Skip to content
All articles

Google Ads keyword match types: reach, control and search-term review

Laptop displaying a browser address bar above the Google homepage
Image by 377053 on Pixabay

Google Ads match types control how a keyword relates to a search before an advertisement can enter an auction. Choose a setting by the meaning you intend to reach and the searches you can responsibly assess. Exact match still covers meaning and intent; negative keywords use different rules from positive keywords.

A useful review therefore has two sides: searches you want to remain eligible and searches you intend to exclude. Write both before changing settings. This article provides a worked matching exercise rather than a recommendation to widen every campaign. For the wider campaign brief, start with Google Ads and a destination that accurately explains the offer.

What do broad, phrase and exact match actually mean?

Broad match can include searches related to a keyword without its direct meaning. Google’s matching documentation describes a wider relationship than simply containing the words you typed. A broad keyword should therefore be reviewed against the business task, landing page and actual searches, rather than a word-for-word expectation.

Phrase match can include implied meaning and more specific searches. Exact match can match the same meaning or intent. These definitions explain why quotation marks and brackets do not turn positive keywords into literal text filters. A customer can use different wording while expressing a relevant request.

Phrase match uses quotation marks and exact match uses square brackets. Record the chosen setting explicitly in a review sheet instead of relying on punctuation copied from another document. An administrator should be able to distinguish the keyword’s saved setting from an example query discussed in a meeting.

Google identifies Smart Bidding as critical when using broad match. Treat that documented dependency as part of the decision, alongside the meaning of the keyword. Do not widen a campaign merely because broad match contains the word broad. Agree the conversion definition and review responsibilities before changing the way searches can enter the campaign.

Why are negative match types a separate decision?

Negative keywords do not expand to close variants. Google’s negative-keyword guidance distinguishes them from positive matching. This means a reviewer cannot reverse the positive exact-match definition and assume the negative equivalent removes every expression of that meaning. Inspect the actual words you want excluded.

Negative broad match blocks searches containing all its terms in any order. Negative phrase match requires the negative terms in the same order. Negative exact match requires the terms in order without extra words. Those Search campaign rules make exclusion scope a concrete choice. They should not be transferred unchanged to Display or Video planning.

A negative term can also be too broad for the business reason behind it. Suppose a service does not offer classroom training but does offer staff onboarding within an installation project. An indiscriminate exclusion of every search containing training would deserve a careful review. Define the unwanted task before selecting the negative wording.

Negative keywords automatically account for casing and misspellings. The negative-keyword reference distinguishes that treatment from its limitation on close variants such as singular and plural forms. Keep this exception in the review notes so the team does not create unnecessary duplicate entries while missing a genuinely different term.

Original positive-and-negative matching laboratory

Use the following hypothetical example in a review document: a business sells running shoes but does not want searches for used running shoes. The negative phrases below illustrate Google’s documented Search matching rules. “Not blocked by this negative” means only that this particular exclusion does not block the query; it does not predict an impression or establish eligibility under other settings.

Proposed negative Example query Result under this negative alone Review implication
Broad: used shoes used running shoes Blocked: both negative terms appear Check whether this breadth matches the unwanted task
Broad: used shoes shoes used for running Blocked: order does not matter Consider whether the query is actually unwanted
Phrase: “used shoes” buy used shoes nearby Blocked: the phrase remains in order Additional words do not remove the exclusion
Phrase: “used shoes” used running shoes Not blocked: phrase is interrupted The intended exclusion may need different wording
Exact: [used shoes] used shoes Blocked: no extra words A narrowly defined exclusion
Exact: [used shoes] buy used shoes Not blocked: extra word present Do not mistake exact negative for a meaning-wide ban

The second row is deliberately uncomfortable. “Shoes used for running” can describe new products and a use case. A rule that removes it may be mechanically correct but commercially wrong. This is why the reviewer needs both the matching explanation and a statement of the customer request the business wants to reject.

Add a singular version to the exercise, such as used shoe, and ask the reviewer to explain the negative-variant limitation before approving a list. Keep spelling, word order and meaning as separate columns. The objective is to catch misunderstandings in a document before they become settings applied across relevant campaigns.

Now prepare the positive side with a proposed keyword such as running shoes. For each match type, record the intended meaning and several desired and undesired searches. Treat these as expectations to investigate against Google’s meaning-based definitions, rather than a promise of exactly which auction the system will enter.

How should you review search terms after a change?

Create a change note containing the saved keyword, previous setting, proposed setting, business reason and intended conversion. Attach the matching exercise. Record who can approve the change and who will inspect its consequences. This makes a later reversal intelligible without requiring someone to reconstruct the reasoning from a screenshot.

Read unexpected searches in context. Ask whether the query expresses the intended task, whether the advertisement answers it and whether the destination supports the promise. A word that looks unfamiliar can still describe a relevant request. A familiar word can appear in an unrelated question. Keep the interpretation beside the evidence instead of treating keyword overlap as the decision.

Use separate actions for separate problems. A relevant but poorly answered query may require page content. An irrelevant task may require an exclusion. An overly broad service group may require a clearer intent brief. A conversion definition that does not represent the desired outcome requires measurement work. Changing the match type should not conceal those distinctions.

Set review timing around the campaign’s actual decision needs and available evidence; this exercise does not invent a universal number of clicks or days. Preserve uncertainty where evidence is insufficient. If somebody proposes a performance conclusion, ask what observation supports it and whether other settings changed at the same time.

What should the approval record contain?

Keep the final record concise: the offer, desired searches, unwanted searches, positive settings, proposed negatives and their scope. Include a counterexample showing that a wanted request remains outside the intended exclusion. A reviewer should be able to explain the outcome of each worked row using the documented rule.

Before activation, inspect the destination on the devices your intended customers use and confirm that the proposed promise is supported. Review relevant website development and SEO work separately from keyword settings. If you want to discuss the matching exercise and measurement requirements, get started with the change record. Keep approval of the worksheet distinct from permission to modify a live advertising account.

By Spout Team. AI-assisted drafting by Codex using the linked primary documentation.

Make your next move.

A website, a store or a clearer digital plan. Let’s start with your business.

Start your brief