Only API keys on our Conversations platform are eligible to use this API version. Refer to the Platforms section of our Platform & API Concepts documentation to learn which platform your API keys are on.

The Conversations API lets you programmatically retrieve and submit Bazaarvoice Conversations data for use in your applications. To learn more, go to the Conversations API documentation home page.

Contents

(+ show- hide)

Returns Reviews and Rating and related data.

Examples

The following examples demonstrate common Conversations API display requests. For a complete overview of using the Conversations API to display data, refer to the Display tutorials.

Demonstration purposes only. Do not reuse the API passkeys below in your application.

Requesting all reviews for a particular product

This example call shows how to use multiple filters. This call only returns the reviews that have comments on them. The reviews are sorted in descending order by Rating.

https://stg.api.bazaarvoice.com/data/reviews.json?apiversion=5.4&passkey=caB45h2jBqXFw1OE043qoMBD1gJC8EwFNCjktzgwncXY4&Filter=ProductId:data-gen-moppq9ekthfzbc6qff3bqokie&Sort=Rating:desc&Limit=10

Requesting all reviews for a particular product with review statistics (inc. average rating)

This example call shows how to use the include and stats parameters to retrieve product information and review statistics (including average rating). Note: Stats=Reviews must be used in conjunction with Include=Products.

https://stg.api.bazaarvoice.com/data/reviews.json?apiversion=5.4&passkey=caB45h2jBqXFw1OE043qoMBD1gJC8EwFNCjktzgwncXY4&Filter=ProductId:data-gen-moppq9ekthfzbc6qff3bqokie&Include=Products&Stats=Reviews

Requesting a single review by review ID for a particular product

You can request multiple reviews by separating the review IDs with a comma - e.g. filter=id:83,84,85

Requesting a single review by review ID including product information for a particular product

You can request multiple pieces of information by separating them with a comma - e.g. Include=products,categories,authors

Requesting 20 reviews, sorted by rating in ascending order for a particular product

You can sort by multiple attributes by separating them with a comma - e.g. Sort=Rating:desc,SubmissionTime:asc

Requesting reviews using a filter and returning statistics for the filtered results for a particular product

You can get statistics specific to the filter applied to a request using the "&FilteredStats" parameter

Requesting all native (non-syndicated) reviews

Native (non-syndicated) content only. Continue reading for more details.

Requests for this content type that do not include a ProductId, AuthorId, CategoryAncestorId, SubmissionId, or Review Id filter will only work with API keys that are configured to exclude syndicated content.

We recommend using our Displayable Content Export service to download all of your native and syndicated content. If you still prefer the Conversations API, then you may contact our support team to have your Conversations API keys re-configured.

Requesting reviews for a product that is associated with a product family

The product families feature allows you to share reviews among related products. Learn more about product families at the Bazaarvoice Knowledgebase.

The parameter “include=products” will result in the API returning product information in the response. This information can be used to identify about which product a review was originally written. Use the ‘ProductId’ field on each review to look up the relevant product information in the “Includes.Products” object.

You will notice that several reviews in the response are associated with "Product4" despite using a filter in Request for ProductId:data-gen-a718alq7rap86nqc7v05h304. That’s because these products are in a product family. The ProductId:data-gen-a718alq7rap86nqc7v05h304 and Product4 are associated with a product family and therefore returning the reviews associated with both the products.

