Internal AI · Business teams
Analytics agent for business teams
The problem
The people who needed revenue and funnel numbers every morning couldn't write SQL, so they asked whoever knew the warehouse, and answers took a few hours. Two people asking the same question in slightly different words would get different numbers, and then it was unclear which one was correct.
What I built
I built two lanes, which was more work to set up.
The numbers that need to be the same for everyone are computed by the same scheduled SQL, at the same time, every day, and sent out as a digest. If two people screenshot the morning message, the figures match.
Non-technical staff ask questions in plain language over Telegram. The agent has the schema and enough business context to answer follow-up questions, e.g. why a number moved, or what to look at next.
The reason to keep them apart is that an agent will eventually invent a number. If people plan around that number, they stop trusting the digest as well. I did not want the digest to grow into a long report covering every possible question.
What changed
INR 3L of additional revenue in the first 15 days. People who used to wait on an analyst now work off the same day's numbers.
What I'd do differently
I would show the SQL behind every agent answer from the start. In hindsight, the digest was trusted quickly because it was the same query every morning. The agent took much longer, mostly because people couldn't see how it arrived at an answer.