For Customers
With relevant answers and personalized content, empower your customers to self-serve and continue enjoying your products or services without reaching out for help.
Coveo Smart Snippets
Create your own rich snippets to drive service success with instant answers
When a user types in a question, instantly serve up the response they need. Offer better customer support with relevant answers in fewer clicks.
Coveo Smart Snippets provides users with answers to their queries directly on the results page by displaying a rich snippet of the most relevant result item.
Leverage contextual cues from a person’s search and interactions to provide relevant, personalized answers right on the results page. Smart Snippets extract relevant answers from documents and content, so customers and agents don’t have to click into different articles to find the right one.
Add Smart Snippets to your website, community portal, chatbot, in-product help - anywhere you have text - to surface your best question-answering content wherever people turn for help. Plus, Coveo carries customer signals for seamless customer support from channel to channel.
Configure Smart Snippets AI models with clicks, not code, and start delivering rich snippets to your customers sooner. There is no need to train labeled data and can use existing indexed content, giving you faster time-to-value.
You can even customize the look and feel of Smart Snippet answers with our JSUI and headless frameworks, and APIs.
Rich snippets make self-service easy
For Support Agents
Let your support team ask questions and get answers too, directly in the insight panel. So, they can save time and focus on delivering truly great service.
Right answers run deep
Flexible Content Optimization
You can use JSON-LD formatting for optimal results, but Coveo’s machine learning Smart Snippet model will work with any HTML.
Designed for question-answering
Using an open-source, pre-trained sentence-transformer model (all-MiniLM-L6-v2), the Smart Snippets’ question-answering machine learning model converts headers and user query texts into embeddings, generates similarity scores, and retrieves the appropriate content snippet for each answer box.
Context preservation for deeper personalization
Our question-answering model works with other Coveo machine learning models, such as Automatic Relevance Tuning, to leverage contextual information and provide more refined and personalized snippet answers.
