Verint Agent Factory: Can Hybrid Human-AI Workforce Orchestration Actually Work?

Verint Agent Factory and the Case for Hybrid Workforce Orchestration

13
Verint Agent Factory, CX Today
Contact Center & Omnichannel​Explainer

Published: July 29, 2026

Rebekah Carter

Verint’s Agent Factory, launched at Engage 2026 alongside a completed Calabrio merger, is trying to become the control layer for a contact center workforce that now includes AI agents alongside people, and its strongest evidence isn’t the agent-building tools, it’s proof that a promised action actually got completed somewhere else in the business.

Like it or not, the contact center has acquired a second workforce. It might not appear on the org chat or ask for time off, and half of the business probably still creates it like a collection of features, but it’s there. That’s already causing trouble.

AI agents now summarize calls, answer customers, search knowledge, update records, shift schedules, and trigger work in other systems. Once several of them start acting at the same time, somebody has to decide who owns the task, which data they’re allowed to touch, and when a person needs to take over. That’s where hybrid workforce orchestration starts earning its name.

This isn’t just Verint’s story to tell. In a survey of more than 320 service and support leaders reported by CX Dive in January 2026, 85% said they were expanding human agents’ responsibilities rather than cutting headcount, a pattern Gartner VP analyst Kathy Ross attributed to workforce redesign rather than outright role elimination.

The awkward truth is that most contact centers still manage humans, bots, copilots, and workflow tools in separate places. NiCE and Genesys are both pushing toward one operating model. Verint has gone further into the machinery with Verint Agent Factory, launched at Engage 2026. The question is, will this solution give contact centers the management tools they really need?

TL;DR: Is Verint’s Hybrid Workforce Bet Working?

TL;DR

  • The merger is done and the leadership choice was telling: former Calabrio CEO Dave Rhodes now runs the combined company, and Calabrio’s WFM brand stayed in the portfolio rather than getting parked.
  • The AI shift was financially visible before Agent Factory shipped: Verint’s AI ARR reached $372 million (+21.2% YoY) while non-AI ARR fell 5.6%.
  • The strongest idea is outcome verification, not agent-building: Quality Intelligence checks whether a promised action (like a refund) actually happened in another system, not just whether the right words were used.
  • The open question: whether customer-built and third-party agents get controls as deep as Verint’s own products, and whether both WFM brands get equal investment going forward.

In This Article

Why Are Contact Centres Moving Towards Hybrid Human and AI Workforces?

Contact centers are moving toward hybrid human and AI workforces because AI agents now touch live customer records, schedules, and service workflows, creating a management job most businesses haven’t designed for yet.

Contact centers are moving toward hybrid human and AI workforces because AI agents now touch live customer records, schedules, service workflows, and employee decisions. That creates a management job most businesses haven’t properly designed for yet: deciding who handles the work, what each agent can change, and when a human must step in.

Buying an AI agent is easy, the trouble starts when several of them work on the same customer, each pulling from different systems and following different rules. At that stage, AI agent orchestration becomes essential, allowing companies to divide work across specialist agents handling everything from identity checks to follow-up.

Verint’s ecosystem is designed to help companies more effectively manage the complete rolodex of human agents, bots, planners, supervisors, copilot, and background automation tools.

Customers aren’t the only ones who benefit. Verint’s 2026 research found that 31% of agents expected to leave within six months. Another 61% saw their jobs becoming more technical or complex, while just 8% expected AI to replace them.

Key Takeaways

  • 31% of agents surveyed by Verint expect to leave within six months, while only 8% fear outright AI replacement. The bigger problem is how the job is designed.
  • Agents spent close to three minutes searching for an answer during 45% of calls, a fragmented-tools problem AI orchestration is meant to fix.

Learn more about the challenges of human and AI workforce management here.

What Changed When Verint and Calabrio Came Together?

The Verint-Calabrio deal created one of the biggest workforce technology estates in the market, with former Calabrio CEO Dave Rhodes now running the combined company and its WFM brand kept in the portfolio.

The Verint-Calabrio deal created one of the biggest workforce technology estates in the contact center market. It also put two competing WFM, quality, recording, and analytics portfolios under one owner. That scale strengthens Verint CX automation, although customers now need much clearer answers about product direction and investment.

Thoma Bravo closed its roughly $2 billion Verint acquisition on November 26, 2025, then merged it with Calabrio. The combined company took the Verint name on February 18. Six days later, former Calabrio CEO Dave Rhodes became CEO, while Dan Bodner moved into an advisory role. That decision matters. Calabrio wasn’t tucked away after the deal. Its CEO took charge, and its WFM brand remained.

