How to keep a #product-feedback Slack channel from becoming a graveyard
Every company creates a #product-feedback channel, and the lifecycle is always the same. Month one: sales and support post enthusiastically. Month three: messages get a 👀 from a PM and nothing else. Month six: people stop posting, because posting there is where feedback goes to die.
The channel doesn't fail because people stop caring. It fails because Slack is a great capture surface and a terrible database — messages scroll away, nothing is counted, and posters can't tell whether anyone read them. Keeping the channel alive means fixing those three things, not writing a sternly worded channel topic.
#shared-customer-channel
Ranked topicsStep 1: make the posting format light, not bureaucratic
Some teams respond to channel chaos with an intake form via Workflow Builder — Slack's own feedback-tracker guidance leans this way. Forms produce tidy data and quietly suppress volume: a salesperson between calls will not fill out six fields.
A better trade for most teams is a one-line convention in the channel topic:
Post: what the customer said (quote if you can), who they are, link to the conversation. One item per message — thread the discussion.
So a good post looks like:
"Honestly the only reason we haven't rolled this out wider is there's no SSO." — Priya, IT lead @ Northwind (renewal in Q2). From today's call, recording: gong.io/call/8842
One item per message matters more than any other rule, because reactions and downstream tooling operate per-message. A five-request dump in one message can't be triaged.
Step 2: acknowledge every post within a day
Posters abandon channels when posting feels like shouting into a void. The cheapest fix is an emoji protocol, applied by whoever owns triage that week:
- 👀 = seen
- 📥 = filed to the tracker
- 🔁 = duplicate, count incremented
- 🚢 = shipped (the best one to hand out)
This costs seconds and is the single highest-leverage habit in this guide. A salesperson who gets a 🔁 with a thread reply — "that's the fifth SSO ask, it's moving up the list" — posts again next week. One who gets silence doesn't.
Step 3: sweep weekly, and empty the channel into a real system
The channel is an inbox, not the record. Once a week, someone sweeps everything since the last sweep: classify it, check it against existing themes, and file it where feedback is actually counted — your feedback queue or tracker, with the Slack permalink and requester attached.
Savio's guide to Slack feedback makes the core point well: feedback that stays in Slack can't be ranked against everything else. The channel's job is capture; the queue's job is memory. If you don't yet have a queue, how to centralize customer feedback covers setting one up.
Automation removes the sweep entirely. This is what we build — Modem reads connected Slack channels, extracts the feedback (including requests buried in ambient conversation nobody formally posted), dedupes it against everything from support and calls, and files it with the requester attached. The filing lands in a context graph rather than a flat list, so a request raised in the channel and again in a support ticket stays one topic with two receipts. We build Modem, so weigh that accordingly; an emoji-reaction workflow that pipes messages into a form or tracker is the manual alternative, and at a few posts a week it's enough. Other tools in this lane are compared in best tools to mine customer feedback in Slack.
Step 4: report back into the channel
The sweep should produce a short weekly post back into #product-feedback:
This week: 14 items filed. Movers: SSO (7 total, +2), CSV export (6, +1), Slack notifications (3, new). Shipped from this channel: bulk archive — thanks @dana for the original report.
This is the channel's heartbeat. It proves the pipeline exists, teaches people what a countable request looks like, and gives the team a shared sense of what's rising.
Step 5: close loops publicly
When something posted in the channel ships, go back to the original message, react 🚢, and reply in-thread tagging the poster: "shipped in today's release — tell Priya." Public loop-closing is what turns the channel from a complaint drop into a system people trust; it's the same principle as closing the loop with customers, applied internally first.
What kills these channels, for reference
- Feedback posted with no requester attached ("a customer wants X") — uncountable, unfollowupable.
- Debating priority in the channel instead of threading — scares off casual posters.
- A form so heavy people DM the PM instead.
- Silence. Always silence.
The smallest version you can start this week
Pin the one-line format, adopt the four emoji, and put a 20-minute Friday sweep on one person's calendar with a two-line report-back. That's the whole system. If the channel's already a graveyard, don't archive it — post the first weekly report and watch it come back.
