Back to Insights
How SaaS Services Raise Prices

How SaaS Services Raise Prices

7 min read

A more flexible Pro plan for modern teams - Vercel

What options does Gamma offer for teams? | Gamma Help Center

Recently, Vercel and Gamma raised their prices by about $5 per month. This is a 25% increase from the existing cost.

Similarly, Notion is also raising its prices. Adobe, of course, as well. This is partly because features have been added and there are no alternatives, but it is also implemented when certain conditions are met.

When prices go up, naturally, customers decrease. In the short term, they almost unconditionally decrease. But despite this, SaaS services raise prices. If you don't know these conditions, you have no choice but to repeat foolish mistakes in pricing policy.

Recently, while helping with various SaaS or Product work, I was surprised to find that there are very, very many people who do not know why they do this. I even saw an approach where, when asked for the reason, they said, "Since the competitor is this much, we are this much cheaper."

It's so obvious that I wonder if I even need to tell you, but how on earth do you calculate prices?

Cost-Plus Type

It's the simplest. It's a method of stacking the price with Cost (Cloud/Labor/Third-party API) + Target Margin. It is a method borrowed from manufacturing and distribution industries, but it is difficult to say it is a good method in the digital market where intangible value is sold.

The advantage is that calculation is easy and the finance team and sales organization feel relieved. The disadvantage is that it is disconnected from customer value. Customers don't care about our costs. They open their wallets for the 'value of problem-solving'. Therefore, Cost-Plus is only a minimum defense line and cannot determine the price ceiling.

  • When to use: Very early stage (when cost uncertainty is high), Reselling structure (passing on external API unit costs), B2B where legacy quotation practices are strong.
  • Caution: There is no need to automatically lower the price even if costs go down. Value is an independent variable.

Competition-Indexed Type

"The competitor is $30, so we are $25." Very common but dangerous. If the opponent's strategy, cost, and segment are different from ours, we get tied to the wrong anchor. This is burdensome when changing pricing policy later, and since the standard itself is highly dependent, it is close to a bad move. If someone suggests setting the price like this, it is worth suspecting if that person is a spy.

  • When to use: When alternatives are clear, and the category battle is a 'Similar Spec vs. Cheaper' structure.
  • Caution: If your Differentiation Point (Data Security, Localization, Support SLA, Ecosystem) is clear, consider Premium Positioning instead of following.

Value-Based Type

This is the standard. A method of anchoring to the economic/emotional value the customer gains. For example, if "10 hours saved per week × $30 per employee hour = $1,200 monthly value", you capture a part of that (e.g., 10-25%) as the price.

Most of the methods mentioned earlier are composed like this.

  • How to execute: Define Key Use Cases → Quantify Value Drivers (Saved Time, Conversion Rate Improvement, Error Reduction, Regulatory Risk Mitigation, etc.) → Survey Willingness-to-Pay (WTP) by Segment (Van Westendorp, Gabor-Granger, Conjoint, etc.) → Differentiate prices with Good-Better-Best packages.
  • Caution: "Value Delivery" storytelling and Evidence (Case Studies, ROI Calculators) must follow.

Bundle/Packaging Strategy

Packaging determines ARPU and NRR more than the price itself. You can think of it as the DLC concept in games, but to use this, several more strict conditions must be met. If you touch packaging or bundles without preparation, issues of fairness and value evaluation with other features arise.

  • Good-Better-Best (3 tiers): Gather the majority of users in Better (Behavioral Economics 'Center Stage' effect).
  • Feature-Based vs. Workflow-Based: Explain not as a list of buttons but as a bundle of results (e.g., "Content Team Automation Bundle").
  • Add-ons: Compliance, Security, Audit Logs, Dedicated VPC, SSO, Premium Support as high-margin add-ons.

Usage-Based / Metered Type

Charge proportionally to performance/consumption such as API calls, processed tokens, storage/transfer amount, number of tasks, etc. It has become especially common in the Generative AI era. Places that organized this amazingly well in the early days were Postman and AWS, and DataDog was a place that did it haphazardly in the beginning. Now, of course, it is tremendously sophisticated and well-established. Their method later inspired the pricing plans of Snowflake and Databricks.

