FollowUp Local: a Gemma model on your laptop that tells you why the customer went quiet
Description and use
FollowUp Local is a local tool that uses a Gemma model to analyze chat sales conversations, identify why a customer went quiet, and draft a tailored follow‑up, keeping all data on the seller’s machine.
Problem
Small sellers lose deals when customers go silent after saying they will check with someone, but no tool helps them understand why or how to follow up.
Outcome
The tool runs on a CPU‑only laptop, delivering a 4B model inference in ~19 s and a 1B model in ~6–9 s, with no data leaving the machine.
Workflow
- Paste the chat into the UI or CLI
- The local Gemma model (via Ollama) processes the text
- It outputs structured data: stage, real concern, evidence, next message and timing
- The seller reviews the draft and sends it manually
Evidence
Nothing leaves your machine. No API key, no account, no per-message cost.
I tried two sizes on the same CPU-only machine (8 vCPUs, no GPU): gemma3:1b ~6–9 s, gemma3:4b ~15–24 s
The model is asked to copy one customer line as evidence. If the quote it returns isn't actually in the chat, I replace it with the last real customer line.
- Record date: 2026-10-15T03:41:05.573Z