Kaleb Jensen
Contact

001

Feature Request Matcher

Enterprise CSM proof.

Which clients asked for what just shipped.

The one-line pitch

I turned everyday client calls and daily release notes into an automated loop that tells me each morning which clients asked for what just shipped, with an email ready to send.

Kaleb Jensen, Enterprise Customer Success Manager at Connecteam. $1.76M portfolio across about 80 enterprise accounts.

Situation and task

Across about 80 enterprise accounts, clients ask for features constantly during day-to-day calls. Those asks are easy to lose, and when a feature finally ships, it is hard to remember which client asked for it. Meanwhile, multiple release note emails arrive every day. Without a system, the moment to go back and say "you asked, we delivered" slips by. My task: make sure no request gets lost and every client hears back when their feature ships.

Action

I spotted the gap and built the automation myself, with ChatGPT's help. It captures feature requests straight from my client calls, logs and tallies them by client in a Google Sheet, and each morning checks the latest release notes against every request on record. It tells me which clients asked for a feature that just shipped and drafts a short email so I can let them know.

How it works

Call to draft email

  1. 01

    Capture from calls

    The Attention AI API pulls my day-to-day client calls and extracts the feature requests mentioned in them.

  2. 02

    Log in one place

    Each request goes into a Google Sheet organized by client, who asked for it, and what the feature is.

  3. 03

    Match and tally

    New requests are checked against existing ones and matched up, keeping a running tally of how often each type of feature is requested.

  4. 04

    Morning release check

    Gmail is connected to ChatGPT through a scheduled action. Every morning it reviews the release note emails from the last 24 hours against all-time feature requests.

  5. 05

    Recommend and draft

    It recommends which clients asked for features that are now released and drafts a basic email explaining the release, so I know who to contact and can send a quick "the feature you asked for is live" note.

Stack

  • Attention AI APICall capture and extraction
  • ChatGPTBuilt the system and analysis
  • Google SheetsRequest log and tally
  • Gmail + ChatGPT scheduled actionDaily release check and drafts

Result

  • By the next morning after a release, I know exactly which clients to tell.
  • Closes the loop with clients who took the time to ask for something.
  • Scales proactive, personal outreach across a large enterprise book.
  • Draft outreach is ready to review and send, not written from scratch.
  • The request tally shows which asks come up most across accounts.

Illustrative mockups

Not live product shots

Mockup of a Google Sheet with columns for client, who asked, feature, request type, and status. Names are redacted.
Mock: Google Sheet columns (client / who asked / feature)
Mockup of a ChatGPT morning briefing that recommends which clients to contact after a release.
Mock: Morning ChatGPT recommendation
Mockup of a draft email saying the requested feature is live. The client name and address are blurred.
Mock: Sample draft email (blurred client names)

Next

Harden the request matching and reduce the manual steps before sending.

DownloadOne-pager PDF