Parameters

Name Description Required Default Value
ApiVersion The API version, e.g. 5.4. Yes
[FORMAT] Response format (json) Yes
PassKey API key is required to authenticate API user and check permission to access particular client's data. Yes
Attributes Attributes to be included when returning content. For example, if includes are requested along with the &attributes=ModeratorCodes parameter, both the includes and the results will contain moderator codes. In order to filter by ModeratorCode, you must request the ModeratorCodes attribute parameter. No
Callback Callback function name used with JSONP. Value is a string consisting of the following characters: a-z,A-Z,0-9,_ (excluding comma). See the JSONP tutorial for more information. No
ExcludeFamily Boolean flag indicating whether to exclude content from other products in the same family as the requested product. For example, "&filter=productid:eq:1101&excludeFamily=true" limits returned content to just that of product 1101 and not any of the products in the same family. If a value is not defined, content on all products in the family will be returned. No false
Filter Filter criteria for primary content of the query. Multiple filter criteria are supported. No
Filter_[TYPE] Filtering option for included nested content. TYPE can be any included nested content. i.e. Comments for Reviews. No
FilteredStats Statistics based on the filters used to perform queries. The filtered statistics are returned in their own block and can be returned any place that a regular statistics block can be currently returned. Available content types are: Reviews, Questions, Answers. Note: FilteredStats must be used in conjunction with "Include=Products". The filtered stats block will be identical to a regular stats block if a filtered stat is requested but a filter has not been specified for the particular type of stat being requested. No
IncentivizedStats If set to true, displays the number of incentivized reviews for each product or author returned in the response within an IncentivizedReviewCount element. This parameter must be used in conjunction with the Include parameter to include authors or products in the response, and also with the FilteredStats or Stats parameters to include statistics for reviews in the response. No false
Include Related subjects to be included (e.g. Products, Categories, Authors, or Comments). No
Limit Max number of records returned. An error is returned if the value passed exceeds 100. No 10
Limit_[TYPE] Limit option for the nested content type returned. TYPE can be any nested content. i.e. Comments for Reviews. An error is returned if the value passed exceeds 20. No 10
Locale Locale to display Labels, Configuration, Product Attributes and Category Attributes in. The default value is the locale defined in the display associated with the API key. No
Offset Index at which to return results. By default, indexing begins at 0 when you issue a query. Using Limit=100, Offset=0 returns results 0-99. When changing this to Offset=1, results 1-100 are returned. The maximum supported value is 300000. No 0
Search Full-text search string used to find UGC. For more information about what fields are searched by default, see the Display Fundamentals page.
Results can not be sorted when using Search.
Search is only available to requests filtered by one product ID.
No
Sort Sort criteria for primary content type of the query. Sort order is required (asc or desc). Multi-attribute sorting for each content/subject type is supported. No
Sort_[TYPE] Sorting option for nested content. Sort order is required (asc or desc). TYPE can be any nested content. i.e. Comments for Reviews. No
Stats The type of statistics that will be calculated on included subjects. Available content types are: Reviews, Questions, Answers. Note: Not all statistical content types apply to every possible include. No

Filter options

  • Each filter argument specifies the attribute to filter on followed by a comma-separated list of values. For instance, "Filter=Rating:eq:3,4" will match all reviews with rating values 3 OR 4.
  • Advanced operators can be used to define filters. For instance, "Filter=Rating:lt:3" will match all reviews with a rating value less than 3. All advanced operators are documented on the Conversations API page.
  • If a filter value contains a comma or a colon, that character needs to be escaped with a backslash (\, or \:). If a filter value contains an ampersand (&), the ampersand must be encoded in the filter value by replacing & with %26.
  • Multiple filters are allowed as URL parameters in which case filters are AND'ed. For instance, "Filter=Rating:gte:4&Filter=IsFeatured:eq:true" will match all featured reviews with a rating of 4-star or higher.
  • Time-based filters can be used for SubmissionTime, LastModificationTime and LastModeratedTime. Dates in time-based filters are calculated as the number of seconds since January 1, 1970, 00:00:00 UTC. In a future version, we will be adding support for comparing date/time string values. The following example returns content that was submitted on November 9, 2009: &filter=SubmissionTime:gt:1257746400&filter=SubmissionTime:lt:1257832800

The following table lists the attributes available for filtering.

