Having spent the last several years deeply entrenched in the observability and APM space, where pricing models are constantly scrutinized for value alignment, I find Fireflies.ai's current per-seat subscription structure to be fundamentally misaligned with its core utility and the modern, asynchronous meeting culture it aims to serve. In an ecosystem where tools like Datadog charge per host or per million log events, and synthetic monitoring tools price based on check frequency and locations, the rigid per-user model feels like an artifact from a bygone era of enterprise software.
The primary issue is that meeting intelligence, unlike a seat in a project management tool, is not a resource consumed uniformly across an organization. Consider two scenarios within the same company:
* A sales team of five individuals, each conducting 15 hours of external client calls per week, generating dense, action-packed transcripts requiring analysis and CRM logging.
* A department of twenty engineers, where perhaps only two engineers attend cross-functional stakeholder meetings requiring transcription, totaling maybe 5 hours per week.
Under a per-seat model, both groups incur cost based on headcount, not usage. This creates a perverse incentive to limit access to the tool, directly counteracting its collaborative benefits. Team members become "second-class citizens" without access, or admins are forced to constantly rotate credentials, which is a security and workflow nightmare.
A more equitable and modern pricing framework would decouple cost from simple user count and tie it to actual consumption or capability. Two models immediately present themselves as superior:
* **Per Meeting Hour (Tiered):** This is the most logical and transparent model. It directly correlates cost to value derived. Tiers could be structured like:
* Base tier: 50 meeting hours/month, includes core transcription and search.
* Pro tier: 300 meeting hours/month, adds AI summaries, action item detection, and integrations.
* Enterprise tier: 1000+ meeting hours/month, includes advanced analytics, custom vocabularies, and SOC2 compliance.
This allows an entire department to share a pool of hours, promoting widespread adoption and collaboration without punitive linear cost scaling.
* **Per Feature Module (Unbundled):** Alternatively, a platform core could be offered at a low flat rate (covering transcription and basic playback), with add-on modules priced separately. For example:
* Core Transcription Engine: $X/month for the first N hours.
* AI Summary & Insight Module: +$Y/month.
* Advanced Integration Pack (CRM, ticketing systems): +$Z/month.
* Administrator & Team Analytics Dashboard: +$A/month.
This approach, common in observability platforms, lets teams pay only for the capabilities they need. The engineering team in our earlier example might only need the core engine, while the sales team activates all modules.
The current seat-based model is a significant barrier to organic, organization-wide adoption. It fails to account for the sporadic nature of meeting attendance and the collaborative essence of the tool itself. For Fireflies.ai to scale effectively and become as ubiquitous as it deserves to be, a shift to a consumption or feature-based model is not just preferable—it's an operational necessity. The parallels to the evolution of pricing in the observability industry are too stark to ignore.
— Billy
So you want to apply APM logic to a meeting tool. The problem is, "per meeting hour" just swaps one bad metric for another. It incentivizes shorter, less useful meetings, or worse, turning off the recorder to save budget during the crucial heated debate. You'd end up gaming the system instead of getting work done. Pricing is always going to feel misaligned somewhere.
Trust but verify.