Updates to the AWS Well-Architected Framework
AWS updated the Well-Architected Framework, splitting guidance into foundations and workloads while reducing workload best practices from 307 down to 216.
In short
Cloud engineers and architects · 216 workload best practices · 2026 · Review updated pillars
By Dillip Chowdary • Nov 16, 2026 • Source: AWS Architecture Blog

At a glance
- Who it is for
- Cloud engineers and architects
- The number
- 216 workload best practices
- Available
- 2026
- What to do
- Review updated pillars
- Foundations count
- 45 best practices
- Foundations capabilities
- 7 capabilities
In this post
Amazon Web Services announced updates to the AWS Well-Architected Framework, restructuring guidance across all six pillars into two distinct areas and reducing workload-specific best practices from 307 down to 216, according to AWS Architecture Blog's report.
This analysis details the separation of organization-wide foundation guardrails from workload architecture patterns for cloud engineers, architects, and platform teams. Technical leads can immediately review the streamlined catalog to run faster audits and automate architectural evaluations using artificial intelligence agents.
01Separation of foundations and workloads
AWS restructured the framework into Well-Architected Cloud Foundations and Well-Architected Cloud Workloads to eliminate redundant assessments. The update separates account-level and organization-level best practices from workload rules so teams evaluate multi-account environments once rather than repeating baseline checks for every application review.
Well-Architected Cloud Foundations establishes environmental guardrails that every application inherits, spanning multi-account design, identity setup, and cost governance. Well-Architected Cloud Workloads directs how teams design, construct, and operate individual software deployments, incorporating updated architecture blueprints mapped to targeted business outcomes.
The revision sharpens its focus on technical architecture, retiring non-technical practices concerning organizational process, business strategy, and people. By grounding remaining rules in configuration and telemetry decisions, the framework enables teams and automated agents to evaluate systems systematically against measurable criteria.
02Seven capabilities in cloud foundations
Well-Architected Cloud Foundations consolidates 45 best practices across seven core capabilities to govern entire cloud estates. Of these entries, 33 originate from relocating or merging existing recommendations previously scattered across the individual pillar documents of the original framework.
The seven capabilities comprise Multi-account environment, Identity and access management, Central observability management, Resource and inventory management, Security and compliance, Cloud financial management, and Networking and connectivity. The framework authors rewrote these relocated recommendations from an environment-wide perspective rather than an individual application viewpoint.
Newly authored foundational rules establish baseline operational visibility and traffic standards across member accounts. These additions introduce explicit requirements for centralized event monitoring, maintaining an auditable cloud inventory record, and implementing centralized domain name system management across distributed corporate infrastructures.
03Workload guidance consolidation and trimming
Advertisement
Tech Pulse Daily
Get tomorrow's pulse first
Join engineers who read Tech Pulse before stand-up. Free, weekday mornings.
Workload guidance now contains 216 best practices, down from 307, while maintaining full technical architecture coverage across applications. AWS achieved this reduction by moving environment-wide rules into foundational capabilities, merging overlapping recommendations, and retiring non-technical practices that relied entirely on organizational judgment.
Service citations throughout the documentation reflect active operational portfolios, removing retired products or services that no longer accept new customers. Concurrently, the framework incorporates design recommendations for newer services deployed since the previous major release of the architecture guidelines.
Each remaining practice addresses a technical configuration that architects or automated software agents can inspect, verify, and remediate. This structural cleanup speeds up review cadences by removing procedural survey questions that configuration checkers cannot parse or validate.
04Pillar changes for security and operations
The Operational Excellence pillar introduces guidance mapped to the operations lifecycle, addressing observation, insight generation, change flow management, deployment, support, incident response, and continuous evolution. It adds a requirement for managing external dependency rollouts alongside guidelines for applying artificial intelligence assistants to telemetry instrumentation, dashboard creation, and incident investigation under human supervision.
The Security pillar relocated root account protection, sign-in mechanisms, emergency access, centralized logging, and forensic readiness into foundational capabilities. For individual workloads, it introduces new best practices covering centralized workload permissions management, dedicated detection metrics and dashboards, and secure coding guidelines.
The Reliability pillar consolidated three previous networking rules concerning subnet allocation, hub-and-spoke topologies, and non-overlapping IP address ranges into a unified network strategy in foundations. Within workload reviews, Reliability adds explicit requirements for cyber resilience of backup data to defend systems against ransomware and external attacks.
05Efficiency metrics and sustainability updates
The Cost Optimization pillar relocated billing tools, account hierarchies, and cost controls into the Cloud financial management foundational capability. For workload reviews, it combines overlapping maintenance checks into a unified resource lifecycle management practice while adding a new recommendation for deploying generative artificial intelligence to drive operational efficiency.
The Performance Efficiency pillar shifted policies and reference architectures into foundations while orienting workload evaluation around validating performance requirements with key performance indicators. The update folds compute configuration directly into compute selection workflows and merges load testing procedures into broader performance evaluation routines.
The Sustainability pillar converted earlier general advice into verifiable best practices tracking resource provisioning, measurement, workload efficiency, and data processing. It mandates selecting resource-efficient computational methods, such as smaller machine learning models, and guides teams to benchmark usage metrics using the AWS Sustainability Console, application programming interfaces, and AWS Data Exports.
Frequently asked questions
What are the two main sections of the updated AWS Well-Architected Framework?
The framework is now split into Well-Architected Cloud Foundations, which covers environment-wide guardrails across 45 best practices, and Well-Architected Cloud Workloads, which covers individual application design across 216 best practices.
How did AWS reduce the number of workload best practices from 307 to 216?
AWS relocated 33 environment-wide practices into the Foundations area, combined overlapping recommendations into single best practices, and retired non-technical practices that could not be evaluated through technical telemetry or automated agents.
What new security and resilience best practices were added to workload guidance?
The update adds new workload practices for secure coding guidelines, centralized workload permissions management, detection metrics and dashboards, and cyber resilience to protect backup data against ransomware threats.
Sources
Author
Dillip Chowdary
Writes Tech Bytes coverage of AI, engineering, and the tools that actually ship. Editor of Tech Pulse Daily.
Related on Tech Bytes
Advertisement