Name Description
Id
The identifier of the Review.
AdditionalField_[FIELD_NAME]
Additional field to filter by, e.g., filter=AdditionalField_[FIELD_NAME]:eq:[FIELD_VALUE]
AuthorId
The identifier of the author who wrote the content
CampaignId
The identifier of the Campaign associated with the content
CategoryAncestorId
The identifier of the Product Category ancestor of the Product that the Review was written on. Syndicated content will not be returned when using this filter.
ContentLocale
Locale of the content to display. If this filter is not defined, all content regardless of its locale is returned. To return specific content by locale, define the value in the filter. A wildcard character “*” can be used to define the value, e.g., “en*” returns all content in English (en_US, en_CA, en_GB, etc.) or you can use a single ContentLocale code (e.g., “fr_FR”). ContentLocale codes are case-sensitive.
ContextDataValue_[DIMENSION_EXTERNAL_ID]
The context data value for the content. DIMENSION_EXTERNAL_ID can be age, gender, etc. e.g. filter=contextdatavalue_age:under21
HasMedia
Boolean flag indicating whether content has either photos or videos
HasPhotos
Boolean flag indicating whether content has photos
HasTags
Boolean flag indicating whether content has tags
HasVideos
Boolean flag indicating whether content has videos. For more information on inserting the returned VideoUrl into HTML, see the Conversations API page.
Helpfulness
Helpfulness represents how helpful a review has been to consumers. The Helpfulness value takes into account several factors including feedback votes and how many approved reviews an author has written. It will be a positive decimal value between 0 - 1.0, or null if not enough data exists to calculate it.
IsFeatured
Boolean flag indicating whether content is featured
IsRatingsOnly
Boolean flag indicating whether the review was a ratings only review
IsRecommended
Boolean flag indicating whether the user would recommend this product
IsSubjectActive
Boolean flag indicating whether the content subject is active
IsSyndicated
Boolean flag that allows you to select whether to return syndicated or non-syndicated reviews. If IsSyndicated:eq:true, the response only returns syndicated content and a SyndicationSource block with the details of where the syndication is coming from. If IsSyndicated:eq:false, the response only return native (non-syndicated) content. Note: The API key must be configured to show syndicated content.
LastModeratedTime
The date/time of the latest moderation of the content. See the Introduction for an example of using advanced operators for filtering.
LastModificationTime
The date/time of the latest modification of the content. See the Introduction for an example of using advanced operators for filtering.
ModeratorCode
String value indicating the moderator code for content that was at one time rejected e.g., &Filter=ModeratorCode:eq:CR returns UGC that contains the CR (Competitor Reference) code. Multiple codes can be entered in a comma-delimited list, e.g., &Filter=ModeratorCode:eq:CS,IU returns UGC with either the CS (Customer Service Complaint) or the IU (Inappropriate/Unusable Content) code. For a list of all Moderator Codes, see the Conversations API page. Note that the ModeratorCodes attribute parameter should be included in the request. This will ensure the moderated code(s) for each request will be included in the response. If the ModeratorCodes attribute is missing the filter will still perform, but there is not indication of which elements have been tagged with which codes. See the Parameters section above. A more detailed discussion on how moderation works can be found here.
ProductId
The identifier of the Product that the Review was written on.
Rating
The Review Rating, usually between 1 to 5.
SecondaryRating_[RATING_NAME]
Secondary rating value to filter by, e.g., filter=SecondaryRating_Durability:gte:1. Note: All advanced operators can be used for secondary ratings comparisons.
SubmissionId
Submission identifier assigned to the content when it was initially submitted
SubmissionTime
The submission date/time of the content. See the Introduction for an example of using advanced operators for filtering.
Tag_[TAG_NAME]
The tag name to filter by, e.g., filter=tag_[TAG_NAME]:eq:[TAG_VALUE]
TotalCommentCount
The number of comments the Review has
TotalFeedbackCount
Number of feedbacks received
TotalNegativeFeedbackCount
Number of negative feedbacks received
TotalPositiveFeedbackCount
Number of positive feedbacks received
UserLocation
Location of the author

Sort options

  • Sort order is required (asc or desc). There is no default.
  • Multi-attribute sorting is supported by using a comma separated list of sort criteria for a content/subject type. e.g. Sort=Rating:asc,SubmissionTime:desc

The following table lists the attributes that are available for sorting.

