Seat-based pricing (per-user) is one of the worst ways to run a SaaS business, but this one is next-level.
Software as a Service will not “disappear” overnight due to LLMs or Lovables and Replits. And large CRMs or PM systems will not evaporate either.
However.
The larger the disparity between customer needs and SaaS offerings, and the gap between consumers and an existing business (and the competitive landscape), the more fuel poured into the fire.
DevriX uses a home-brewed time tracking system. We made this decision a decade ago. It’s paid off ages ago.
I use an accounting and FP&A Electron app that is custom and superior to market leaders.
I’ve met dozens of teams, from startups to large enterprise, rebuilding project management or CRM systems for different needs. In fact, even as a HubSpot partner on the DevriX side, one of my firms manages WP-CRM, a CRM plugin for WordPress with many hundreds of active licenses.
What most critics see in a post about “rebuilding Slack or HubSpot in a day” is failing to replicate the full SaaS suite of features. This is correct: ACL, user management, permissions, scalability, security for 3rd parties, and hundreds of other layers are missing.
But does a team of 40 need an app that supports 100,000 users in the open web… Or a VPN-gated system hosted locally for the in-house team?
This is the pivot point where custom rebuilds, especially in the “vibe” era of AI, are getting more competitive.
Seat-based pricing has been unfair to managers who log in once a month or finance teams who only need to download an invoice. That’s never made any sense whatsoever.
And billing over half a dozen spare licenses because of a company policy is another primer on brand policy that’s going against core business principles and definitely not favoring customers.
Again.
It’s not about the final dollar amount. It’s a contractual and relationship matter.
And in business, especially product-led growth freemium startups cannot afford to behave this way when the bubble popped 3 years ago and is still deflating.
