Pricing is the decision founders put off the longest. It feels permanent and personal. It is neither. Your first price is a starting point you will change, probably more than once.
What I want you to avoid is the most common mistake I see: picking a low number because it feels safe, then never touching it again. A low price does not just reduce revenue. It attracts customers who are less committed, it makes your product look less serious and it leaves you with no room to grow.
Start with the customer's problem, not your costs
Your hosting bill, your AI credits and your hours are real, and you need to cover them. But customers do not care what it costs you to run the app. They care what the problem costs them.
Ask a few people who have the problem:
- How do you handle this today?
- How much time does it take each week?
- What are you paying for the tools or people involved now?
- What happens when it goes wrong?
If your product saves a small business owner several hours a month, or replaces a spreadsheet that keeps causing billing mistakes, that is your pricing frame. You are not selling software. You are selling that outcome.
Know your floor
Still, do the math so you know your minimum. List your monthly costs per customer:
- Database and hosting usage that scales with users.
- AI or API calls per user, if your app uses them.
- Payment processing fees. Stripe and others take a percentage plus a fixed fee per transaction. Check current pricing.
- Email, SMS or other per-message services.
- Your own support time.
If an AI-heavy feature can cost you real money per use, you need usage limits or a price that covers the heaviest users. I have seen founders lose money on their best customers because they did not check this.
Pick a value metric
A value metric is the thing you charge for that grows as the customer gets more value. Common ones:
- Per user or seat: good for team tools where more people means more value.
- Per usage unit: projects, contacts, bookings, documents, AI generations.
- Per location or account: useful for multi-location businesses.
- Flat rate: one price for everything. Simple to sell, but leaves money behind as customers grow.
The best value metric is easy to understand, easy to predict and matches how the customer thinks about their own business. A campground owner understands "per site." A consultant understands "per client." Pick the unit your buyer already counts.
Use tiers, but not too many
For a first product, two or three tiers is plenty. A pattern that works for many small SaaS products:
- Starter: for individuals or very small users. Core features, lower limits.
- Pro: the plan you expect most people to choose. More limits, the features serious users need.
- Business or Team: higher limits, more users, priority support.
Separate tiers by limits and by features that matter to bigger customers (team access, integrations, reporting). Do not hold back basic functionality so the cheap plan feels broken. That just creates frustrated customers.
Free trial, free plan or neither
This is one of the bigger decisions, and it depends on your product.
- Free trial (time-limited, full access): works well when the product shows its value quickly. It creates a natural decision point.
- Free plan (permanent, limited): works when free users spread the product, like when they invite others or share outputs. It also creates support load and infrastructure cost for people who may never pay.
- Paid from day one, maybe with a money-back guarantee: works for niche B2B tools where buyers are serious and volume is low.
For most first-time founders building a niche tool, I lean toward a trial or paid with a guarantee. A free plan is a distribution strategy, and it only pays off if you have one.
Monthly and annual
Offer monthly so people can start without much risk. Offer annual with a discount so committed customers can save and you get cash up front. Many SaaS products discount annual by roughly the cost of a couple of months. Pick something simple that is easy to explain.
Set the actual number
Here is the process I use when I have no other data:
- Look at what alternatives cost. Not just direct competitors, but spreadsheets plus a part-time assistant, or the agency they hire now. Your price should make sense next to those.
- Estimate the value. If the product saves a few hours a month, what is that time worth to this buyer?
- Pick a price that is a clear deal against the value but not a rounding error. Many founders land well below a fair share of the value they provide.
- Check it against your floor so you are not losing money on heavy users.
- Say the price out loud to a potential customer and watch the reaction. If nobody hesitates, it may be too low.
A little hesitation is normal. Instant agreement from everyone is a signal to go higher next time.
Grandfather early customers
Your first customers take a risk on you. I like to reward that by locking their price for a set period or for as long as they stay subscribed. It makes the first sales easier and it means you can raise prices for new customers later without upsetting the people who got you started.
Be clear in writing about what is locked and for how long.
Test and adjust
Pricing is something you revisit, not a one-time decision. Some ways to learn:
- Change prices for new signups and compare conversion over a few weeks.
- Ask every customer who cancels why. Price is often not the real reason.
- Ask every customer who upgrades what triggered it. That tells you which limits and features matter.
- Look at which plan people pick. If nearly everyone picks the cheapest, the tiers may need to be redrawn.
Set up your products and prices in Stripe so changing them is easy. Keep the plan names and limits in your database, not hard-coded in a dozen places in your app. That makes adjusting later a configuration change instead of a rebuild.
Pricing page basics
- Show the price. Hiding it behind "contact us" for a small product loses buyers.
- Highlight the plan you recommend.
- List limits clearly so nobody is surprised.
- Answer the common questions right on the page: can I cancel, what happens to my data, do you offer refunds.
Pricing is one of the sessions in our live cohorts, where we work through real numbers for each builder's product.
What would have made it better?