Name Description
Id
The identifier of the content/subject type
AdditionalField_[FIELD_NAME]
Additional field to sort by, e.g., sort=AdditionalField_[FIELD_NAME]:desc
AuthorId
The Identifier of the author who wrote the content
CampaignId
The identifier of the Campaign associated with the content
ContentLocale
Locale value of the content.
Besides asc or desc, you can also specify the order of locales with ContentLocale. You can only specify upto 5 locales. Multi-attribute sorting is not supported with this customized order. For more information, please refer to the internationalization page.
ContextDataValue_[DIMENSION_EXTERNAL_ID]
The context data value for the content. DIMENSION_EXTERNAL_ID can be age, gender, etc. e.g. sort=contextdatavalue_age:desc
HasPhotos
Boolean flag indicating whether content has photos
HasTags
Boolean flag indicating whether content has tags
HasVideos
Boolean flag indicating whether content has videos. For more information on inserting the returned VideoUrl into HTML, see the Conversations API page.
Helpfulness
Helpfulness represents how helpful a review has been to consumers. The Helpfulness value takes into account several factors including feedback votes and how many approved reviews an author has written. It will be a positive decimal value between 0 - 1.0, or null if not enough data exists to calculate it.
IsFeatured
Boolean flag indicating whether content is featured
IsRatingsOnly
Boolean flag indicating whether the review was a ratings only review
IsRecommended
Boolean flag indicating whether the user would recommend this product.
IsSubjectActive
Boolean flag indicating whether the content subject is active
IsSyndicated
Boolean flag indicating whether the content has been syndicated.
LastModeratedTime
The date/time of the latest moderation of the content
LastModificationTime
The date/time of the latest modification of the content
ProductId
The identifier of the product
Rating
The Review Rating, usually between 1 to 5
SecondaryRating_[RATING_NAME]
Secondary rating value to sort by, e.g., sort=SecondaryRating_Durability:desc
SubmissionId
Submission identifier assigned to the content when it was initially submitted
SubmissionTime
The submission date/time of the content
TotalCommentCount
Number of comments associated with the content
TotalFeedbackCount
Number of feedbacks received
TotalNegativeFeedbackCount
Number of negative feedbacks received
TotalPositiveFeedbackCount
Number of positive feedbacks received
UserLocation
Location of the author

Response format

This is a sample response for requesting reviews. Use the links above to see live examples.

{
    "Includes": { },
    "HasErrors": false,
    "Offset": 0,
    "TotalResults": 224775,
    "Locale": "en_US",
    "Errors": [ ],
    "Results": [
        {...}
        {...}
    ],
    "Limit": 2

}

Response elements

Name Description
Includes
Included results related to primary results. For example, with the request reviews.json?include=Comments, the Include node will be populated with comments associated with the reviews in the Results node.
HasErrors
Boolean value indicated if one of more errors have occurred. See Errors below.
Offset
Dataset offset used for pagination (passed as URL parameter in a query request). The maximum supported value is 300000.
TotalResults
Total number of records matched.
Locale
Indicates the language by region (language_region) associated with the client instance. This is the value used for field labels returned by the API, not the region from which the reviews originate.
Errors
Error section is populated instead of other fields if there is an error with a query syntax or problem executing a query.
Results
Section containing an array of primitive type object references matched by a query.
Limit
The total number of results returned, specified by user in the URL. Defaults to 10 and has a maximum of 100.

Error codes

Value Description
ERROR_ACCESS_DENIED

Insufficient privileges to perform the operation

ERROR_PARAM_INVALID_API_KEY

Invalid API Key value

ERROR_PARAM_INVALID_CALLBACK

Invalid JsonP callback function name

ERROR_PARAM_INVALID_FILTER_ATTRIBUTE

Invalid filter attribute name

ERROR_PARAM_INVALID_INCLUDED

Invalid parameter value

ERROR_PARAM_INVALID_LIMIT

Invalid limit value

ERROR_PARAM_INVALID_LOCALE

Invalid locale code

ERROR_PARAM_INVALID_OFFSET

Invalid offset value

ERROR_PARAM_INVALID_SORT_ATTRIBUTE

Invalid sort attribute name

ERROR_REQUEST_LIMIT_REACHED

Rate limiting error, i.e. too many requests per time interval

ERROR_UNKNOWN

Unknown error (internal server error, for instance)