By March, Calabrio customers could access Genie Bot through Conversation Intelligence, TimeFlex inside Calabrio WFM, Agent Copilot Bots, and Verint IVA. That’s useful evidence of cross-portfolio delivery, rather than another acquisition presentation full of future-tense promises.

Our Verint AI Pivot report captured the shift. AI ARR reached $372 million, up 21.2% year over year, while non-AI ARR dropped 5.6% to $356 million. Total subscription ARR was $728 million. Those figures cover the quarter ending July 31, 2025, and shouldn’t be mistaken for recognized 2026 revenue.

That crossover matters because it shows Verint CX automation was gaining ground before Agent Factory appeared. It also exposes the pressure underneath the deal: the older business was shrinking.

Agent Factory now has a second job. It needs to make the combined estate feel like one strategy before customers start wondering which products are living on borrowed time.

Customers have already offered their verdict. In January 2026, Verint earned two TrustRadius Buyer’s Choice awards based entirely on customer reviews. Verint CX Automation won the Contact Center category, while Verint WFM took Workforce Management.

Key Takeaways

  • Former Calabrio CEO Dave Rhodes now runs the combined company; Calabrio’s WFM brand stayed in the portfolio rather than being retired.
  • Verint’s AI ARR hit $372M (+21.2% YoY) while non-AI ARR fell 5.6% to $356M, the shift was visible before Agent Factory even launched.
  • Verint won two January 2026 TrustRadius Buyer’s Choice awards (Contact Center and Workforce Management categories), decided by customer reviews rather than Verint’s own claims.

How Does Verint Agent Factory Coordinate Human and AI Work?

Verint Agent Factory puts agent creation, prompt control, model choice, data access, and human handoffs in one operating environment, with Quality Intelligence checking whether a promised action actually happened elsewhere.

Verint Agent Factory coordinates human and AI work by putting agent creation, prompt control, model choice, data access, and human handoffs in one operating environment. Companies can use prebuilt Verint agents or create their own, then connect them to live workflows with rules covering who acts, what they can change, and when judgment moves back to a person.

A prompt change can affect thousands of customer interactions. Leaving prompts scattered across separate tools is asking for trouble.

Agent Factory centralizes prompt controls across agents and applications. It works with OpenAI, Gemini, Anthropic, and Meta Llama models, as well as bring-your-own-model deployments. Verint also supports MCP, A2A, and ACP connections to outside systems. Buyers get model choice without handing control to the model provider.

The governance piece is crucial too, particularly as customer journey orchestration affects CX security. Buyers should demand named agent identities, restricted access, approval thresholds, complete action logs, and rollback controls.

Agent Factory becomes more interesting when it connects with Verint’s other Engage launches, too.

Workforce Intelligence can adjust staffing and task assignments as conditions change across Verint’s WFM portfolio. Desktop Intelligence converts field entries, navigation paths, and process steps into structured data, exposing how employees actually complete work. Quality Intelligence then compares the conversation with the action recorded in another system.

There’s a trap here. Verint CX automation could make a poor workflow run much faster. Desktop Intelligence needs to expose bad process variants before Agent Factory copies them across the human and AI workforce management model.

That connection is Verint’s strongest idea. The unanswered question is whether customer-built and third-party agents receive controls as deep as Verint’s own products.

Key Takeaways

  • Agent Factory supports OpenAI, Gemini, Anthropic, and Meta Llama models plus bring-your-own-model, with MCP/A2A/ACP for external system connections.
  • Quality Intelligence’s refund example is the sharpest proof point: catching a promised-vs-delivered mismatch same-day, not just checking language quality.

CX Awards 2026 Categories Announced

CX Today has announced all 25 categories for the CX Awards 2026, recognizing excellence across customer engagement, contact center, analytics, and AI-powered CX platforms.

Explore the full CX Awards 2026 category list

Can CX Automation Work Without Replacing the Existing Contact Center Stack?

Open CX automation lets companies add AI agents and workflow controls across their current ACD, CRM, and WFM systems, with Verint pitching a contained-use-case start rather than a multi-year platform migration.

Open CX automation lets companies add AI agents, workflow controls, and decisioning across their current ACD, CRM, WFM, and business systems. Verint’s pitch is that large enterprises can start with one contained use case, prove the outcome, then expand without turning the project into a multi-year platform migration.

