How to Price a SaaS Product for Your First Customers

To price a SaaS product for your first customers, start from what the problem costs them today, charge for the unit they value, put the paywall right after they see the result, and offer one plan at a price that feels slightly too high. Then ask real people to pay, by hand, and adjust from what they do. Your first price is an experiment. Its main job is to tell you whether you've built something worth paying for.

Here's how to set it, and how to use it to learn.

Your first price is a test, not a decision

Founders delay pricing because it feels permanent. It isn't. Steven Sinofsky's a16z essay on early SaaS pricing frames early prices as an iterative search, and expects them to change. With your first ten or twenty customers, you can change the price every week if you need to. What you can't do is learn anything about value while the product is free.

A price turns vague interest into a clear answer. "I'd use that" costs nothing to say. Entering a card number is a decision. That's why I'd rather see a founder ask for money too early than too late, and why pricing belongs in validating the idea as much as in launching it.

Step 1: Start from what the problem costs them

Don't start from your costs or from what feels fair. Start from what your customer loses without you.

Write down, for one specific customer:

1. What they do today instead. A spreadsheet, a freelancer, an agency, a competitor, or nothing.
2. What that costs them. Money spent, hours lost, revenue missed, mistakes made.
3. How often it happens. Daily problems justify subscriptions. Yearly ones usually don't.

If your product saves a small business several hours a week, the value is far above a few dollars a month, even if your hosting bill is tiny. Your price should sit well below the value and well above your costs. That gap is the space you're pricing in.

Then check comparables. Look at what your customers already pay for nearby tools. You don't have to match them, but being far below every alternative tells buyers your product is minor, and being far above needs a clear reason.

Step 2: Choose what you charge for

The unit you charge for matters more than the number. It should grow as the customer gets more value.

Charge per…	Fits when	Watch out for
Flat monthly fee	Value is roughly the same for everyone; you want the simplest offer	Heavy users get a bargain, light users feel overcharged
Seat (user)	Value grows with the number of people using it, like collaboration tools	Customers share logins to avoid paying
Usage (credits, messages, reports)	Value and your costs both grow with use, like AI features	Bills feel unpredictable; people ration use
Outcome (booking, lead, transaction)	You can track the result directly	Harder to build, harder to explain
One-time payment	The problem is solved once, or it's a tool people use occasionally	No recurring revenue; you need a steady flow of new buyers

If your product has AI features with real per-use costs, don't offer them unlimited on a flat plan. Usage limits or credits protect you from your most enthusiastic users.

Step 3: Put the paywall at the value moment

Where you ask for money matters as much as how much.

• Too early: before the user has seen anything work. They're being asked to pay for a promise.
• Too late: after they've got everything they needed for free. There's nothing left to pay for.
• Right: just after they've seen the result once and want it again, or want more of it.

So define the value moment first: the first report generated, the first booking made, the first match. Let users reach it. Then charge for repeating it, for more of it, or for the version that saves them the most time. I wrote more about charging for the wrong thing at the wrong moment, which is one of the most common reasons apps don't make money.

Step 4: Pick a number

For the first customers:

• Offer one plan. Tiers are for when you know which customers value which features. You don't yet.
• Avoid very low prices. A tiny price attracts customers who are shopping on price, makes every support request unprofitable, and loses a bigger share to payment fees. Stripe's standard US card rate is 2.9% + 30¢ per charge: on a $5 charge that's about 9% of the payment, on a $29 charge about 4%.
• Go slightly above comfortable. If every person you ask says yes immediately, you've learned your price is too low, not that you've found product-market fit.
• Use round, simple numbers. Clarity beats clever price points at this stage.

Step 5: Ask in person, and watch what people do

Your first customers shouldn't come through a pricing page. They should come from conversations. Show the product, let them reach the value, then say the price plainly:

"It's [price] a month. Do you want to start?"

Then stop talking. What happens next is your data:

• Yes, right away: good sign, and possibly too cheap. Raise it for the next person.
• Yes, after questions: the price is in range. Write down the questions; they're your pricing-page copy.
• "That's too expensive": ask "compared to what?" The answer tells you which alternative they're comparing you to.
• "Let me think about it": usually a no. Ask what would make it an easy yes.

If you're talking to several prospects, you can borrow the four questions from Van Westendorp's Price Sensitivity Meter, a pricing research method from 1976: at what price would this be so cheap you'd doubt the quality, a bargain, getting expensive, and too expensive to consider? The method is designed for surveys of real potential buyers, so with a handful of conversations treat the answers as a rough range, not a precise number.

Make a founding offer, with an end date

Early customers take a risk on an unfinished product. Reward them, but don't lock yourself into a price you'll regret.

• A founding price, clearly labeled as one, below the price you plan to charge later.
• A clear limit: the first 20 customers, or until a specific date.
• What they get in return for the risk: direct access to you, input on the roadmap, fast fixes.
• A decision you've made in advance about whether their price is locked forever or for a year.

A founding offer also gives prospects a real reason to decide now instead of "later".

When to raise the price

Raise it when you see any of these:

• Almost nobody pushes back on price.
• Customers say "this is cheap" or tell you what they'd have paid.
• Your best customers use the product heavily and would clearly pay more for more.
• You've added the thing that removes the most common objection.

Raise it for new customers first. Existing customers can keep their price for a while, or get notice and a clear reason. Raising prices once you've proven value is how you find the real price.

Pricing mistakes that cost early founders the most

• Staying free "until it's ready". It never feels ready. Ask for money once users reach value. More on what to do in the weeks after launch in the first 30 days after your MVP.
• Pricing from costs. Your costs set the floor, not the price.
• Copying the lowest competitor. You'll inherit their price-shopping customers.
• Launching with three tiers and an enterprise plan before you have ten customers.
• Unlimited AI usage on a flat fee. One heavy user can wipe out the margin of many light ones.

Common questions

Should I start with a free plan?

For most early products, no. A free plan means supporting users who may never pay, and it delays the signal you need most. A time-limited trial or a money-back guarantee lets people reach the value without turning free into the default.

Monthly or annual billing?

Start with monthly. It's easier to say yes to and gives you faster feedback. Add annual billing with a discount once customers stay long enough that committing for a year feels natural to them.

How do I know if my price is wrong?

Watch behavior, not opinions. If people reach the value moment and still don't pay, the problem is usually value, audience or timing, not the number. If they pay instantly and never question it, the price is probably too low.