We’ve migrated our documentation to a new site, which means some URLs have changed.
Audience

Match Types

Description

Keyword match types allow you to control how relevant and accurate your purchased keywords are to the user's query. The rules are essentially the same as Google's match types with the key difference that broad match in Cxense Advertising requires all words to be present in the query string (i.e. an AND match instead of Google's OR). This dramatically improves relevance and is essentially the same as placing the broad match modifier in front of each word.

Match types are defined around criteria that describe how a user's query relates to purchased keywords. A match type is defined by having one or more of the following properties:

  • Additional words in query? Whether the query can contain additional words and still match the advertisement.

  • Variant normalisation? Operations such as removing diacritics from characters ("café" to "cafe") or transforming to the base form ("fuß" to "fuss")

  • Lemmatisation? Reduces the word to its lemma or canonical form (e.g. "boxing" to "box").

  • Synonyms? Whether matches against synonyms are accepted.

  • Order sensitive? Whether words must be in the same order in both the query and purchased keyword.

  • Contiguous? Whether the words must be next to each other in one "run" or "phrase" or whether other words can be found in the middle of the phrase.

The table below shows the various transformations and allowances made for each match type.

Match Type

Additional words in query?

Variant normalisation?

Lemmatisation?

Synonyms?

Order sensitive?

Contiguous?

Exact

No

Yes

No

No

Yes

Yes

Phrase

Yes

Yes

No

No

Yes

Yes

Broad

Yes

Yes

Yes

Yes

No

No

Note: The Negative match type can be used as modifier to any of the Match Types above. The negative keywords behave like a phrase, so a negative keyword consisting of multiple words will only match if all terms in the negative phrase occur in the specified order.

Inline match types

When entering keywords into the user interface or uploading via an API or CSV you can prefix and suffix keywords with characters that correspond to each match type:

Match Type

Prefix

Suffix

Exact

[

]

Phrase

"

"

Broad

 

 

Negative

-

 

Examples

The table below presents various combinations of queries related to a user's interest in car insurance. Each Match Type column below identifies if there would be a match for a keyword expression of car insurance defined for an advertisement and explains why the advertisement would or would not match.

Query

Exact Match: [car insurance]

Phrase Match: "car insurance"

Broad Match: car insurance

Negative Match: "car insurance" -house

car insurance

Matches.
The query contains the same terms in the same order with no additional terms.

Matches.
The query contains the same terms in the same order.

Matches.
The query contains all purchased terms.

Matches.
The query contains all terms and does not match the "-house" clause.

cheap car insurance

Does not match.
The query contains the additional term "cheap" which is not permitted in exact match.

Matches.
Phrase match allows additional query terms to be present.

Matches.
The query contains all purchased terms.

Matches.
The query contains all terms and does not match the "-house" clause.

car and house insurance

Does not match.
The query contains the additional terms "and" and "house" which is not permitted in exact match.

Does not match.
The term "car" is not contiguous with "insurance".

Matches.
The query contains all purchased terms.

Does not match.
The term "house" is present in the query which matches the negative term "house".

car seat

Does not match.
The query does not contain the term "insurance".

Does not match.
The query does not contain the term "insurance".

Does not match.
The query does not contain the term "insurance".

Does not match.
The query does not contain the term "insurance".

insurance for cars

Does not match.
The query does not contain the term "car". Note that exact match does not perform lemmatisation.

Does not match.
The query does not contain the term "car". Note that phrase match does not perform lemmatisation.

Matches.
The term "cars" is lemmatised to "car" and thus matches. Note that broad match is not sensitive to word order.

Does not match.
The query does not contain the term "car". Note that phrase match (the "inclusive" clause) does not perform lemmatisation.

insurance for my car

Does not match.
The query contains additional terms.

Does not match.
The terms are not contiguous and are out of order.

Matches.
Broad match accepts additional terms and is insensitive to word order.

Does not match.
The inclusive clause is based on a phrase match and the query terms are not in the correct order and are not contiguous.

Last updated: