Webex Workforce Optimization and quality management

Verified 2026-09-05 · 63 claims · sources tier 2 · 1 disputed

Also known as A-FLEX-AI-QM, Advanced Sentiment, Auto QM, Basic QM, Basic WFM, Calabrio, Enterprise Analytics, Interaction Summary, MyTime, Quality Management, Trending Topics, WFO, Webex WFO, Workforce Optimisation, Workforce Optimization.

Webex Workforce Optimization (WFO) offers 2 streamlined tiers, Basic WFM and Basic QM, alongside its full Workforce Management and Quality Management offerings 53. Cisco positions these basic packages for customers utilizing basic call recording without full quality management and for organizations scheduling with spreadsheets 36.

Packaging and Feature Tiers

Basic WFM includes forecasting across voice, digital, and back-office channels 37. Full Workforce Management adds advanced notifications, a vacation planner, an overtime manager, open API integrations, a chat bot, self-scheduling, Insights, standard payroll integrations, and 1 historical plus 1 real-time connector 45.

Basic QM provides call recording designed to secure and store interactions for compliance, training, and evaluations, along with screen recording captured alongside audio 33​34. It also provides the ability to create and edit evaluation forms and view standard evaluations without customizable workflows 32. In contrast, full Quality Management adds interaction export, Insights reporting, quality management workflows, custom metadata, QM Views, a contact queue, live screen monitoring, evaluation calibration, evaluation appeal, and an unspecified capability designated as HR and training contact tagging 44​49.

Ordering Basic WFM and Basic QM requires dedicated SKUs and provisioning information submitted at order time 35. However, Webex WFO packaging documentation names SKU categories without publishing alphanumeric SKU codes 47.

Documentation for Basic WFM and Basic QM exists under 2 distinct articles, nmzg380 and cfe6jj, which present different scopes and feature sets rather than duplicate text 52. The nmzg380 article targets Webex Contact Center, while cfe6jj targets Webex Contact Center Enterprise, Contact Center Enterprise, and Contact Center Express 40. Within the cfe6jj scope statement, Contact Center Enterprise is named twice out of the 3 listed products, which reads as an error in the source 41. Under cfe6jj, Basic QM evaluates voice interactions only and omits both Text Search and digital channels, whereas the Webex Contact Center document evaluates both voice and digital channels and includes Text Search 38​39.

Enterprise Analytics and Advanced Sentiment

Enterprise Analytics provides AI conversation analysis and is described as an entry point into AI suited to on-premises customers 43. Auto QM, a component of this capability, can evaluate up to 100 per cent of interactions against consistent criteria, and agents are able to appeal Auto QM scores 30​31.

The Analytics plus Transcription base tier contains:

• Predictive evaluation scoring and predictive net promoter scoring 27​48
• Non-advanced sentiment analysis 27
• Phrase manager, interaction summary, and transcription 27
• Live screen monitoring, agent explorer, desktop analytics, application usage, and desktop manager 27

Enterprise Analytics adds Auto QM, trending topic analysis with impact sentiment and notifications, and enhanced advanced sentiment to that base tier 42.

Advanced Sentiment operates only on voice interactions, producing distinct sentiment scores for the agent and the customer 8​4. It requires a manual 1-time customer enablement, after which legacy sentiment ceases to display for new interactions 6. Step-by-step instructions for activating Advanced Sentiment are hosted on wfohelp.com rather than help.webex.com 7. In the field, role-based online help across operational roles (agents, team leaders, forecasters, schedulers, intraday analysts, evaluators, and administrators) is hosted on wfohelp.com rather than cisco.com or help.webex.com 63. Sentiment trend tracking across conversations and identification of sentiment shift moments require either the Enterprise Analytics SKU or a WFO Bundle with a 100 interactions add-on SKU 5.

Control Hub AI Quality Management Configuration

Within Control Hub, purchasing AI Quality Management requires either the standalone add-on SKU A-FLEX-AI-QM or the bundled AI Assistant and AI Quality Management SKU A-FLEX-AI-QMMB 24. Users must have AI Assistant enabled in their profile access via the Desktop Experience setting 19. Features are configured under Services > Contact Center > AI Features under Desktop Experience 15. AI Quality Management features are configured at organisation, queue, or agent level, depending on the feature 26.

Sources disagree regarding the configuration scope of Auto CSAT: the feature table marks Auto CSAT as agent-scoped, whereas the configuration walkthrough instructs administrators to enable it at the queue level for all queues or specific queues 29.

Sentiment, Auto CSAT, and real-time transcriptions are also bundled within a separate AI Assistant SKU 23. To start real-time transcription, administrators must place a Start Media Stream activity after the AgentAnswered event in Flow Designer and publish the flow 50. Sentiment analysis requires active transcription on target queues and updated flows 51, while Auto CSAT requires call recording due to its dependency on post-call processing 28. The Control Hub AI transparency technical notes list only 1 note for Auto CSAT despite covering 5 separate AI QM features 16.