  • Pros: Adoption threshold↓, Value and Cost align naturally.
  • Cons: Predictability↓, so Enterprise procurement may be reluctant.

Hybrid is safe. "Base Seat (Stable Revenue) + Excess Metering (Upside)" structure.

Per-Seat or Module/Workflow Based

Per-Seat is advantageous for expansion (land & expand), but automation/agent-type products may have a mismatch between the number of seats and value. In this case, you should consider switching to value proxy metrics such as Number of Domains/Projects/Agents, Number of Executed Jobs, Organization Unit, etc.

This is a method frequently seen these days in Make, Zapier, etc. However, this is also operable only if the scale supports it. Seeing a company called Unito or some others disappear after leveraging unlimited offers... Recently, places like Dify and n8n are also borrowing some of these.


Conditions Where Price Hikes 'Work' (Checklist)

  • Product-Market Fit (PMF) Clear: Key use case dominance, high replacement cost/switching risk.
  • NRR ≥ 100%: Expansion revenue can cover net retention/upsell.
  • Feature/Quality Superiority: Performance, Accuracy, Security, Manageability, Localization, etc. (The most basic)
  • Packaging Redesign Accompanied: Must be tied with a value story, not just a simple 'price hike'.
  • Communication Rules:
    • Prior notice to existing customers (Usually 30-60 days)
    • Grandfathering (Choose between Limited Time/Permanent) or Early Renewal Discount
    • Honestly explain the justification for the price increase (Performance, Reliability, Security, Enhanced Support, Infrastructure Cost Realization)
    • Annual Prepayment Conversion: Cash flow, Churn reduction.
    • Customer Success Team Prep: Churn/Downgrade prevention talk tracks, offering alternatives.

Practical Playbook

  1. Segment Separation: Individual/SMB/Mid-Market/Enterprise, Region (KRW/USD), Industry (Regulation), Partner/Education/Non-profit, etc.
  2. Value Driver Modeling: Hypotheses and ranges of 'Reduction/Increase' by feature (Conservative-Aggressive Scenarios).
  3. WTP Survey & Price Experiment: PSM (Range), Gabor-Granger (Point), Conjoint if necessary (Package).
  4. Packaging Refactoring: Feature → Result centered; Add-on definition; Upgrade path design.
  5. Price Book & Guardrails: Discount limits, Free extension rules, Quote approval lines.
  6. Rollout Plan:
    • New Customers: Apply immediately
    • Existing Customers: Notice + Options (Grandfathering/Early Extension/Upgrade Credit)
    • Metric Board: ARPU, NRR/GRR, Conversion Rate, Trial→Paid, Downgrade Rate, Tier Mix, Churn Reason by Cohort.
  7. 2-4 Week Unit Review: Monitor reaction → Fine-tune Packaging/Message.

While giving GTM advice or helping with interviews recently, I often feel regret that there are more people than expected who don't know at all how to unravel the story while keeping the value, even if there are gaps in the calculation formula. The failing patterns really frequently seen in the field are...

  • Copying Competitors: Announcing '$X → $X+5' without our own value description.
  • Absence of Message: Repeating only "Because prices rose." (It may be true, but it must be connected to customer value to be convincing).
  • Ignoring Local: Missing KRW price, VAT, Tax Invoice, Domestic Payment Methods → Perceived price increase is exaggerated.
  • Excessive Discounts: Raising the base price and offering a 30% coupon at all times. Customers learn the discounted price as the base price.
  • Absence of Measurement: No churn reason codes, downgrade reasons, or reaction by segment, so judgment of good or bad policy is impossible.
  • 'Seat=Value' Illusion: Insisting on Per-Seat for automation/agent products → Failure in monetization.

Price is not a number but a narrative method of delivered value. Even if customers may decrease in the short term, value-aligned prices increase the long-term survival rate. If you are going to raise the price, do you know what to do first? If what you are making is a one-time service, selling a fixed license, or simply providing specific features, it actually doesn't matter much what you attach. But attaching 'aaS' means you intend to continue the business as a service provider, right? Let's think from this part.

How SaaS Services Raise Prices | Oswarld Boutique Firm