Intro: pricing is the product's second version#
SaaS pricing is a design decision, not a finance one. Which limit is paid, what counts as premium, how the price curve climbs — these directly shape user behavior. Wrong pricing creates a wrong product feeling.
This article lists the 6 decisions that build SaaS pricing: model selection, tier count, which limits to set, free tier yes/no, trial length, price-bump strategy. We give a concrete framework for each.
Core principle: pricing must come from market data, not intuition. The first version is a hypothesis; iterate via A/B testing. Teams attached to a fixed price lose value over time.
1. Model selection: per-seat / usage / feature / hybrid#
Per-seat: the most common model. Cost rises with user count. Pro: predictable, easy to understand. Con: customers saying "only one person will use it" resist growth. Slack, Notion, Asana use this.
Usage-based: billed on API calls, data volume, message count, etc. Pro: value = payment. Con: hard to forecast; customers fear surprise invoices. Stripe, Twilio, OpenAI use this.
Feature tier: tiers differ by feature set. Basic / Pro / Enterprise. Pro: clear. Con: "which tier has what?" confusion. Most B2B SaaS use this.
Hybrid: per-seat + usage limit + premium features. Most flexible. Like "5 users + 10K API calls + AI feature". Most mature SaaS evolve toward this combination.
Selection criterion: what is the product's value measured against? Per-seat for per-person value; usage for transaction value; tier for feature differentiation; hybrid for mixed.
2. Tier count: 3 is the magic number#
3 tiers is the SaaS industry standard. "Basic / Pro / Enterprise" or "Starter / Growth / Scale". Less feels like insufficient choice; more creates decision fatigue.
Display order: (a) leftmost = cheapest / fewest features (decoy effect), (b) middle = highlighted with "most popular" (the actual seller), (c) rightmost = enterprise (high price, customizable, sales-led).
Anchor effect: the most expensive tier makes others look cheap. "Enterprise: $999" makes "Pro: $99" look attractive. So even an exaggerated top tier serves a purpose.
Exceptions exist: early-stage / niche products may use a single tier (Linear started this way). Complex enterprise products may make custom quotes the main model.
3. Which limits to set? Tie value to payment#
Limit selection shapes user behavior. Wrong limits = low adoption or lost monetization. Right limits create the value-as-you-grow → pay-as-you-grow link.
Good limits (naturally crossed as the user grows): user count, project / workspace count, storage, AI request count, integration count. These metrics correlate with success.
Bad limits (blocking core usage): export count (creates "can't access my data" feeling), user count for solo (1 person on $50 is heavy), feature lockout ("button doesn't work"). These lose customers.
Soft vs hard limits: soft = warning + upgrade nudge ("approaching AI request limit, switch to Pro"). Hard = full block ("limit reached, can't proceed"). Soft converts better; hard only for expensive resources (AI compute, large files).
4. Free tier: yes or no, how generous?#
Three models: (a) no free tier (only trial), (b) limited free tier (1 user + 100 items), (c) generous "forever free" tier.
Free tier is good when: there's a network effect (Slack — once a team adopts, others join), viral growth is possible, the freemium → paid path is clear.
Free tier is bad when: per-user operational cost is high (each free user loses money), B2B enterprise sales (free users pollute the sales pipeline), niche high-value product (gets positioned as cheap).
Practical: 2-5% conversion is typical from freemium to paid. 10K free users ≈ 200-500 paid. If you can't hit that ratio, free tier is a cost center; switch to trial.
5. Trial length: 7 / 14 / 30#
14 days is the SaaS standard. Long enough for value, short enough that "I'll come back later" procrastination is rare.
7 days: creates urgency, raises conversion. Works for simple / fast-value products (if value is reachable within an hour of signup).
30 days: for complex products (CRM, ERP). Adoption is slow; value isn't visible without team-wide engagement. Trial extensions are rarely granted (encourages slack culture).
Credit card vs no-CC: requiring a CC reduces signups 50-70%, but 80% of trial completers convert. Without CC, trial counts are higher but paid conversion drops to 10-20%. Which gives more paid users? Test.
Reverse trial alternative: user starts with paid features; reverts to free tier when trial ends. Notion and Linear use this. Lifts conversion.
6. Price-bump strategy: planned, not ad-hoc#
Initial pricing wants to start low (adoption); but raising the price after 12 months is inevitable. Unplanned hikes cause churn; planned hikes are accepted as natural.
Good practices: (a) grandfathering — existing customers stay on the old price (loyalty reward), (b) new customers on new price, (c) 30-60 day advance notice (no surprises), (d) annual prepay discount (20% off for yearly — softens the bump).
Price-hike rule: 10-20% per year is typical. Higher = churn risk; lower = loses to inflation. The US market sometimes sees 15-25% per year (high-value SaaS).
Plan migration: shifting from monthly to annual (20% discount) is essentially a soft hike. "Annual customers pay 20% less" instead of "monthly customers pay 25% more". Most users move to annual; cash flow + LTV grow.
Conclusion: pricing is iterative, not static#
The first pricing model is a hypothesis. Ship to market, collect data, adjust. Pricing 6-12 months later may differ entirely from version 1; that's natural. Teams attached to a fixed pricing lose value.
Metrics to watch: ARPU (revenue per user), MRR (monthly recurring revenue), churn rate, expansion revenue (growth from existing customers), CAC payback period.
If you want to clarify your SaaS product's pricing structure, get in touch via our SaaS development page — we set up a structure based on market data + customer interviews + iterative testing.
Related services
Related articles
Other articles that support the same decision
SaaS
The First 7 Mistakes Teams Make in SaaS Architecture
Why SaaS products get expensive when tenant isolation, permissions, billing, and observability are not planned early.
SaaS
SaaS Development Roadmap 2026: 8-Phase Plan
From pre-MVP to enterprise, an 8-phase roadmap. Investment, metrics, team scale, hidden traps and typical duration per phase.
AI Tools
What is Claude AI? A Comprehensive 2026 Developer Guide
Comprehensive guide to Claude AI's 2026 capabilities — Claude Code, MCP, API integration, KVKK compliance, and enterprise case studies for software developers.
Next step
If you are planning a similar project, we can clarify the scope and shape the right proposal flow together.
Start a project request