terms report daily. Add irrelevant queries as negative keywords. Note which terms are actually generating enquiries or sales.

After two weeks, the data will guide you. If specific search queries are converting reliably, add those as exact match keywords in a separate ad group. This creates a layered structure where the same keyword exists as both phrase match and exact match. The exact match version typically delivers lower CPC and higher conversion rates because the ad-to-query alignment is tighter.

What you should avoid: starting with broad match on a limited budget. The learning period will consume a disproportionate share of your spend before you gather enough data to make informed decisions. With £1,500 a month, you cannot afford the experimentation that broad match demands.

Scenario 2: Mature Campaign, Growth Target

Your campaigns have been running for 3-6 months. Conversion tracking is properly configured. You know which keywords drive results. Now you want to scale. Keep your existing phrase and exact match keywords as they are. Alongside them, create a new ad group (or campaign) with the same keywords set to broad match. Enable Smart Bidding — Target CPA or Target ROAS — for this broad match campaign. Review the new search terms that broad match discovers on a weekly basis.

The critical point here: keep the broad match ad group separate. Do not mix broad match keywords with your phrase and exact match keywords in the same ad group. Google may prioritise the broad match keyword and you lose the control you have built up over months of optimisation.

Scenario 3: E-commerce, Large Product Catalogue

You sell hundreds or thousands of products. Adding individual keywords for every SKU is impractical. Use category-level phrase match keywords: “women’s running shoes”, “men’s casual shoes”, “children’s school shoes”. For specific, high-converting products, add exact match: [Nike Air Max 90 women’s]. For general discovery, test broad match in a separate ad group with a capped daily budget.

In e-commerce, the purchase journey involves multiple searches. Someone buying a sofa might search 10-15 times before converting — “corner sofa”, “L-shaped sofa grey”, “sofa deals UK”, “DFS vs IKEA sofa”. A broad match campaign captures these variations while your phrase and exact match campaigns handle the queries you already know work.

Scenario 4: Service Business, Location-Based

Law firms, dental practices, plumbing companies, estate agents — local service businesses where geography matters. Phrase match with location terms is your strongest option: “solicitor Manchester”, “dentist Kensington”, “emergency plumber Brooklyn”, “estate agent Bristol”. Add your brand name as exact match separately.

Broad match is risky in this sector. If you set “London solicitor” as broad match, expect your ads to appear for “solicitor salary UK”, “how to become a solicitor”, “solicitor job vacancies London” and “TV solicitor drama”. These are real examples pulled from search terms reports. Every one of those clicks costs you money and brings zero business value.

Scenario 5: B2B with High CPCs

Software companies, consultancies, financial services — sectors where a single click can cost £8-£25 ($10-$30). In these environments, budget discipline is not a preference; it is a necessity. Phrase match combined with an aggressive negative keyword list is the baseline. Exact match for proven converters. Broad match only if your monthly spend exceeds £5,000 ($6,000) and you have at least 30-50 monthly conversions to power Smart Bidding.

One £20 click on an irrelevant query in a B2B SaaS campaign might not sound like much. Multiply that by 15-20 irrelevant clicks per week and you are burning through £300-£400 a month with nothing to show for it. That amount could fund an entirely separate exact match campaign for a high-intent keyword group.

Transitioning from One Match Type to Another

If your current campaigns rely entirely on broad match and you are experiencing budget waste, do not make the switch overnight. A sudden change will cause impression volume to drop sharply and force Google’s algorithm into a re-learning phase. Follow this graduated approach instead:

1. Download the last 90 days of search terms data from your account.
2. Identify every query that generated a conversion.
3. Add those converting queries as phrase match keywords in a new ad group.
4. Take the top 5-10 performers and add them as exact match in another ad group.
5. Gradually reduce the budget on your broad match ad group while increasing it on the new groups.
6. Monitor performance over 2-3 weeks before making further adjustments.
7. Keep the broad match group running on a minimal budget for ongoing discovery rather than shutting it off entirely.

This graduated transition protects your impression share while shifting spend toward better-performing match types. We have run this exact process for accounts spending anywhere from £2,000 to £50,000 per month. The pattern holds regardless of budget size.

Negative Keyword Strategy

No matter which match type you choose, negative keywords are non-negotiable. If match types answer the question “which searches should trigger my ad?”, negative keywords answer the equally important question “which searches should definitely not trigger my ad?” The two systems work as a pair. Running campaigns without negative keywords is like fitting a door but leaving it permanently open.