AI Architectures and Underlying Models

Webex AI Quality Management relies on fine-tuned BERT models, statistical classifiers, and third-party large language models via Microsoft Azure OpenAI Service 14. These fine-tuned evaluation models are trained per organization, are never shared between customers, and remain inside the Webex Contact Center cloud 17. Its sentiment score ranges from -100 to 100, broken into negative (-100 to -45), neutral (-45 to 45), and positive (45 to 100) 22. These boundaries overlap, as -45 falls into both negative and neutral bands, and 45 falls into both neutral and positive bands 21. The tool is restricted to the supervisor desktop, requiring a supervisor license and Control Hub permissions 25. Supervisors can override AI evaluation scores, which feeds automatic model retraining 18.

Webex WFO enhancements rely on AWS Bedrock using Anthropic and Amazon Nova models along with Cisco proprietary transformer models 55. Calls to AWS Bedrock are stateless and do not store or use customer data for model training 54. These 5 enhancement features provide opt-in and opt-out controls at agent, group, and team levels 59.

Webex WFO Agent Assist is a self-service generative AI assistant embedded in the WFO workforce management module via the MyTime interface 13. It supports 135 languages and uses Azure OpenAI models exclusively without additional vendor models or model fine-tuning 9​10. Agent Assist event logging is opt-in and stores records in a tenant-specific database 11. Real agent interactions are used for prompt refinement only with explicit client permission; otherwise, synthetic conversation logs are used 12.

WarmTransfer's reading of the sources is that 3 capabilities marketed under adjacent Workforce Optimization and Quality Management branding rest on 3 distinct model stacks: AI Quality Management uses Azure OpenAI with BERT; WFO enhancements use AWS Bedrock with Anthropic, Amazon Nova, and Cisco transformers; and WFO Agent Assist uses Azure OpenAI alone 61.

The 3 AI transparency notes for these products contain documentation artifacts: each footer lists copyright 2021 immediately preceding a second, correct copyright year, and each references section ends with a stray trailing fragment reading "Las" 57​58.

Architecture Guides and New WFM

Documentation articles for the Webex WFO cloud design guide, the New WFM design and installation guide, and the New WFM Webex Contact Center integration guide are stubs containing only PDF viewer placeholders without article prose 56. Field reports suggest that the cisco.com document URLs ending in .pdf redirect to help.webex.com HTML articles rather than delivering raw PDF files 60. Additionally, the New WFM design and installation guide shares the identical one-sentence description with the integration guide despite having different titles 62. The documentation titles reference New WFM without explaining what New WFM is, what distinguishes it from other variants, or which variant is current 46.

See also

• Employment decision safeguards on automated agent scoring
• Webex Contact Center Enterprise add-on ecosystem
• Cisco per-model AI transparency technical notes
• Third-party model and speech vendors behind Webex contact centre AI
• Webex Contact Center recording and contact data retention
• Webex Contact Center Enterprise as a distinct product
• Classic WFM and New WFM variants of Webex Workforce Optimization
• Webex WFO role based operational documentation on wfohelp.com
• Webex Contact Center platform generations and predecessors
• Migration from Classic WFM to New WFM
• Webex Contact Center Functions as a development platform
• Webex Engage as the digital channel product
• Webex Contact Center Analyzer and the legacy reporting split
• The New WFM workforce management API surface
• Failure and rollback behaviour of the Classic to New WFM transition

Applicability

Across the claims in this article, the evidence covers Cisco Webex Workforce Optimization, Webex Contact Center, Webex Contact Center Enterprise, Contact Center Enterprise, and Contact Center Express on-premises deployments, verified as of 2026-09-05.

What remains uncertain

What Classic WFM and New WFM are and which is current is not covered by any claim in this article beyond the presence of the New WFM term in guide titles. The contents of the role-based WFO operational documentation on wfohelp.com are not covered by any claim in this article. How Auto QM computes an evaluation score is not covered by any claim in this article. How evaluation appeal works for AI-scored evaluations is not covered by any claim in this article beyond the fact that appeals can be submitted. The cost of the Basic WFM and Basic QM packages is not covered by any claim in this article. The operational behavior and functional role of HR and training contact tagging are not covered by any claim in this article. Edge components, data flows, storage mechanisms, and encryption architectures for WFO cloud deployments are not covered by any claim in this article. How Webex WFO attaches to Unified CCE is not covered by any claim in this article.

WarmTransfer. (2026, September 5). Webex Workforce Optimization and quality management. WarmTransfer. https://warmtransfer.net/knowledge/webex-wfo-quality-management

@misc{warmtransfer-webex-wfo-quality-management,
  title  = {Webex Workforce Optimization and quality management},
  author = {{WarmTransfer}},
  year   = {2026},
  url    = {https://warmtransfer.net/knowledge/webex-wfo-quality-management},
  note   = {Verified 2026-09-05}
}