Product display page v2

No. Bazaarvoice does not record a rating until the user completes all required fields in the progressive submission form and select Submit.

No, the review summary section stays left-aligned.

No. Bazaarvoice doesn't have plans to implement a color gradient for the rating star icons in the new PDP review display at this time. Instead, we provide helper text to explain the meaning of the stars. For example, selecting fifth star shows 5 out of 5 stars selected, product is Excellent right below. Generally, having this text is a more accessible solution than using a red/yellow/green gradient. It also helps maintain the brand colors and user experience.

Yes. You can move questions into the first step of the progressive submission form by marking them as required. Otherwise, the form is designed to capture the core parts of the review in the first step and allows consumers to provide additional details in subsequent steps.

Yes. For details, refer to Enable V2 Ratings & Reviews and single product submission.

We won’t automatically enable the new display for clients. Instead, we want to offer them the chance to test the experience before deployment.

No. We will be replacing the Write a Review button with star rating buttons, as they drive more consumer engagement and result in more reviews collected. You can manage the submission of the form through the config hub.

If you choose to disable the submission, you have two options, both of which require additional effort:

• Create a custom button: You can generate an ssURL and transform it into a button on your website. This button can be product-specific and positioned near the review display on the product page.
• Use a hybrid approach: If you are an Enterprise client, you can use API submission to create the button or form for onsite reviews while continuing to use Bazaarvoice's hosted solution's display JavaScript. 

However, these options are not recommended. Our in-page submission feature typically doubles the number of organic reviews collected from product pages by directly engaging users and collecting star ratings from the PDPs, thus eliminating additional steps for consumers.

Yes. For those not using multi-product submission and directing consumers to a product or container page using an ssURL, the new V2 submission form will be displayed, provided they have upgraded their product page reviews display. If you use ssURLs, ensure that Reviews V1 remains ON in the Style Editor.

Yes. The new photo-first reviews display is designed to provide an accessible shopping experience by meeting a range of accessibility guidelines, including levels A and AA, for various aspects of the ratings and reviews experience.

Yes. Galleries are typically placed above the reviews container, creating a visual separation from the visual reviews carousel. This separation is maintained by elements such as the ratings histogram, summary information, and star buttons, which usually provide enough space between the gallery and the carousel. 

Yes. The feature provides 4 or 5 AI-generated topic suggestions based on the product’s name, description, and category.

No. These suggested topics provide guidance to help consumers improve the quality of their reviews. Including them in the reviews is optional, not mandatory..

No. It is turned off by default.

No. They are based on the product name, product description, and product category..

No. Content Coach currently supports only English-specific locales.

Yes. You must have a minimum of 30 approved reviews combined across English, French, German, and Spanish, regardless of the source (native, family, or syndicated), to generate the review summary and ensure a broad range of consumer feedback.

Yes. Review summaries is available on the Top consumer sentiment on my site dashboard. To access it, ensure the Review Summary flag is enabled in Configuration and then navigate to: Portal > Analytics > Ratings & Reviews > Consumer Sentiment. 

No. The AI review summary is included in the conversation API at no extra cost for Advanced and Enterprise clients.

No. Review Summaries are only available upon request through your Bazaarvoice point of contact. It takes up to 4 weeks to generate the summaries after the client has been onboarded. This time frame allows for full integration of your data, system setup, and ensuring the AI model has sufficient information to create accurate and meaningful summaries.

Review summaries are generated for all active products with at least 30 approved reviews combined across English, French, German, and Spanish, regardless of the source (native, family, or syndicated) at the time of the request. Newly added products will be included in the next AI model cycle, ensuring they are regularly updated and reflected in the summarization process.

Yes. The summaries are up to 100 words in length.

Yes. The AI review summary model has been evaluated for relevancy and representativeness against the original product reviews. It aims to capture the overall tone of the reviews, including positive, negative, or neutral feedback. While summaries are concise and may not cover every topic in all reviews, especially for products with a large volume of reviews, we use extensive evaluation metrics to ensure the summary reflects the range of opinions of the product's consumers.

No. Negative reviews cannot be excluded. The AI model considers all displayable reviews of a product, including positive, negative, and neutral reviews. The summaries generated are based on all reviews, ensuring they are completely authentic.

Yes. It analyzes all review content to identify themes as unbiased topics, which are then represented as filters to help consumers quickly find related reviews (e.g., "taste" for food or "texture" for beauty products). These keywords are listed in descending order based on the number of reviews in which they appear.

Yes. Locale-level configuration is supported for review summaries. You can select which locales to enable. 

No. You cannot select specific products for the summary to display on. If the feature is enabled, it applies sitewide to all eligible products with a minimum of 30 displayable english reviews.

Yes. You can preview the review summary in the staging environment before deploying it to production. The staging environment uses test content. However, to generate a summary based on real customer feedback for a specific product ID, use the Review Summarization API  with production API keys. The results are identical in both environments, as the summary is generated based on production reviews. Ensure that the product ID has at least 30 approved reviews combined across English, French, German, and Spanish, regardless of the source (native, family, or syndicated).

Yes. You can share the Bazaarvoice Use of AI documentation, which covers our approach to security, privacy, and quality.

Yes. The feature supports English, French, and German, with Spanish support added in Q3 FY25.

No. Configuration is managed via ConfigHub at the implementation level and applies globally to all locales within that specific site implementation. Locale-specific enabling or disabling is not supported.

No. If a locale does not yet support keyword generation, the filters will simply not appear on the Product Detail Page (PDP). There is no negative impact on page performance.

Yes. The feature does not make a regional distinction for supported languages. If enabled, keywords will display across all locales for that language (e.g., both en_US and en_CA), provided the review volume threshold is met.

Yes. A product must have a minimum of 10 approved reviews written exclusively in a supported language (English, French, German, or Spanish).

No. The data pipelines that refresh the keywords run periodically (typically once every 4 weeks). If a product reached the 10-review threshold after the last pipeline run, the filters will not appear until the next automated refresh.

No. Keyword Filters are built on our legacy sentiment dataset which excludes incentivized content. Only native and network reviews (including product variants and intra-site syndicated content) are used.

No. Keyword generation is currently not supported on staging (similar to Review Highlights). Testing must be done in production.

Yes. The underlying Natural Language Processing (NLP) pipeline runs automatically on a monthly basis. Clients do not need to take manual action to trigger a refresh.

No. Manual keyword exclusion or blocklisting is not supported at this time.

No. There is no cap on the total number of backend keywords generated, as it depends on the themes discussed in the reviews.