This is an industry analysis of a public Zapier customer story about Rebrandly—not Tiercel Labs client work. Rebrandly’s support team used a documentation-grounded chatbot and automation to cut ticket volume while keeping escalations visible in Slack and Jira.
The constraint
Rebrandly supports hundreds of thousands of users with a lean team. Prior support relied on rigid decision-tree routing to static help articles. It did not scale with volume or with the variety of questions customers asked.
The useful move was not a general-purpose assistant. It was a chatbot wired into help documentation, with clear escalation when the bot could not finish the job.
What they shipped
According to Zapier’s write-up, the chatbot is grounded in Rebrandly’s help docs and handles 50–100 conversations per day. The team reports 90–95% response accuracy and more than 16,000 conversations resolved automatically since launch.
Ticket volume fell by about 50%. Agents could spend more time on higher-complexity work instead of repeating answers already covered in documentation.
Escalation stayed human
Enterprise and urgent paths were automated into Slack and Jira rather than left inside the chatbot. Logging alerts and high-priority tickets notify on-call channels; enterprise requests create Jira issues and Slack updates so nothing silent-fails.
That pattern matters: the model answers from docs; people and systems own exceptions.
Lesson for other teams
Ground the bot in owned documentation. Measure ticket deflection and accuracy, not vanity chat counts. Design Slack/Jira (or equivalent) paths before launch so enterprise customers never depend on a dead-end reply.
Again: this is Rebrandly’s public story via Zapier, not a Tiercel engagement.