Skip to content
ClaudeClaude

How Slack uses Claude for AI search and summaries

Slack processes billions of messages a day. Claude powers the AI features that help users find what they need: search that answers natural-language questions, conversation summaries, and daily recaps. In this video, Slack's team shares how they went from early experiments with Claude to shipping features used across the platform. Read the full customer story: claude.com/customers/slack

Mar 9, 20261mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Slack integrates Claude to reduce noise with smarter search summaries

  1. Slack targeted information overload by using Claude to deliver better in-app search answers and summaries.
  2. Early experiments with Claude produced strong results quickly, creating an “instant fit” for Slack’s search use cases.
  3. Slack tracks outcomes like query success rate and user-perceived noisiness, and reports major improvements in both.
  4. Beyond customer-facing features, Slack uses Claude Code internally to fix bugs and accelerate engineering workflows.
  5. The shift enables engineers to spend more time on planning and architecture, with creativity becoming the main constraint.

IDEAS WORTH REMEMBERING

5 ideas

Slack’s AI effort is anchored in reducing information overload.

They frame the problem as “distilling through the noise” so users can find what matters, aligning with Slack’s product principle of being a great host.

Search and summarization were the initial, clearly defined AI targets.

Rather than adding AI broadly, Slack focused on two high-impact workflows: answering search queries and summarizing large volumes of content.

Claude delivered strong early performance that validated the direction.

The team describes an immediate “holy cow” moment when search answers came back, suggesting rapid proof-of-value during experimentation.

Slack measures AI impact with both behavioral and sentiment metrics.

They track query success rate alongside the number of users who self-report Slack as “noisy,” and claim significant improvements on both.

AI features translate into daily time savings for users.

Automated answering and summarization are positioned as giving users “minutes every single day,” emphasizing cumulative productivity gains.

WORDS WORTH SAVING

5 quotes

Some of the challenges we faced within Slack is traditionally information overload.

— Unknown

When we started to get our first search answers back, we saw like an immediate almost like mind meld with Claude. That was our first holy cow moment.

— Unknown

The system automatically takes care of answering things for me, summarizes information for me, and that just gives me minutes every single day.

— Unknown

The limitation now is creativity and imagination. It's not really a question of how to build the thing. It's a question of what you wanna build.

— Unknown

This is the most excited I've been to be in technology in, in my entire career.

— Unknown

Information overload and “be a great host” principleAI-powered search answers in SlackHigh-volume summarizationMetrics: query success rate and perceived noisinessClaude Code for internal engineering productivityShift from coding to planning/architectureTime savings and user productivity impact

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome