Enterprise CX teams are under pressure to treat AI as a platform reset. Yet the continued growth of NICE and Genesys suggests many CX platform buyers are taking a more pragmatic route: extending trusted systems before replacing them.
Adrian Swinscoe, Author, Blogger and Podcaster at Punk CX, told CX Today that the “SaaSpocalypse” narrative risks assuming that major enterprise software vendors will stand still while AI-native challengers move around them. In his view, that is not what the market is showing.
NICE reported 8% year-over-year revenue growth in the third quarter of 2026 and said it exceeded its revenue and earnings-per-share guidance ranges. Genesys, meanwhile, said earlier this year that fiscal 2026 revenue was nearly $3 billion, with Genesys Cloud annual recurring revenue nearing $2.6 billion and growing more than 35% year over year.
Those figures do not settle the long-term question of who will own AI-powered customer orchestration. They do, however, show why the next enterprise buying decision may be less about replacing incumbents and more about whether established platforms can absorb AI safely, economically, and at scale.
CX Platform Buyers Are Still Choosing Familiar Ground
Swinscoe argued that the most durable CX platforms are responding to disruption by adding AI capability, acquiring specialist technology, and expanding inside existing enterprise accounts.
This is important for buyers, because a new AI-native tool may promise speed, specialization, or lower friction, but enterprise procurement rarely starts from a blank sheet of paper. Data security checks, approvals, integrations, operational dependencies, and user familiarity all shape the real cost of switching.
Swinscoe said that context helps explain why existing providers can retain buyer attention even as AI-native rivals attract market excitement:
“If you're broadly happy with where their platforms at, you're buying extra from them. That feels like an easier and simpler sort of sale rather than going, ah, I'm now going to pick one from a thousand kind of these new apps, new ventures, new startups.”
That does not mean incumbents get a free pass. Swinscoe also pointed to acquisitions, including NICE’s acquisition of Cognigy, as evidence that large vendors are buying capabilities as well as building them. For enterprise buyers, that creates a second question: whether the acquired capability becomes a coherent part of the platform or another layer to govern.
AI Pricing Is Getting Harder to Forecast
The pricing model behind contact center technology is also shifting. Seat-based or license-based models are giving way to usage, automation, and outcome-based structures, particularly as AI handles more customer interactions.
On paper, outcome-based pricing should align technology spend with business value. In practice, Swinscoe warned that the definition of an “outcome” can become difficult to pin down, especially when AI systems rely on multiple agents, inference costs, monitoring, orchestration, and human oversight.
That is where the economics of AI can become uncomfortable. A contact center may reduce some headcount or automate a large portion of tier-one queries, but still add licensing costs, model usage costs, specialist talent, governance work, and managed-service spend. Swinscoe described the cost picture as unresolved:
