video1.0<iframe src="https://www.loom.com/embed/227c87f4d8ca4bdf900e43a0acb5bfee" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/os/loom-media-content/loom/83c409c2-80f4-498a-ac49-9ce2c7e9b5c0/5ddff2c6-a97d-4634-98b4-ccfca111e308/sessions-thumbnails-227c87f4d8ca4bdf900e43a0acb5bfee-74acfc811ffc2f1f.gif/binary?authToken=eyJraWQiOiJtaWNyb3MtZ3JvdXAvdGRwLW9zLWFyY2hldHlwZS85Z2ZkM3NvZnVsYWNyYzRxIiwiYWxnIjoiUlMyNTYifQ.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.POE9Y7CiU01GIcP-fc1_PHSBEoELRMVeMglazmufbr7eaeDkMV_czT7ZKZ4OKd9XjHP5F72RkiCGBGeOgsATuE0_BcnxQkHGuPuhDkb4zCnb6_kPPbrgnF9bhQcaHD5CJCYbe7sB-VZtngSSYadr_XPVo20u53zbXN_gEPe-7X5s_mPegh6rUsfmyJ_SAKrgsw-rKDFKM2m24TaQRGJ5VaBr32d8uxGd4AIAjyADMjjOhiyfYvkDusmz1xVQ5Wj7guFKLwDcEmwQHncoA4J5JCOp3AA8AYqvTul2sE0jgh0w4xYLwgGBo0_fAbGzi1DqM2ru8u18CnB151_aJ7bkdg&Expires=1790610343&Signature=YojC34JoRfnvTHg%7EHlaG46Rj3jC41OKEVKmqMVp4Upz7%7ESwE8dio8pt0oUeUvsxk5GSP6cPRbMxJDO4WbCeYk3f5RVFZDiwTj2cwGG%7E1g7MU0fiMD6tOQHZ8UpfxX5MSIpJTcQydmmySDtp41lqKfP99HWsxqBM9VbtokQm-O%7E7Azzj7%7Eo0DtNaDKjrEm9bWii7iUffAkdS8k9g%7ECFLZgKp0nf9P7HTShnnNF3MHYTRhVMn40D5lGIQsuZpeMm5CZnKeBXsRo6FsCzB9nbZciXmI9TNqVJRyKAbhAnbCDZDgMI9%7EKsyw9jA%7ERFfLphawTOEGaCP0fd2jOUOgRyWGVQ__&Key-Pair-Id=KQOSYIR44AIC0614.173333This Loom explains how to fully automate customer call quality assurance and voice analytics using an always-on workflow. It schedules a trigger to check Google Sheets for new unprocessed calls, waits for transcription readiness, converts each conversation into a transcript, then uses an LLM to extract sentiment and a category (such as cake quality, delivery issues, or pricing) before saving results to an internal database and marking rows as done. It also computes daily or date-range metrics like total call volume, count of negative calls, and the exact negative call percentage. Finally, it sets up an interactive chat agent that can answer management questions such as how many negative calls were received last week within about 2 seconds using a tool-driven subworkflow and an executive summary response.