Negative Keyword Match Types

Negative keywords have their own match types, and they behave differently from their positive counterparts. Understanding these differences is essential for plugging budget leaks.

Negative broad match: The default type. If all words in your negative keyword appear in the search query (in any order), the ad is suppressed. Adding “red shoes” as a negative broad match means “red running shoes” is blocked, but “red dress” and “blue shoes” are not blocked because only one of the two words appears in each query. All words must be present for the block to activate.

Negative phrase match: The search query is blocked if it contains your negative keyword in the same word order. Adding “red shoes” as negative phrase match blocks “red shoes prices” and “cheap red shoes UK” but does not block “shoes red colour” because the word order is reversed.

Negative exact match: Only blocks the exact query. Adding “red shoes” as negative exact match blocks only the search “red shoes”. It does not block “red shoes price” or “red shoes sale” because those queries contain additional words.

In practice, you will use negative broad match for the vast majority of your negative keywords. Negative phrase and exact match are reserved for situations where you need surgical precision — for example, blocking a specific query that is similar to a high-value keyword but carries the wrong intent.

Universal Negative Keywords Every Account Needs

Certain search patterns generate junk traffic in every industry. Adding these before your campaigns go live provides day-one budget protection.

Job seekers and career browsers:

• jobs, vacancy, vacancies, career, careers, hiring, salary, apprenticeship, internship, graduate scheme

Information seekers with no purchase intent:

• what is, how to, definition, meaning, wiki, wikipedia, tutorial, course, certification, free

File and download seekers:

• pdf, download, template, free download, torrent

Complaint and review traffic:

• complaint, complaints, review, reviews, trustpilot, glassdoor, reddit

These lists are starting points. Your search terms report will reveal industry-specific junk queries that need to be added over time. The list is never finished — it grows with every review cycle.

Industry-Specific Negative Keywords

E-commerce:

• second hand, used, pre-owned, refurbished (if you sell new products)
• rental, hire, lease (if you sell, not rent)
• repair, fix, service centre (if you are a retailer, not a repair shop)
• DIY, homemade, how to make

Professional services:

• Competitor brand names (unless you are deliberately targeting them)
• Irrelevant location names (if you serve London only, add Manchester, Birmingham, Leeds etc.)
• DIY, do it yourself, self-help (if you sell professional expertise)
• pro bono, free consultation, no win no fee (if these do not apply to your firm)

SaaS and technology:

• free, open source, crack, pirated, nulled
• Competitor names
• integration, API, developer documentation (if targeting end-users not developers)

Shared Negative Keyword Lists

Google Ads allows you to create negative keyword lists in the Shared Library and apply them across multiple campaigns. This is a significant time-saver for accounts with more than a handful of campaigns.

Recommended list structure:

• Universal Negatives: Applied to all campaigns. Contains the job-seeker, information-seeker and download-seeker terms listed above.
• Industry Negatives: Sector-specific terms. Your e-commerce list, your B2B list, your services list.
• Brand Protection List: Competitor brand names that you do not wish to target.

When you add a new term to a shared list, it automatically applies to every campaign that uses that list. For accounts running 5, 10 or 20+ campaigns, this eliminates the need to update each campaign individually. It is one of those small operational efficiencies that compounds over time.

Search Terms Report Analysis

The search terms report is the X-ray of your Google Ads account. It shows you exactly what users typed into Google before clicking your ad. You will find it in the Google Ads interface under Keywords > Search Terms. Every advertiser should be intimately familiar with this report because it is the primary feedback loop for match type decisions and negative keyword additions.

The report provides three critical insights:

1. Where your money is actually going. Which queries received clicks? How much did each click cost? Did any of those clicks lead to a conversion? This is where you discover that your “dental implants London” keyword triggered clicks for “dental nurse jobs London” at £6.50 per click with zero conversions.
2. Which queries actually convert. Your keyword might be “accountant Manchester” but the query that actually generates phone calls could be “small business tax accountant Salford”. These nuggets are hiding in your search terms report.
3. New keyword opportunities. Queries you never thought to target but that are converting at strong rates. These should be added as phrase or exact match keywords in their own right.

