March 10, 2026 • 4 min read
AWS Introduces Predictive Insights to Modernize Customer Support Models
Director of Content & Market Research
March 10, 2026
AWS has unveiled Predictive Insights, a new tool for its Amazon Connect CCaaS platform, which is now available in preview.
The tool helps contact centers anticipate what customers are likely to want next and adapt the experience that comes after.
To understand how this works, it’s critical to note that Predictive Insights is built on Amazon Connect Customer Profiles.
Customer Profiles consolidate data from across the enterprise pertinent to a customer and arrange it into a single view.
From there, the Predictive Insights tool kicks in to determine what the customer wants next, identifying patterns such as:
- Customers who are likely to churn
- Customers who may soon need help with another issue
- Situations where completing one action means another step will likely follow
Insights like these could fundamentally change how agents, human and AI, engage with customers in real time and proactively.
On the latter, contact centers can spin up new cases to save relationships and pre-emptively solve issues.
As such, service teams can fulfil more customer success tasks and pivot away from traditional reactive customer support models.
With 80% of organizations planning to move “some” customer service agents into new roles due to increasing automation, per Gartner, this tool also offers Amazon Connect customers a new way to maximize their agent talent pools.
Yet, it’s not just about giving service agents customer success responsibilities. Predictive Insights may also help turn service reps into sales reps, suggesting additional products the customer appears to need, based on cross-departmental data, in real time.
Again, this use case may help reposition the contact center as a revenue engine and support more individualized customer experiences.
More New Capabilities for Amazon Connect
Alongside Predictive Insights, AWS announced three additional capabilities for Amazon Connect at Enterprise Connect 2026. These also aim to bolster an increasingly automated customer support function.
1. Assistant for Managers (in Preview)
AWS has developed a unified interface for managers to assess the performance of AI and human agents together.
Critically, managers can interact with that interface using natural language to understand which queries are best automated and which necessitate more of a human touch.
For instance, a manager may ask: how does customer sentiment compare when an AI and a human agent handle refunds? The answer could then inform the contact center’s orchestration strategy.
Moreover, managers can track how changes to automated resolution flows impact customer outcomes and scale when they start delivering with parity with humans.
The Assistant may also share this insight with resource planners, who can then better forecast for AI.
2. Testing and Simulation (Generally Available)
While the Assistant for Managers tracks the performance of customer-facing AI agents after they go live, AWS’s new Testing and Simulation solution helps optimize their performance ahead of deployment.
Indeed, it simulates thousands of interactions with the agent to catch issues before a customer does.
As many AI agent implementations stall due to fear of something going wrong, the solution aims to support organizations in validating their designs and governance strategy ahead of time, accelerating deployments.
Such tools are becoming increasingly widespread in customer service, with some support teams also using customer simulations to train human agents in a safe environment before they start on the contact center floor, bringing them up to proficiency faster. The same principle applies here.
3. Conversational Analytics for Email (Generally Available)
Finally, in more of a brick-in-the-wall announcement, AWS shared that it is extending its conversational analytics solution - Amazon Connect Contact Lens - to email.
As a result, it’s available in all major AWS regions where Amazon Connect is.
That means contact center teams can reduce the amount of time they spend manually reviewing customer conversation transcriptions.
Additionally, they may automatically categorize each email, score every conversation, support compliance management, and more.
The Hot Take: AWS Is Focused on AI-Driven Outcomes
With these latest updates, AWS is effectively moving the conversation from 'AI experimentation' to 'AI-driven business outcomes,' which is something organizations have been demanding this year, according to Zeus Kerravala, Principal Analyst at ZK Research.
“By bridging the gap between automated deflection and proactive revenue generation, Amazon Connect is challenging the industry’s long-standing view of the contact center as a cost center,” he added.
However, Kerravala also stressed that the most critical evolution here isn't just the AI's ability to drive sales; it’s the new governance and observability layer.
“The introduction of a unified manager interface and expanded simulation capabilities addresses the ‘trust gap’ that has stalled enterprise AI adoption. Enterprises are no longer forced to choose between innovation and safety; they can now observe, manage, and validate human and machine performance in a single, predictable environment.”
“This is exactly the kind of maturity that enterprise leaders have been waiting for to move AI from the pilot stage into full-scale production,” concluded Kerravala.