NEW YORK -- Aug. 31, 2026 -- Optimizely unveiled Virtual Teammates, a lineup of role-specific AI agents designed to work as persistent digital coworkers rather than one-off prompt tools, at its Opticon conference on Aug. 31, 2026.
The launch responds to a fragmentation problem the company documented in a global study of more than 2,000 B2B marketing leaders: 81% said they personally switch between two or more disconnected AI tools weekly, and 19% juggle four or more. Optimizely CEO Alex Atzberger framed the shift as a departure from task-based automation. "A Virtual Teammate learns how your organization works once, then keeps going," he said, calling it the biggest advance since AI agents first appeared in marketing software.
Five roles ship at launch, from Chief of Staff to CRO Manager
Optimizely is rolling out five initial Virtual Teammates: a Chief of Staff that tracks follow-ups and delivers recurring briefings; an SEO & AI Search Analyst that audits AEO performance and search visibility; a Marketing Analyst that converts acquisition and conversion data into decision-ready reports; a Personalization Strategist that builds and monitors targeted campaigns; and a CRO Manager that analyzes experiments to flag optimization opportunities.
Unlike generic chatbots, each teammate carries defined responsibilities and organizational permissions that persist across sessions. Marketers configure recurring triggers—weekly briefings, scheduled audits, pre-meeting summaries—without re-explaining company context each time, according to Optimizely.
Identity-based permissions aim to solve AI accountability gaps
Kevin Li, SVP of Product at Optimizely, said stalled AI adoption often traces back to unclear accountability: teams cannot always verify who did what or trust an AI system enough to let it act unsupervised. Optimizely built each Virtual Teammate with its own identity and permission set, tying every action to an audit trail and letting organizations designate checkpoints requiring human review.
That governance layer distinguishes the product from standalone AI writing or analytics tools that lack traceable execution history. Companies can assign work, involve teammates in existing workflows, and adjust responsibilities using plain-language instructions rather than custom-coded automation scripts.
Teammates come from a directory, not a build-from-scratch model
Rather than requiring marketing teams to construct agents and workflows internally, Optimizely is distributing Virtual Teammates through a directory where organizations select and configure pre-built roles. Teams connect each teammate to the specific marketing systems it needs to operate, then customize scope and responsibilities without engineering resources.
The approach mirrors a hiring model: teams "hire" a teammate suited to a functional need, then adapt its remit over time as priorities shift. Optimizely positions this as a lower-friction alternative to the current landscape, where marketers manage multiple disconnected point solutions for SEO, personalization, and experimentation separately.
Product extends Optimizely's existing Agent Platform
Virtual Teammates build on Optimizely's Agent Platform, which already combines content management, commerce, personalization, and analytics in a single system. The company frames the addition as a move from isolated, prompt-driven AI tasks toward continuous, proactive support embedded directly in marketing operations.
Pricing details and enterprise rollout timelines were not disclosed in the announcement. Availability of the initial five roles begins immediately for Optimizely customers accessing the platform's directory.