Lyra Thinks Your Meeting Notetaker Should Actually Do the Follow-Up Work

The Macro: Meeting Transcription Was Just the Beginning

The meeting notetaker category had its moment. Otter.ai proved that real-time transcription worked. Fireflies built a business around searchable meeting archives. Fathom showed you could give away a great free product and grow virally. Grain made it easy to clip and share highlights. And then the market got crowded, fast.

By now, automatic transcription is table stakes. Every video conferencing platform has it built in. Zoom has AI Companion. Teams has Copilot. Even the standalone notetakers have mostly converged on the same feature set: join the call, transcribe, summarize, done.

But here is the thing nobody in the transcription wave solved: the work that comes after the meeting. A sales rep finishes a discovery call and still has to manually update the CRM, draft a follow-up email, create tasks in their project management tool, and brief their manager on what happened. The transcript sits in some archive. The summary gets skimmed once and forgotten. The actual operational work that the meeting generated still falls on the human.

This is the gap. Transcription solved the recording problem. Nobody solved the action problem. And for revenue teams specifically, the action problem is where all the time goes. Sales reps spend roughly 65% of their time on non-selling activities, according to Salesforce’s own research. A huge chunk of that is post-meeting administrative work: updating records, sending emails, scheduling next steps, documenting what was agreed to.

The question is whether you can build a meeting platform that does not just capture what was said but actually does the work that the conversation created. That is a fundamentally different product than a notetaker, even though it starts in the same place.

The Micro: Meetings That End With Outcomes, Not Notes

Lyra is a four-person team out of San Francisco building what they call an AI-native meeting platform. The founders are Courtne Marland (CEO) and Henry Kwon, and they came through Y Combinator’s Spring 2025 batch. The positioning is deliberate: this is not a notetaker. It is a workspace where meetings and work happen together.

The core product lets you run video calls with 4K individual microphone transcription, collaborate on documents in real time during the call, and have AI generate follow-up actions as the conversation unfolds. That last part is what separates Lyra from the transcription tools. It is not summarizing your meeting after it is over. It is detecting action items, commitments, and decisions in real time and turning them into executable work.

The automation layer is where this gets genuinely useful for sales teams. You can tell Lyra to update CRM stages, send follow-up emails, or trigger actions across your tool stack based on what was said in the conversation. So when a prospect says “send me the proposal by Friday,” Lyra does not just write that down. It can draft the email, attach the relevant materials, and queue it for review. The rep does not have to context-switch into their CRM and email client after every call.

Pre-meeting preparation is another smart feature. Lyra runs research on attendees before the call starts and drafts personalized talking points. If you are a sales rep doing five discovery calls a day, the time savings on prep alone are meaningful. Today, most reps either skip preparation entirely or spend 10 minutes Googling the prospect right before the call. Neither approach is great.

The workspace knowledge base ties it all together. Every conversation becomes part of a searchable repository with citations back to specific moments in specific calls. When a new rep joins the team, they can query the entire history of conversations with an account instead of asking their colleagues to recount what happened over the last six months.

New users get 1,000 free minutes, which is roughly 30 meetings. That is a generous trial for a product that needs several calls to demonstrate its value.

The Verdict

I think Lyra is right about the direction of this market. Transcription is commoditized. The value is moving downstream into action, context, and workflow automation. Otter and Fireflies will have a hard time pivoting from “we record your meetings” to “we do your post-meeting work,” because those are fundamentally different products that require different architectures.

The competitive challenge is real, though. Gong and Chorus (now part of ZoomInfo) already own the revenue intelligence space for larger sales teams. They have deep CRM integrations, coaching features, and years of customer data. Lyra needs to carve out the segment of teams that want meeting automation without the enterprise overhead and pricing of a Gong deployment.

The four-person team is lean for the scope of what they are building. Video calling, real-time transcription, document collaboration, CRM integration, email automation, knowledge management. That is a lot of product surface. Marland and Kwon will need to be ruthless about which features matter most in the first six months and which can wait.

What would convince me this is working: usage data on the automation features specifically. If teams are using Lyra primarily for transcription, it is just another notetaker. If teams are actively using the CRM updates, email drafting, and follow-up automation, then Lyra is a new category. The ratio between those two usage patterns will tell the whole story.