My Client Used to Scramble Through Notebooks Before Every Call. Now He Walks In Prepared in 10 Seconds.
Same client, few weeks later. The price monitoring system is running. He's happy. Then he says something that opens up a whole new project:
"I called a supplier yesterday and couldn't remember what we talked about last week. I made the same offer twice. It was embarrassing."
His "CRM" was a physical notebook and KakaoTalk threads. He'd scroll back through messages trying to piece together what he discussed with who. Sometimes he'd find it. Sometimes he wouldn't. And when he couldn't, he'd walk into a negotiation blind.
For a steel trader, relationships are everything. His entire business runs on knowing who has what inventory, who's flexible on price, who needs to move product by end of month. Forgetting a detail doesn't just waste time — it costs trust.
The Real Problem
He was already recording his calls (just a habit). But he never listened back because who has time to replay a 20-minute call to find one detail?
So I built him something simple:
He uploads a call recording. Within minutes, he has a full transcript, a summary of what was discussed, and a list of follow-up tasks with due dates.
But the real value isn't the transcript. It's the context.
Every time he opens a contact's page before making a call, he sees: last 5 interactions summarized, what was promised, what's pending, what prices were discussed. The AI reads all prior conversations and builds a briefing.
He went from spending 10 minutes scrambling through notebooks to being fully prepared in 10 seconds.
What's Next: Real-Time Recommendations
We've already started talking about the next step: real-time recommendations during calls. Imagine he's on the phone with a buyer who wants 304 coil. The system could surface: "We have 3 items matching this spec in the current auction, scored as good deals. Similar product sold for a certain price last week." He wouldn't need to check anything. He could focus entirely on the relationship while the system feeds him the data.
That's where we're heading. And it started because I asked one question: "What do you do right before you pick up the phone?"
The Takeaway
Don't start with the technology. Start with the moment in your client's day where they feel unprepared, overwhelmed, or like they're flying blind. That's your project. The AI is just how you solve it.