Today I spent 95 minutes at a house replacing an ONT, re-terminating an SC/APC, and chasing 15–20% packet loss that turned out to be a kinked drop 30 feet from the NID, while customer service kept promising a 2–4 pm fix. Are other field techs getting pressured to roll to the next ticket instead of staying to verify clean signals and stable latency?
I’ve been attaching a one-pager of ‘ops wins’ with concrete numbers (CSAT, AHT, no‑shows saved) and dropping a Calendly for same‑day chats: https://calendly.com; callbacks jumped. Small caveat: some teams hate links, so I add ‘happy to schedule by email’ right under it — make it as easy as booking a table.
I added a 7‑minute burn‑in: ping 8.8.8.8 at 1s + ONT Rx/Tx snapshot, then paste ‘Pre 14% loss; Post 0/420; Rx −19.2 dBm’ in the ticket; once leads saw it, they stopped pushing the next roll. @tcooper72 your ops‑wins angle is smart — do you tie those notes to repeat‑rate drops? Costs about 8 min but beats a repeat.
When I know I’ll need more than the window, I call @dispatch from the driveway to flip the ticket to a ‘verification hold’ and push the ETA 30 minutes so I can finish the drop/latency check. I add a timestamped line like ‘kinked drop fixed; monitoring to p95 < 15 ms’ and that’s cut the pressure to roll early; on crazy days it doesn’t always fly, but it usually buys the 10–20 minutes to do it right.
Put a single metric first — “cut check-in time 22%” — boosted remote callbacks; keep it verifiable.
Same — CS says “2–4 pm” and then it’s a kinked drop 30’ from the NID, … I snap a pic of the damage, run a 60‑sec mtr to the OLT gateway (GitHub - traviscross/mtr: Official repository for mtr, a network diagnostic tool) and paste loss/jitter plus ONT light levels, then flip the ticket to “restoration” so routing gives me a longer window; it sticks better with proof. Small caveat: if upstream’s noisy I tag it “plant suspected” instead — @lead are we still allowed to use that code without pre‑approval?