How to Work Through the Report Efficiently

When you open the search terms report, you might see hundreds or thousands of rows. Going through every single one is not practical. Use this filtering approach:

First: find the expensive non-converters. Set the date range to the last 30 days. Sort by cost, descending. Look at rows where the cost column is high but the conversions column shows zero. These queries are actively consuming your budget and returning nothing. Add them as negative keywords immediately.

Second: identify the converting queries. Sort by conversions, descending. Which searches are generating form submissions, phone calls or purchases? If any of these queries are not already in your keyword list, add them as phrase or exact match keywords. They have proven their value with real data.

Third: review CTR anomalies. If a search term receives a high number of impressions but the CTR is below 1%, the query and your ad are misaligned. Either adjust your ad copy to address that query better or add the term as a negative keyword. Low CTR on high-impression queries drags down your overall Quality Score, which raises your CPCs across the board.

How Often to Check

During the first month of a new campaign: daily. If you are running broad match, check every single day for the first week. You will catch irrelevant queries before they consume significant budget. For mature campaigns: weekly. Set aside 15-20 minutes every Monday or Friday to scan the past week’s search terms. For low-budget, exact match-heavy campaigns: fortnightly reviews are usually sufficient.

One important caveat: Google has been reducing the visibility of the search terms report since 2020. Low-volume queries are hidden under privacy justifications. The report typically covers 60-75% of all search queries that triggered your ads. The remaining 25-40% is invisible to you. This makes proactive negative keyword lists — the ones you build before launching campaigns — even more valuable. You cannot rely on the search terms report alone to catch every irrelevant query.

Smart Bidding + Broad Match: The 2026 Playbook

Google has been pushing the broad match + Smart Bidding combination since 2024, and by 2026 it is their official recommendation for accounts with sufficient conversion data. The logic is straightforward: broad match widens the net, Smart Bidding uses machine learning to decide how much to bid on each individual auction. The two together mean broad match no longer blindly spends on every query. Instead, the algorithm assigns higher bids to queries it predicts will convert and minimal bids (or no bid at all) to queries with low conversion probability.

The pitch sounds compelling. But this combination has prerequisites that many advertisers overlook.

Prerequisites

Sufficient conversion data. Smart Bidding strategies — Target CPA, Target ROAS, Maximise Conversions — depend on conversion tracking data. Google recommends a minimum of 30 conversions in the last 30 days, ideally 50 or more. Below that threshold, the algorithm lacks the data density to make reliable predictions. Performance becomes erratic.

Correct conversion definitions. If your conversion action is defined as “page view” or “add to basket” rather than an actual purchase or lead form submission, Smart Bidding will optimise for those micro-actions instead of genuine business outcomes. Make sure your primary conversion action reflects what you actually consider a successful outcome — a completed sale, a submitted contact form, a phone call that lasts more than 60 seconds.

Adequate budget. The broad match + Smart Bidding combination goes through a learning period of 2-3 weeks. During this phase, performance fluctuates. CPA may temporarily spike. Your budget needs to absorb this learning phase without forcing you to make panicked changes. At £500-£1,000 per month ($600-$1,200), this strategy is too risky. At £3,000+ ($3,500+), it becomes viable. The larger the budget, the faster the algorithm learns.

Implementation Steps

Do not modify your existing phrase match campaigns. Create a separate campaign or ad group. Add the same keywords with broad match. Select Target CPA as your bidding strategy and set the CPA target at 10-20% above your current campaign’s average CPA. This gives the algorithm room to experiment during the learning phase.

For the first two weeks, keep your expectations low. The algorithm is gathering data. From week three, results start stabilising. After 4-6 weeks, you can make a clear comparison between your phrase match campaigns and your broad match + Smart Bidding campaigns. If the combination outperforms, shift budget gradually toward it. If it underperforms, pause it and refocus on phrase match.

One approach that works well for UK and US accounts spending £5,000-£20,000 per month: run both structures simultaneously with a 70/30 budget split. Seventy percent on your established phrase and exact match campaigns, thirty percent on the broad match + Smart Bidding experiment. Adjust the split based on results over the following months.

When This Combination Does Not Work

Not every account benefits from broad match + Smart Bidding. If your monthly conversion volume is below 15-20, the algorithm simply does not have enough data points. Niche industries with very specific audiences also struggle because the broad net catches too many irrelevant queries even with Smart Bidding’s filtering. Accounts with very low budgets cannot sustain the learning period without significant waste.

