Amazon Kendra
Enterprise search software
- Features
- Ease of use
- Ease of management
- Quality of support
- Affordability
- Market presence
Take the quiz to check if Amazon Kendra and its alternatives fit your requirements.
Pay-as-you-go
Free Trial
Free version unavailable
Small
Medium
Large
- Healthcare and life sciences
- Education and training
- Transportation and logistics
What is Amazon Kendra
Amazon Kendra is a managed enterprise search service on AWS that indexes content from common repositories and enables natural-language search across an organization’s documents and knowledge sources. It is used by IT and application teams to embed search into internal portals, intranets, and customer support experiences. The service emphasizes connector-based ingestion, relevance tuning, and access control alignment with source systems, and it can be integrated with other AWS services for authentication, analytics, and application development.
Managed AWS-native deployment
Amazon Kendra is delivered as a fully managed AWS service, reducing the need to provision and maintain search infrastructure. It fits organizations that already standardize on AWS identity, networking, and monitoring. Consumption-based pricing and AWS account governance can simplify procurement and operational ownership for cloud teams.
Broad connector ecosystem
Kendra provides built-in connectors for several enterprise content sources and supports custom ingestion pipelines. This reduces time to index content compared with building and maintaining bespoke crawlers. It also supports incremental updates and metadata ingestion, which helps keep results current and filterable.
Relevance and access controls
Kendra includes relevance features such as query understanding, synonyms, and result ranking controls to improve findability for business users. It supports document-level access control by mapping permissions from source repositories, helping ensure users only see authorized content. Administrative tools allow tuning and monitoring of search behavior for specific domains and collections.
AWS dependency and lock-in
Kendra is tightly coupled to AWS for deployment, identity, and surrounding services. Organizations with multi-cloud or on-prem-first strategies may face additional integration work and governance constraints. Migrating search workloads away from Kendra can require re-implementing ingestion, ranking logic, and operational processes.
Cost can be hard to predict
Pricing depends on indexed content size, query volume, and enabled features, which can make budgeting difficult for large or highly active deployments. Connector usage and frequent re-indexing can increase ongoing costs. Teams often need careful capacity planning and monitoring to avoid unexpected spend.
Limited UI and app layer
Kendra primarily provides APIs and administrative configuration rather than a complete end-user search application. Organizations typically need to build or customize the front-end experience, analytics dashboards, and feedback loops. Compared with products that bundle knowledge management or workplace content workflows, Kendra focuses more narrowly on search services.
Plan & Pricing
| Plan | Price (base index + components) | Key features & notes |
|---|---|---|
| GenAI Enterprise Edition | $0.32 per hour (base index). Storage units: $0.25 per hour per unit. Query units: $0.07 per hour per unit. Connectors: flat $30 per index per month (includes up to 500 sync hours); $0.35 per hour when syncing; $1 per 1M documents scanned while syncing. | Includes up to 20,000 documents or 200 MB of extracted text; ~0.1 QPS (≈8,000 queries/day). Built for production, high availability, recommended by AWS. Discounted pricing available for high-volume workloads (request quote). Charges accrue from index creation until deletion. |
| Basic Enterprise Edition | $1.40 per hour (base index). Storage units: $0.70 per hour per unit. Query units: $0.70 per hour per unit. Connectors: $0.35 per hour when syncing; $1 per 1M documents scanned while syncing. | Includes up to 100,000 documents or 30 GB of extracted text; ~0.1 QPS (≈8,000 queries/day). Supports adding storage/query units. |
| Basic Developer Edition | $1.125 per hour (base index). Storage/Query units: Not applicable (cannot add units). Connectors: $0.35 per hour when syncing; $1 per 1M documents scanned while syncing. | Includes up to 10,000 documents or 3 GB of extracted text; ~0.05 QPS (≈4,000 queries/day). Intended for proof-of-concept; not recommended for production. |
Additional notes:
- Connector charging details: For GenAI Enterprise Edition connectors AWS charges a flat $30/index/month (includes up to 500 sync hours); for other editions connectors are charged $0.35/hour while syncing plus $1 per 1M documents scanned.
- AWS provides pricing examples and notes that once an index is created you are charged for storage and query unit hours until it is deleted.
Seller details
Amazon Web Services, Inc.
Seattle, Washington, USA
2006
Subsidiary
https://aws.amazon.com/
https://x.com/awscloud
https://www.linkedin.com/company/amazon-web-services/
