Amazon Kendra review
A capable AWS search service with clear hourly pricing, but it is cloud-only and closed to new customers.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Amazon Kendra is a managed enterprise search service for organizations that need to find answers across structured and unstructured content. It combines machine learning, natural-language question answering, semantic ranking, FAQ matching, extracted passages, query autocompletion, and document-level access controls. Teams can search through web and API interfaces, while administrators can use search analytics, relevance tuning, and Visual Experience Builder with IAM Identity Center integration.
Pricing is usage-based rather than per-user. The Basic Developer Edition costs $1.125 per hour and supports up to 10,000 documents or 3 GB of extracted text, with 0.05 QPS and approximately 4,000 queries per day; it is designed for proof-of-concept solutions. The Basic Enterprise Edition costs $1.4 per hour and supports up to 100,000 documents or 30 GB of extracted text, with 0.1 QPS and approximately 8,000 queries per day for production workloads. The GenAI Enterprise Edition costs $0.32 per hour, supports up to 20,000 documents or 200 MB of extracted text, and includes a Retriever API for RAG workflows. A 30-day trial includes 750 hours during the first 30 days, excluding connector usage.
Its ecosystem fit is strongest for AWS-centered teams. Native connectors include Amazon S3, Amazon RDS, Amazon FSx, Atlassian Confluence Cloud and Server, Box, Microsoft SharePoint, Salesforce, ServiceNow, Google Drive, and Web Crawler, with scheduled synchronization and permission-aware filtering. Kendra suits teams seeking managed search, citations, analytics, and RAG retrieval without building those capabilities from scratch. It is a poor choice for organizations requiring on-premises deployment or new access, because AWS states that the product is no longer open to new customers.
Amazon Kendra pros and cons
- Where it wins
- Natural-language answers with extracted passages and citations
- ACL filtering, scheduled connectors, and semantic relevance tuning
- Retriever API supports retrieval-augmented generation workflows
- Where it doesn't
- Cloud-only deployment
- No longer open to new customers
- Usage limits vary substantially by edition
Amazon Kendra fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update