Every vendor claims openness. The useful test is whether the customer can keep its existing systems, models, data rules, and deployment choices once the project reaches production.

That means checking where the action is executed, how identity is resolved, which decision rules apply. Latency deserves the same scrutiny. Authentication and in-session decisions may need sub-second responses. Chat or next-action use cases might tolerate a few seconds. “Real time” tells a buyer very little unless the vendor can show performance under production load.

A modular approach suits companies that can’t rip out their contact center and start again. They might automate quality reviews, call wrap-up, knowledge retrieval, or one self-service process first. Broader automation can follow once the controls hold.

Verint says focused deployments can reach production in weeks. Its own guidance is more realistic about larger programs, which can take months once integration, testing, training, and governance enter the picture.

The “no rip-and-replace” promise still needs hard questions. Buyers should ask which features require Verint Cloud, which connectors are native, whether customer-built agents get the same controls, and how pricing changes as model use climbs. Contract language matters too. Support today doesn’t guarantee equal investment across both Verint and Calabrio WFM products in 2027.

Key Takeaways

  • Openness should be judged by the event-identity-decision-action chain holding together, not by a generic “we have APIs” claim.
  • Verint says focused deployments can reach production in weeks; larger programs realistically take months once integration and governance are added.

What Evidence Shows That CX Automation Is Producing Measurable Results?

Verint has solid ROI examples across quality, coaching, and scheduling deployments, though the figures are self-reported and haven’t been independently audited.

The proof of CX automation is a measurable change in workload, revenue, retention, or completed customer actions. Verint has solid examples, though buyers shouldn’t copy the ROI figures without seeing the baseline, costs, and calculation. These numbers haven’t been independently audited. They’re self-reported by Verint or named customers.

A large North American industrial distributor offers one of the stronger examples. Using Verint Quality Bot, it reportedly evaluated 97% of more than 200,000 monthly calls, lifted supervisor capacity by 80%, and avoided hiring over 200 analysts. Verint put the annual saving at $12.5 million.

BT offers a different kind of proof. Its coaching deployment produced a reported $9 million increase in annual cross-sell revenue and cut new-hire time to competency by three weeks. That matters because it ties Verint CX automation to commercial performance, rather than another vague productivity claim.

The workforce examples are just as useful. An online bank using Request Anything reduced incoming scheduling requests by 25% and saved around 200 hours each week, equivalent to five full-time employees. That is modular CX automation doing a very specific job well.

Dominion Energy’s Genie Bot case is smaller, but revealing. Analysts could examine up to 100 calls in roughly 30 minutes, compared with 15 to 20 before. During a website redesign, that wider sample showed complaints represented less than 1% of total call volume. The tool didn’t merely make analysis faster. It stopped the business from overreacting to a loud minority.

A bot can “contain” a contact and still leave the customer stuck. Buyers should pair containment with repeat contact, confirmed completion, human correction rates, and cost per resolved outcome.

Key Takeaways

  • North American industrial distributor: 97% call coverage across 200,000-plus monthly calls, 80% more supervisor capacity, and an estimated $12.5 million annual saving.
  • BT: $9M annual cross-sell revenue increase and 3 weeks faster new-hire competency, real commercial performance, not just a productivity claim.
  • Containment alone is a weak success metric; pair it with repeat contact, human correction rate, and cost per resolved outcome.

Is Hybrid Workforce Orchestration Becoming the New Standard for Large Contact Centers?

Large contact centers now need a shared control layer for AI agents and people, and NiCE, Genesys, Cisco, and Five9 have all moved in the same direction within weeks of each other.

Large contact centers now need to coordinate people, AI agents, and automated workflows around the same customer outcome. That’s moved the market beyond standalone copilots. Buyers need a shared control layer for access, handoffs, workloads, quality, and ownership.

NiCE’s June 2026 Workforce Empowerment Suite applies common workforce and performance controls to people and AI agents. Genesys is pushing a similar idea through connected data, persistent context, and orchestration across service workflows.

Cisco’s Webex AI Workforce Engagement Management suite followed within a week, and Five9 extended its Assembled partnership to bring agentic workforce management into its own platform. Sheila McGee-Smith, founder and principal analyst at McGee-Smith Analytics, framed NiCE’s launch as a sign that once humans and AI agents share a workflow, the old rules for managing employees stop applying cleanly.

