How Planet Labs achieved 99.1% CSAT across +65,000 inquiries with Kapa
"Kapa is a great fit for a nuanced company like ours with a complex portfolio of technical products, a diverse user base, and international employees".
Mike Merit
Chief Customer Officer, Planet Labs
Challenge
Planet had already built a strong self-service library, a developer center, video training, and detailed support articles. The problem was speed and accessibility, not content. The team fielded repeat questions ("How do I log into Sentinel Hub?") that ate into time needed for serious customers, while 15% of queries arrived in another language and new hires took up to six months to reach full productivity.
Solution
Chief Customer Officer Mike Merit ran an internal contest: find a solution within four months, against three principles, eliminate repetitive work, get customers answers faster, and gather data to keep improving. The team's pitch was Kapa, and once decided, it took only a few days to deploy it in two places: a customer-facing widget on the website, and an internal knowledge tool for agents, CSMs, and new hires.
Results
65,000+ tehcnical customer inquiries handled per year
26% of support cases are auto-resolved in under 30 seconds
2x faster new hire ramp time (from 6 months to 3 months)
99.1% CSAT score
Won Gold at the 2025 Stevie Awards for Customer Service
Documentation Agent
Self-service answers, the moment a customer needs them
Deployed as a customer-facing widget, it gives Planet's 900+ customers instant, self-service answers instead of a ticket queue, drawing on the developer center, support articles, and video training already in place. Non-English questions, roughly 15% of volume, are handled with translations the international team validated as accurate, one of the first results that won the broader team over. "We started noticing it early with language," Merit says. "15% of our queries are in a different language, and it's really hard to translate sophisticated technical questions. Our international team validated that Kapa's translations were spot-on. That became one of the first things that drew the broader team in."
Play video
Support form agent
Resolving questions before a ticket is ever filed
At the support intake form, it reads the drafted question and surfaces an answer before submission, escalating cleanly to a human when it can't help. This deflects the repeat questions the knowledge base already covers, cutting ticket volume and freeing the team for the escalations that genuinely need a specialist. Tickets that do get filed arrive with more context, so they resolve faster too. "It doesn't take an hour to find the stuff. It maybe takes five minutes," Merit says. "But the difference between five minutes and 30 seconds is a big difference. And a customer submitting a ticket and waiting 30 minutes for a response, with our best-in-class response rate of 30 minutes, versus getting an answer in 30 seconds?"
Play video
MCP SErver
Grounded code for satellite and geospatial workflows
A hosted MCP endpoint feeds Planet's satellite and geospatial product knowledge to coding agents like Claude Code and Cursor as developers work, grounding them in Planet's APIs, data specifications, and imagery formats instead of general-purpose training data. For geospatial workflows, where the wrong API call or data assumption costs real debugging time, reasoning from Planet's own sources keeps generated code closer to correct on the first try, and gives developers building on Planet's data a faster path from idea to working integration.
Play video
Why Kapa
Both deployments draw on the same self-service content Planet had already invested in, so the assistant is only as good as the knowledge base, and improves right along with it. That is how a lean team handles 50,000+ inquiries a year (26% auto-resolved in under 30 seconds), sustains 99.1% CSAT, and won Gold at the 2025 Stevie Awards for Customer Service, all despite 33% annual revenue growth and without adding support headcount.
"Kapa is a great fit for a nuanced company like ours, with a complex portfolio of technical products, a diverse user base, and international employees," Merit says. "We use AI to protect our humanity, not replace it. The team is happier because they're doing the work they were hired to do."
