Pricing a SaaS Product from First Principles

Value-based pricing means pricing on what the customer perceives the product to be worth, rather than on production cost plus a markup (Paddle). To use it, you need a number for that value.

Write a short value model for your best-fit customer:

Customer: a 5-person accounting firm (illustrative)
Problem: staff spend hours each month chasing client documents
Hours saved per month with the product: 20
Loaded cost of one staff hour: H
Monthly value from time saved: 20 x H
Other value: fewer missed deadlines, faster month-end close

Then ask: what share of that value can you capture while leaving the customer clearly better off? There is no fixed rule, but the customer must keep enough of the gain that the purchase is an easy decision. If the value is hard to quantify, that is a signal that you do not understand the customer's problem well enough yet, and more interviews are a better next step than a pricing spreadsheet.

Paddle suggests sorting features by how much customers value them and how much more they would pay for them. Features that customers see as essential but will not pay extra for belong in every plan. Features with high value and high willingness to pay are differentiators that can justify a higher tier (Paddle).

The value metric is the unit you charge for: seats, contacts, transactions, documents, rental units or usage. Chargebee describes it as the foundation of the pricing model and cites Patrick Campbell, founder of Price Intelligently, on what a good metric does: it aligns with customer needs, grows with them and is easy to understand (Chargebee).

Lincoln Murphy of Sixteen Ventures argues that the metric matters more than the price level. If you charge for something customers do not value, such as storage or seats that sit unused, you create what he calls the "shelfware perception", where customers pay for capacity they do not need (Sixteen Ventures).

The Pragmatic Institute gives two examples from its author's own companies. A property management product charged per rental unit managed, so customers paid more only as their own business grew. A roadmap product charged only for editors and gave free access to everyone else, because customers wanted to share roadmaps widely (Pragmatic Institute).

Score each candidate metric against four tests:

Predictability is often overlooked. Chargebee points out that customers dislike paying for a metric they cannot forecast, such as support tickets, and describes a help desk company that tried ticket-based bundles and moved back to per-user pricing (Chargebee).

Do not rely on one metric forever. Pricing consultant Robin Landy notes that pricing on a single metric decouples price from value when customers get very different value per unit, and gives enterprise buyers a single number to negotiate down. A second metric, add-ons or tiers give you more levers (Robin Landy). The opposite risk is too many dimensions. Chargebee cites investor David Skok's guidance to keep pricing axes to a small number so buyers are not paralyzed (Chargebee).

Once you know what you charge for, decide how you charge. The common models, as Stripe lists them, are flat subscriptions, tiered plans, per-user pricing, usage-based pricing and freemium (Stripe).

The pricing consultancy Willingness to Pay frames pricing as four parts that must reinforce each other: the metric (what customers pay for), the model (how it is charged), packaging (how plans and limits are organized) and segmentation (how pricing differs by customer type) (Willingness to Pay). Its list of common mistakes includes packaging features instead of value, which produces tiers with arbitrary limits and unclear upgrade paths.

Three tiers are a common default. The Pragmatic Institute notes that a low, middle and high package tends to anchor buyers toward the middle, and warns that an overly complicated scheme confuses customers and the finance team alike (Pragmatic Institute). A practical packaging approach:

1. Define each tier by the customer segment it serves, not by a feature list.
2. Put the features every customer needs in every tier.
3. Gate tiers on the value metric and on the few differentiators that a larger segment values.
4. Make the upgrade trigger obvious, so a customer knows when they have outgrown a plan.

You can learn a lot about price before anyone pays.

Pricing questions in interviews. In customer interviews, ask what the problem costs today, what they pay for current alternatives, and who would approve a purchase. Then show a price and watch the reaction. Hesitation, a question about discounts and immediate acceptance all tell you something.

The Van Westendorp price sensitivity meter. This survey method, introduced by Dutch economist Peter van Westendorp in 1976, asks four questions: at what price the product is so expensive you would not consider it, so cheap you would doubt its quality, starting to get expensive, and a bargain (Wikipedia). Plotting the cumulative answers gives an acceptable price range, bounded by where the curves cross. SurveyMonkey's guide walks through plotting the "too cheap" and "cheap" answers as inverted curves and the "expensive" and "too expensive" answers as standard curves, then reading off the intersections (SurveyMonkey).

Know the method's limits. The basic version tells you what range buyers find acceptable, not how many will buy at each price. Wikipedia notes that Van Westendorp did not address demand estimation and that later extensions add purchase-likelihood questions to estimate a revenue curve (Wikipedia).

Real offers. The strongest test is an actual price on an actual offer. Quote a price to design partners or early customers and see who signs. Adjust between conversations if you need to.

Keep a simple pricing log while you do this. For every conversation where a price came up, record the segment, the price you quoted, the reaction and whether the deal closed. After twenty or thirty entries, patterns appear: one segment accepts without hesitation, another always asks for a discount, a third never gets past the first call. If almost everyone accepts your first quote without any pushback, that is a sign the price is below what the market will bear, and the next quote should be higher.

Discounts are part of pricing, so decide them on purpose.

• Annual prepay. Many SaaS products offer a discount for paying a year upfront. The Pragmatic Institute advises using this with discretion because over the long term you can leave significant revenue on the table (Pragmatic Institute).
• Promotional discounts. Stripe warns that steep permanent price cuts train users to undervalue the product, and recommends tying any promotional discount to a longer commitment (Stripe).
• Enterprise. Large buyers expect to negotiate. Publish prices for self-serve plans and keep a "contact us" tier where a sales conversation is part of the process (Pragmatic Institute). Give yourself several levers to trade, such as term length, support level and add-ons, rather than only the headline price.

Also check that your price supports your sales model. The Pragmatic Institute notes that an expensive field sales team needs customer lifetime value high enough to pay for it. A low price that can only be sold through costly sales calls does not work.

Early pricing is almost always too simple and too low, and that is acceptable while you are learning. Willingness to Pay describes how the gap between the value a business creates and what its pricing captures grows over time, which consultant Ulrik Lehrskov-Schmidt calls commercial debt. Symptoms include heavy discounting, enterprise customers forced into small-business plans and expansion revenue that stays flat while customers grow (Willingness to Pay).

Watch for signals that the price is too low as well as too high. Too low looks like near-universal acceptance, customers who are surprised by how little they pay, and heavy users who would clearly pay more for the value they get. Too high looks like stalled trials, frequent discount requests and losses to cheaper alternatives that do the core job. Both are fixable, but only if someone is looking.

A simple review cadence:

• Every quarter: check win rates, discount rates and the share of customers at each tier limit.
• Every year: re-run value interviews with recent customers and check whether the value metric still tracks how they get value.
• When the product changes materially: revisit packaging before you add another tier.

When you change prices, the billing platform Maxio recommends communicating early and clearly: what is changing, why, and how it affects each customer (Maxio). Many companies keep existing customers on their old price for a period, which reduces churn risk and gives you time to show the added value.