On top of that, your negative keyword list for the broad match group needs to be substantially more aggressive. Where 50 negative keywords might suffice for a phrase match campaign, the same account running broad match may need 200-300 negatives. The wider the net, the more exclusions you need to keep the catch relevant.

Campaign Structure Examples

Theory becomes practical when you see how match types map onto real business types. The examples below are drawn from common UK and US business scenarios. Adapt them to your specific market and offering.

Example 1: Dental Practice in London

Ad Group 1 — Phrase Match (Core Services):

• “dental implants London”
• “teeth whitening near me”
• “Invisalign London”
• “emergency dentist Kensington”
• “porcelain veneers price”

Ad Group 2 — Exact Match (High Converters):

• [dental implants cost London]
• [Invisalign price London]
• [emergency dentist near me]

Negative keywords: NHS, free, jobs, salary, apprenticeship, course, university, how to become, DIY, home whitening, Manchester, Birmingham, Leeds, Edinburgh (if serving London only)

Dental is a high-CPC sector in the UK. Clicks on “dental implants” related keywords routinely cost £5-£15. In the US, “dental implants” CPCs can reach $15-$40 in competitive metro areas. At those prices, every irrelevant click represents a meaningful cost. Exact match on proven terms significantly reduces waste.

Example 2: UK-Wide E-commerce (Furniture)

Campaign 1 — Category Campaign (Phrase Match):

• “corner sofa”
• “oak dining table”
• “king size bed frame”
• “TV unit modern”
• “velvet armchair”

Campaign 2 — Product Campaign (Exact Match):

• [grey corner sofa with storage]
• [6 seater oak dining table]
• [ottoman bed frame king size]

Campaign 3 — Discovery (Broad Match + Smart Bidding):

• living room furniture
• home furnishings
• contemporary furniture UK

Negative keywords: second hand, used, charity shop, gumtree, freecycle, DIY, how to build, assembly instructions, IKEA, Argos (if not competing on price with these retailers)

In the furniture sector, purchase journeys are long. A customer buying a sofa typically conducts 10-15 searches over days or weeks. Broad match captures the early-stage exploration queries while phrase and exact match handle the high-intent, decision-stage searches. The three-tier structure ensures coverage across the entire purchase funnel.

Example 3: US-Based SaaS Company

Ad Group 1 — Phrase Match:

• “project management software”
• “team collaboration tool”
• “task management app”
• “remote work software”

Ad Group 2 — Exact Match:

• [project management software for small business]
• [best team collaboration tool 2026]
• [task management app pricing]

Negative keywords: free, open source, crack, pirated, GitHub, tutorial, course, certification, Asana, Monday.com, Trello, Jira (competitor names, unless deliberately targeting), student, personal use

B2B SaaS campaigns in the US regularly see CPCs of $8-$25 for competitive terms. “Project management software” can cost $15-$20 per click. At that rate, ten irrelevant clicks per week equals $150-$200 lost monthly. A tight phrase match structure with strong negatives prevents this from happening.

Example 4: Legal Services in the US

Ad Group 1 — Phrase Match:

• “personal injury lawyer Houston”
• “car accident attorney Texas”
• “slip and fall lawyer near me”

Ad Group 2 — Exact Match:

• [personal injury lawyer Houston TX]
• [car accident attorney near me]

Negative keywords: salary, jobs, how to become, law school, pro bono, free consultation (if not offered), malpractice (if not practised), Dallas, Austin, San Antonio (if serving Houston only)

Legal keywords in the US carry some of the highest CPCs on the entire Google Ads platform. “Personal injury lawyer” can exceed $100 per click in competitive markets. At that price point, a single wasted click is painful. Exact match combined with exhaustive negative lists is not a luxury — it is a survival mechanism.

Budget Allocation Across Match Types

When running multiple match types simultaneously, how you split the budget matters. These are starting points, not fixed rules. Adjust based on your account’s performance data.

The increase in broad match allocation for advanced campaigns is contingent on having Smart Bidding powered by sufficient conversion data. Without that data, keep broad match allocation low. A 10% discovery budget on broad match, reviewed weekly, is far safer than committing 40% and hoping the algorithm sorts it out.

For more detail on campaign setup and structure, our guide to running Google Ads covers the full process. For current pricing benchmarks across UK and US markets, see our Google Ads pricing guide for 2026.

Frequently Asked Questions