Recently, Talkdesk held its industry analyst summit in Savannah, GA. I thought this was a fitting venue for the event as Savannah is known for many firsts.
Savannah was the first city in the US to be a planned community, and Savannah Golf Club, which opened in 1792, is known as the first American golf course. Also, in 1819 the S.S. Savannah became the first steamship to cross the Atlantic, sailing from one of the first seaports built and the Girl Scouts of USA were formed in 1912 there.
This was also the first time I had heard the CEO of a Gartner MQ leading vendor come out and call the market in which his company plays a role in, “dead.”
But that’s exactly what happened when Talkdesk’s Tiago Paiva called for the end of CCaaS, which will give way to Customer Experience Automation.
Now to be clear, Paiva was being a bit provocative and isn’t calling for CCaaS to go away but rather for that functionality to become a standardized set of capabilities that are part of CXA.
A good analogy is to consider how there is no longer a “video meetings” or “chat” market, as they are part of the UC suite.
Similarly, while companies will offer CCaaS capabilities, the core functions are becoming standardized, and differentiating oneself in the contact center space will be increasingly difficult.
Enter Customer Experience Automation
The first step in understanding Paiva’s thesis is to define what CXA is.
As the name suggests, CXA is about automating customer experience to deliver an experience that customers actually like. The reality is, for all the chatter of cloud this and AI that, in general, the experience we get from most brands leaves a lot to be desired.
The change signals Talkdesk’s recognition that in the age of agentic AI, the value proposition is moving away from managing human-agent interactions toward automating entire end-to-end customer journeys.
This doesn’t mean the end of agents, but Paiva was up front that the number of human agents will likely decline due to more interactions being completed by machines.
A good way to think about this pivot is pre-CXA, Talkdesk was a contact center provider that used AI to make its products better. With the shift to CXA, it’s an AI company that happens to have a contact center stack to fuel its AI.
While the industry has tossed the platform word around for years, Talkdesk’s CXA is a true platform built on agentic AI capabilities that orchestrates autonomous AI agents to manage complex workflows across the front and back office and deliver measurable outcomes.
At the event, I sat down with Pedro Andrade, VP of AI for Talkdesk, and he outlined a four-part, continuous cycle that defines the CXA platform:
- Discover: Mining and analyzing interaction data to identify pain points and high-value automation opportunities.
- Build: Using low-code/no-code tools to create AI agents and workflows based on the discovered data.
- Orchestrate: Running the multi-agent system to automate customer interactions, agent assistance, and post-call quality analysis across the entire customer journey.
- Measure: Tracking the impact of automations to continuously optimize the experience and translate improvements into measurable business outcomes.
CXA Is a Platform for the Entire Enterprise, Not Just the Contact Center
Perhaps the most significant aspect of the CXA platform is its independence from Talkdesk’s underlying CCaaS stack.
Andrade emphasized that the CXA platform is designed to run on top of existing customer service infrastructure, a critical differentiator for enterprises:
Platform Agnosticism
CXA can be deployed on competing CCaaS platforms – even on-premises solutions from vendors like Avaya and Genesys.
Modernization Without Disruption
This capability allows customers to leverage the benefits of leading AI automation without undergoing a massive, time-consuming “rip and replace” of their core contact center.
This addresses the urgent need for optimization now, rather than waiting months or years for a full cloud migration. Most enterprises are risk-averse when it comes to the contact center, and Talkdesk’s approach de-risks the adoption of AI.
The ability to plug into any contact center and automate cross-organizational processes fundamentally changes the scope of Talkdesk’s service, making it a high-level customer experience AI tool rather than just a service and support tool.
The Go-to-Market Strategy: Partnership, Co-Development, and Vertical Specialization
The shift to CXA requires a dramatic shift in Talkdesk’s approach to customer engagement and go-to-market strategy.
Andrade insists that deploying AI successfully is not a “turn on capability” but a motion of digital transformation requiring deep partnership.
1. Co-Development With Customers
Talkdesk’s new model centers on co-development with the customer, which involves the following:
- Working alongside the customer to identify key use cases that deliver value quickly.
- Iterating with speed to prove value and ROI.
- Reducing the risk and accelerating the time-to-market for AI-driven projects.
This deep engagement is required because AI is a tool whose value is entirely dependent on the specific use case and the goals an organization is trying to achieve.
2. The Shift in Buyer Persona
Because CXA addresses high-level process optimization and modernization, the conversation has moved up the organizational chart. Talkdesk is increasingly selling to CIOs, CTOs, and CEOs, who are tasked with driving innovation through AI.
The challenge is not budget but building trust in the technology’s ability to deliver a guaranteed ROI. The new, partnership-focused strategy is designed to build that trust.
It’s worth noting that the C-Level is not an audience that the CCaaS providers typically interact with. I asked Talkdesk’s CRO, Al Caravelli and the company’s SVP of America, Jen McDonald, about this, and both reiterated that the sales team has already undergone a major shift in talent to bring in people comfortable talking to this audience.
Caravelli also stated it’s taking a close look at the channel it uses to drive more C-Level engagement through them.
3. Vertical Specialization
Talkdesk has always served key verticals like Financial Services (FSI), Healthcare, and Retail, but under the CXA model, this specialization has deepened significantly.
This is not simply about having a sales force for different sectors; it’s about embedding industry-specific intelligence directly into the AI:
- Pre-built and Pre-trained Solutions: Talkdesk brings pre-tested, pre-built solutions to the table, reducing the customer’s risk and accelerating success.
- Specialized AI Models: Andrade confirmed that Talkdesk has invested in building or fine-tuning language models for specific industries. For instance, the AI for FSI will focus on extracting and reasoning about volumes, amounts, and percentages, while Healthcare AI will prioritize details like diagnosis, symptoms, and correlated effects. This blend of the right people (industry experts) and the right technology (verticalized AI) creates a high competitive moat.
A Necessary but Challenging Shift
As an industry watcher, it was good to see Talkdesk make this shift.
Although it’s a Gartner MQ Leader, the reality is that the company is a minority shareholder, and it would be next to impossible to grow that share significantly going to market with the same messaging as the larger vendors.
The only way to move from small player to market leader is to redefine the industry; that’s what Talkdesk is attempting to do.
Laying down the vision and challenging the industry is the easy part. The challenge for Talkdesk is having the strength to stick to the mission even when customers and partners may disagree.
Also, there is no MQ or Wave yet for “CXA,” so Talkdesk might find itself in contradiction to the very document that’s naming it a leader.
Legacy mindsets are hard to change, but customer experience will never deliver on the promise of AI if vendors themselves don’t change.
I’ve long called for the end of CCaaS, and I’m glad to see Talkdesk look at the problem they are solving, not through the lens of their current product, but by rethinking the problem and then building a product that addresses that concern – even if that means disrupting its own core product.
Very rarely are vendors willing to disrupt themselves, which is why market transitions generally lead to new leaders.
