Freemium

Freemium became the default pricing model for product-led B2B and consumer SaaS, Slack, Notion, Figma, Canva, Dropbox, Spotify, Loom, Linear all use some form of it. The model works because the marginal cost of serving a free user is small, the free tier doubles as marketing, and word-of-mouth from free users brings in paid ones. But freemium also kills more startups than any other pricing decision when the upgrade math doesn't work. Choosing freemium is choosing a multi-year bet that conversion + virality will eventually outrun infrastructure cost.

Trial creates urgency; freemium creates patience. Different growth shapes, different tradeoffs.

Limit by capacity, not feature: Slack limits message history; Notion limits block count for teams. Capacity limits feel fair and scale with usage. Feature limits feel punitive.

Make the free tier actually useful: A crippled free tier produces neither conversions nor advocacy. Aim for "useful enough that 80% never upgrade, but those who hit the ceiling have already become advocates."

Aha moment first, gate second: Users must reach the value moment in the free tier. Gating before activation kills both conversion and word-of-mouth.

Visible upgrade triggers: When a user hits the ceiling, the upgrade prompt should be contextual ("you've used 5/5 seats; add another for $X") not interruptive popups.

Team or workspace gating: SaaS often gates collaboration features. The first user is free, but inviting teammates triggers paid.

Usage-based limits: API calls, storage, exports, build minutes, these scale naturally with value.

• Consumer freemium (Spotify, Dropbox): 2–5% free → paid is healthy.
• B2B SaaS freemium (Slack, Notion): 4–10% is typical for product-led businesses.
• Developer tools (GitHub, Vercel): 1–3% is normal but customer LTV is huge.
• Mobile games (free-to-play): <1% but whales drive revenue.

A 5% conversion rate on millions of free users is a great business; 5% on thousands is not.

Low marginal cost per free user: Hosting and serving the free tier doesn't bankrupt you.

Strong network effects: Free users make the product better for paid users (Figma, Slack, Loom).

Viral or word-of-mouth growth: Free users bring in paid users.

Clear upgrade triggers: Natural points where users want to pay (more storage, more seats, advanced features).

Long sales cycles otherwise: When traditional sales is too slow, free trials let buyers self-evaluate.

High cost per free user: Hosting, storage, support, model inference (a real issue for AI startups).

Free tier too generous: Nothing forces upgrade; conversion stays at 0.5%.

Free tier too stingy: Users churn before activation; word-of-mouth dies.

No clear ICP: Random free users don't reveal who actually pays.

Enterprise-only buyers: When the buyer is procurement, not the user, freemium doesn't help, sales still sells.

Reverse trial: Start with a 14-day full-feature trial, then drop to a permanent free tier with limits. Combines urgency and patience.

Free-but-paywall-on-results: Free to use, paid to export or share results.

Time-bounded freemium: A "forever free" tier that's clearly labeled as a beta program.

Team-only freemium: Free for individual users, paid the moment they invite a colleague.

Not measuring free → paid conversion by cohort: Without cohort analysis you can't tell if conversion is improving.

Treating free users as failures: They're marketing. Retention and engagement of free users matters too.

Pricing the paid tier wrong: A $9 Pro tier on a $50 LTV doesn't pencil. Price-to-LTV alignment is critical.

Pulling the rug: Removing free features after launch destroys trust faster than any other move.

Confusing freemium with free trial: They need different funnels and different metrics.