Contact Center and Omnichannel teams are moving into a tougher phase of AI evaluation, as new research and deployment signals show buyers need proof across platform selection, workflow execution, workforce design, QA governance, and measurable value.
TL;DR
- ISG’s 2026 Contact Centers Buyers Guide shows buyers are evaluating AI alongside automation, analytics, integration, governance, and lifecycle ROI.
- Telstra is using Salesforce Agentforce to automate live service, refund, and security-permission workflows across more than 1,000 contact center representatives.
- Cresta’s 2026 CX Workforce Report says only nine percent of customer conversations are fully handled by AI without human involvement.
- ScorebuddyCX finds AI QA adoption is widespread, but trust, review discipline, and coaching conversion remain uneven.
- The signal for CX leaders is clear: AI value now depends on whether platforms can support real operating change, not just feature adoption.
ISG Shows The Buyer Bar Is Getting Wider
The latest ISG Buyers Guide Contact Centers 2026 provides a useful starting point for the week because it shows how contact center buying criteria are expanding.
ISG evaluated 27 software providers and named NiCE, Verint, and Genesys as the top three overall leaders. It also identified Content Guru, Dialpad, Five9, Genesys, Microsoft, NiCE, RingCentral, Salesforce, Sprinklr, Talkdesk, Verint, and Zendesk as Exemplary providers.
The guide weighted Product Experience at 80%, covering capability and platform strength. Capability included areas such as agent workspace and collaboration, knowledge management, experience and interaction analytics, AI and automation, and AI governance and model operations. Platform criteria included adaptability, manageability, reliability, and usability.
Customer Experience accounted for the remaining 20%, with ISG assessing validation and TCO/ROI. That matters because buyer scrutiny is no longer limited to what a platform can do in a product demonstration. Buyers increasingly need evidence that the provider can support adoption, lifecycle value, and outcomes after deployment.
Why It Matters for Contact Center and Omnichannel
ISG’s framework points to a more demanding enterprise buying environment. Contact center platforms are being judged as broader customer engagement environments that need to orchestrate interactions, workflows, analytics, governance, and operational decisions across the customer journey.
That raises the stakes for vendors and buyers. AI capability is now one part of a wider platform question: can the system integrate with enterprise workflows, govern AI behavior, support self-service, improve employee experience, and demonstrate value over time?
The rest of this week’s news make that buyer shift more concrete. Telstra shows what AI execution can look like inside live workflows. Cresta shows how AI changes the shape of human work. ScorebuddyCX shows why trust and feedback loops decide whether AI insight becomes operational improvement.
Telstra Puts Agentic AI Into Live Service Workflows
Salesforce said Telstra is deploying Agentforce to automate high-volume processes across interim services, refunds, and security compliance.
The rollout already covers millions of customer interactions annually and supports more than 1,000 contact center representatives. Telstra has several use cases in production, including an interim service agent that validates eligibility, calculates refund amounts, monitors service restoration, applies refunds, and updates customer records.
Telstra is also using a security permissions agent to monitor user access, system configurations, and permissions across its Salesforce environment. Salesforce said the agent identifies high-risk permissions and analyzes access patterns, reducing the manual effort required for routine compliance and governance activities. Amy Childs, Customer Service Executive at Telstra, framed the deployment as a way to increase service efficiency while keeping accountability with the frontline team:
“Agentic AI is the next phase of AI at Telstra, but our people remain accountable and core to customer service delivery.”
Why It Matters for Contact Center and Omnichannel
Telstra’s use case is important because it reaches into service administration, refund calculation, compliance, and record updates. Those areas sit closer to operating reality than conversational AI alone.
For enterprise buyers, the question is whether AI can perform safely inside workflows that affect money, eligibility, records, and regulatory obligations. Telstra’s deployment suggests the next competitive test for vendors will involve controlled action, not just faster answers.
Salesforce also said future Agentforce capabilities for Telstra include a billing and payments agent and an email triage agent for B2B sales teams. That roadmap points to a broader service model where AI spans customer support, back-office processing, compliance, and revenue-adjacent workflows.
Cresta Says Contact Center AI Still Runs Through Humans
Cresta’s 2026 CX Workforce Report challenges the idea that contact centers are quickly becoming fully autonomous.