So Verint doesn’t get points simply for spotting the trend. Everyone serious about enterprise CX has spotted it. Where Verint Agent Factory looks different is the operational detail underneath the pitch. Verint already sits close to forecasting, scheduling, quality reviews, and the desktop work that follows a conversation. That gives its approach to human and AI workforce management more substance than an agent builder sitting above a collection of APIs.

Dimension / Verint Agent Factory / NiCE Workforce Empowerment Suite / Genesys

Dimension Verint Agent Factory NiCE Workforce Empowerment Suite Genesys
Launched Engage 2026 NiCE World, June 9, 2026 Agentic core unified, February 2026
AI model options OpenAI, Gemini, Anthropic, Meta Llama, plus bring-your-own-model Native agentic AI (Cognigy) built into the platform core Genesys-native agentic virtual agent plus Cloud AI Studio
Coordination approach Centralized prompt control, MCP/A2A/ACP connectors, human handoff rules One operating model spanning WFM, quality, performance, compliance Connected data and persistent context across service workflows
Distinctive claim Quality Intelligence checks whether a promised action was actually completed elsewhere Same evaluation criteria and performance standards applied to humans and AI Orchestration extended across front- and back-office workflows

Key Takeaways

  • NiCE, Genesys, Cisco, and Five9 all expanded hybrid workforce products within weeks of one another. This isn’t Verint’s private talking point. The whole market is moving.
  • Verint’s edge isn’t spotting the trend first, it’s the operational depth underneath: Agent Factory sits close to forecasting, scheduling, quality reviews, and desktop work already, rather than acting as a builder layered on top of someone else’s APIs.

Will Verint Agent Factory Win The Hybrid Agent Orchestration Battle?

The verdict: Verint’s operational depth, not its agent-building tools, is its real advantage, but Info-Tech’s own read still places it as a follower, not the primary market-maker, in agentic AI and CX platforms.

Verint is making some important strides in the hybrid agent orchestration space, but it’s not alone. According to a July 2026 vendor note, Info-Tech Research Group analysts Shashi Bellamkonda and Thomas Randall credited Agent Factory with giving Verint a clearer answer on agentic AI and a larger workforce-engagement footprint after the Calabrio deal.

Still, they also added that Verint remains a follower rather than the primary market-maker in agentic AI and CX platforms, and still has to prove Agent Factory is more than a business-user interface layered on its existing bot-building tools.

The decisive feature may be outcome verification. Plenty of platforms can route work or suggest the next action. Verint is trying to check whether the action promised to the customer was actually completed in another system. For a large contact center, that gap causes repeat calls, complaints, and expensive remediation.

Explore the Best CX Solution (Large) Award

The CX Awards 2026 will recognize the enterprise-scale platforms proving they can coordinate a hybrid human-AI workforce around a verified customer outcome.

Learn more about the Best CX Solution (Large) Award

Frequently Asked Questions

Does Verint Agent Factory replace an existing CCaaS platform?

No. Verint Agent Factory is positioned as an orchestration layer that connects with existing ACD, CRM, WFM, data, and workflow systems, rather than a wholesale platform replacement. That matters for large enterprises with mixed estates. Buyers should still confirm which functions depend on Verint Cloud and which integrations require custom work before assuming full portability.

Are Verint WFM and Calabrio WFM both still supported?

Yes. Verint currently says both WFM products remain supported and that customers won't face a forced migration. Workforce Intelligence is also intended to work across the combined portfolio. The bigger question is whether both products receive equal features and development funding as Verint rationalizes overlapping capabilities over the next year or two.

Does Verint Agent Factory support third-party AI models?

Yes. Agent Factory supports models from providers including OpenAI, Google, Anthropic, and Meta, alongside bring-your-own-model options. Model choice is only part of the buyer test, since enterprises should also check whether external agents receive the same prompt controls, audit records, access limits, and rollback options as Verint's own native bots.

How are TimeFlex and Request Anything different?

TimeFlex gives agents policy-controlled flexibility to adjust their schedules around staffing needs. Request Anything covers a wider range of in-shift changes, including coaching, meetings, training, and breaks. Both support human and AI workforce management, though Request Anything reaches further into the routine scheduling work usually handled by planners directly.

Can Verint CX automation work across on-premises and cloud systems?

Verint says its open CX automation model can connect cloud and on-premises environments without requiring a full platform replacement. Buyers need proof at connector level, since data movement, latency, residency rules, and failover behavior can expose real gaps that a broad 'open platform' claim alone won't reveal.

AI Agents
Featured

Share This Post