I'm Florian, Head of Product at Flunter. My job is to decide what we build, and above all what we don't. In this article, I explain how sales reps' needs reach us, how we sort them and what we shipped in spring 2026 thanks to them. I also talk about what we refuse to build, because that is often where a cold calling tool shows what it is worth.
When people talk about sales productivity on the phone, they usually think about call volume first. What I see in the teams we work with is that time mostly slips away between calls: dialing, waiting, looking for a note, filling in the CRM. Taken one by one, these frictions last a few seconds. Added up over a day, they can get close to an hour. Here are the eight we see most often, a worked example to measure them and, for each one, the fix on the process side and on the tool side.
A calling session is often won or lost before the first call, when the list is prepared. Numbers without a country code, duplicates, switchboards mixed with direct lines, people who asked not to be called again: every dirty row costs time on the phone. In this article, I share what I see with the teams we work with at Flunter: the columns that matter, the right number format, cleaning duplicates and opt-outs, CSV pitfalls and importing from the CRM. The goal is simple: a list you can call without stopping every five minutes.
A calling session almost always ends the same way: you close the tab, move on to email, and forget what was said. That's a shame, because right after the session is when you learn fastest. In this article, I suggest a 15-minute debrief routine: two calls to listen back to, one pattern to look at, three numbers to check, a grid to score yourself and one decision for the next session. I also show how we designed Flunter so this debrief takes a quarter of an hour, not an hour.
A callback is a prospect who told you "not now" rather than "no". It is often the most promising call, and yet it is often the one we get wrong: the prospect remembers you, you no longer remember them. In this article, I share the method I see working in the teams we support: a 30-second checklist before you dial, opening lines that pick up the thread, and a follow-up routine that does not depend on your memory. I also explain how we built the Flunter contact record so that all this context fits on a single screen.
A sales call summary has one job: let anyone pick up the deal without listening to the call again. In most CRMs I see, that is far from the case: "interested, call back", and nothing else. AI summaries solve the typing problem, not always the content problem. Here is the structure we recommend at Flunter, a template to copy, a before/after example and what should actually go to your CRM.
MCP lets an AI assistant like Claude read from and write to your tools, instead of working only with what you paste into the chat. For a team that prospects by phone, that means pulling up a contact's history, preparing a call list or creating follow-ups in one sentence. Flunter publishes an MCP server built on its API. This article explains what MCP is in plain terms, what our server can do today, how to install it and where its limits are.