Sinch research: 60% of executives are confident their AI programs are succeeding. Only 43% of the teams delivering them agree

This reads more like a timing signal than a demand signal: enterprises are still struggling to operationalize AI in customer communications, so the near-term monetization curve for vendors tied to that workflow is likely to be shallower than sell-side narratives imply. The winners are platforms that can sit on top of existing systems with governance, observability, and compliance layers; pure model access or point solutions are more exposed to pilot purgatory and procurement delays.

For Sinch specifically, the second-order risk is not that AI is irrelevant, but that implementation friction pushes out booking conversion and expands services burden before product revenue can scale. That can compress gross margin in the next 1-3 quarters if management chases growth through custom deployments, while leaving long-duration investors disappointed on the speed of AI attach rate. Competitively, broader customer engagement stacks with distribution and workflow control have more pricing power than standalone CPaaS names if buyers want one throat to choke.

The contrarian read is that the confidence gap may actually be bullish for 6-18 months: once operational teams get budget authority and a few reference cases prove ROI, deployments can accelerate from experimentation to standardization. The thesis is falsified if Sinch can show measurable AI-related bookings, shorter sales cycles, or margin-neutral deployment economics in the next two reporting periods; absent that, this is mainly an alert that narrative is outrunning revenue.