The report, based on a survey of 300 U.S. leaders responsible for customer support or CX software decisions, found that 76% of customer conversations are handled by humans and AI together. That includes 35% handled collaboratively and 41% handled by human agents with behind-the-scenes AI assistance.
Only nine percent of conversations are fully handled by AI with no human involvement.
Cresta also found that 93% of leaders say calls handled by human agents are becoming more complex. Meanwhile, 97% say AI has enabled them to redeploy staff into higher-value work, and 92% expect AI adoption to increase demand for skilled human workers. The report summarized the workforce implication this way:
“The future of CX will not be defined by a simple tradeoff between labor and automation, but rather by how well organizations redesign workflows, elevate human roles, formalize emerging AI responsibilities, and build the data and systems infrastructure required to support hybrid operations at scale.”
Why It Matters for Contact Center and Omnichannel
Cresta’s data reframes the labor conversation. The practical impact of AI is a different distribution of work, with routine interactions increasingly absorbed or assisted by AI and human agents handling more ambiguous, emotional, or high-stakes conversations.
That change raises the bar for workforce design. Contact centers need responsibilities around monitoring AI performance, designing human-AI workflows, coaching agents to work with AI, and managing risk.
Cresta also identified two adoption bottlenecks that should worry buyers: 81% cited integration complexity as a top barrier to AI adoption, while 53% cited poor data quality. Only seven percent said conversation data is easily accessible across the business.
Those findings support ISG’s broader buyer lens. AI value depends on integration, data access, role design, and platform readiness. The workforce impact cannot be separated from the architecture underneath it.
ScorebuddyCX Shows QA Is Becoming The AI Control Point
ScorebuddyCX’s Q2 QA & CX Intelligence Pulse Report adds another layer to the week’s contact center AI story: measurement and trust.
The report, based on a survey of 600 contact center professionals in the UK and U.S. conducted by OnePoll between July 9 and July 17, found that nine in ten contact centers now use AI to evaluate interactions. Automated scoring can reach up to 100% of interactions, compared with the two percent or three percent typically reviewed through manual QA.
Yet adoption has not resolved the trust issue. ScorebuddyCX found that 76% of managers trust AI-generated QA scores, compared with 57% of agents. It also found that only 14% always review AI scores, while 47% review them sometimes, rarely, or never. Derek Corcoran, CEO of ScorebuddyCX, described the market shift directly:
“The challenge then was adoption. The challenge now is impact.”
ScorebuddyCX also found that 15% of contact centers generating AI insights get no coaching from them at all. On average, 18% of coaching conversations are triggered by AI insight.
Why It Matters for Contact Center and Omnichannel
QA is becoming more than a performance-management function. In an AI-enabled contact center, QA becomes part of the operating system that checks whether human and AI interactions meet the required standard.
The tool score is not the end point. ScorebuddyCX’s data suggests the value comes after the score: whether someone trusts it, reviews it, explains it, coaches from it, and uses it to change customer-facing behavior.
The finding that AI chat or voice bots fully resolve an average of 19% of queries also offers a useful reality check. Automation is meaningful, but much of the experience still depends on what happens before, during, and after escalation.
For CX leaders, the risk is automated insight that does not change decisions.
The Operating Model Is Now The Proof Point
Taken together, the week’s sources show why contact center AI is entering a more demanding phase.
ISG’s guide shows buyers evaluating contact center platforms through a wider lens that includes AI, automation, analytics, integration, governance, platform reliability, and lifecycle value. Telstra shows AI agents being applied to live operational workflows. Cresta shows the workforce becoming more hybrid and more complex. ScorebuddyCX shows why QA trust and coaching conversion decide whether AI-generated insight actually improves the business.
The signal for enterprise CX teams is practical. Adoption is no longer enough to prove progress. Serious buyers should now ask where AI takes action, how people remain accountable, whether the data is accessible, how AI outputs are reviewed, and whether the platform can support measurable operating change over time.
Join the conversation: Join our LinkedIn community (40,000+ members): https://www.linkedin.com/groups/1951190/
Get the weekly rundown: Subscribe to our newsletter: http://cxtoday.com/sign-up
