IntouchCX's claim that Superpunch Roleplay lifted CSAT by 7% deserves scrutiny precisely because the supporting evidence is incomplete. The case study describes a major US retailer replacing static scripts with AI-driven customer personas that adapt during conversations, and the product has earned a 2026 AI Excellence Award from the Business Intelligence Group. However, the award announcement uses boilerplate language identical to six unrelated winners, and the published results lack cohort sizes, baseline scores, statistical testing, and program costs—the data points any CX leader needs to calculate actual ROI. The strongest evidence comes from a January 2026 retailer deployment showing an 18.9% CSAT advantage in week two that narrowed to 3% across eight weeks, suggesting real but diminishing returns. Yet IntouchCX hasn't reconciled how this connects to the May summary claiming 7% first-month improvement, and a separate TITAN Business Awards submission reports entirely different figures (18% CSAT, 55% faster time-to-proficiency) that don't align with either case study. For teams already running Zendesk or Salesforce Service Cloud, the question isn't whether roleplay training works in principle—it does—but whether IntouchCX's specific implementation justifies the configuration, scenario-writing, and trainer overhead against your existing onboarding infrastructure.
Adoption remains the hidden variable. The March 2026 pilot doubled usage from 9% to 18%, but 82% of eligible agents never engaged with Roleplay, and the self-selection problem is real: agents motivated enough to practice unprompted may already have been your strongest performers. Sezzle's parallel AI push shows what happens when automation scales across the entire operation—68% chatbot deflection, 88% AI-developed code, 98% automated ticket grading—but Sezzle is embedding AI into product development and core business logic, not just training. For contact centre leaders, the implication is clear: a training tool only generates ROI if it connects repeated practice to live performance metrics (CSAT, QA, AHT, transfers, repeat contacts) and if adoption reaches critical mass. IntouchCX's omnichannel positioning is undermined by evidence that focuses almost entirely on voice; buyers should demand proof that scenarios can replicate chat, email, and social interactions using your own policies and customer behaviour patterns before committing. The sensible move is a 60- to 90-day pilot with defined baselines, usage targets, and a cost-per-proficient-agent calculation—not a full deployment based on a 7% headline that lacks the methodological transparency to support it.
IntouchCX Digital CX Says AI Roleplay Lifted CSAT by 7%. Should Companies Be Investing in Superpunch? CX Today
Sezzle AI Push: Can Automation Boost Service and Productivity? TradingView