An opinionated essay on the pricing mistakes solo founders make — cost-plus thinking, fear of charging, the round-number trap, and tier design that ignores value.
Methodology. This essay synthesizes pricing research from Patrick Campbell (ProfitWell), Madhavan Ramanujam, public SaaS founder reports, and the editorial team’s reading of pricing pages across 100+ SaaS companies.
Read enough indie SaaS pricing pages and a depressing pattern emerges. Three tiers, the middle one labeled “most popular,” round-number prices ($10, $20, $50, $99). Tier one has “up to 5 projects,” tier two has “up to 50,” tier three has “unlimited.” The bullets are features the founder is proud of, organized roughly by build order. The prices feel comfortable. They are also wrong — not because the numbers are too low (though they usually are), but because the founders did not derive those numbers from anything. They picked them by feel, by glancing at three competitors, and by the deeply human instinct that says “I would not pay $79 for that, so I will charge $19.” That instinct is the single most expensive mistake in indie SaaS, and it compounds for years before anyone catches it.
Almost every wrong price page has its roots in one or more of five errors. They are easy to name and hard to escape, because each of them feels reasonable in the moment.
The first and most pervasive mistake is founder fear. The founder asks themselves what they would pay for the product, decides that number, and ships it. This is wrong because the founder is the worst possible proxy for their customer. Founders are paying for tools out of personal cash flow on a side-project budget. Their customers are usually paying with company money, often expensing, and almost always have a higher willingness-to-pay than the founder’s gut suggests. The price that feels comfortable to the founder lands in the bargain-basement tier of the market they are actually selling into.
The fix is uncomfortable but simple: assume your gut is wrong by a factor of two. Whatever number feels right, double it. Pieter Levels has written about this rule for years; Patrick Campbell’s ProfitWell research has confirmed it across thousands of SaaS pricing studies. The median company that runs a willingness-to-pay analysis discovers customers would pay 25–40% more for the exact same product. That is a permanent revenue lift left on the table by a single bad assumption.
The second mistake is cost-plus thinking imported from physical products. The founder adds up server costs, AI tokens, support time, and a margin, and produces a price. This is fine for shipping widgets and ruinous for software. SaaS has near-zero marginal cost per customer. The interesting question is not “what does it cost me to deliver this” — the answer is “basically nothing.” The interesting question is “what is it worth to the customer who has the most painful version of this problem?”
Cost-plus pricing for software is like pricing a novel by the cost of paper and ink. The marginal cost is real but it is not the value. Customers do not buy your hosting bill; they buy a result. Price the result, not the hosting bill. Madhavan Ramanujam’s Monetizing Innovation spends most of a chapter on this exact mistake, naming it the “Hidden Gem” failure mode — companies that ship valuable products and price them as commodities because they did not understand what the customer was actually buying.
The third mistake is the race to the bottom. The founder googles three competitors, sees they charge $19, $29, and $39, and prices at $24 to be in the middle. This treats pricing as a market-clearing exercise instead of a positioning decision. Worse, it borrows the competitor’s implicit assumption about who the customer is and what value they get — assumptions which may be wildly different from the ones that fit your product and your customer.
Competitor prices are useful as a sanity check, not a starting point. If your competitors are all at $19 and you decide your value justifies $79, that is a positioning statement — you are claiming a different customer segment, a different value tier, a different category. That is fine and often correct. What is not fine is assuming the competitor’s number is the right number because it is the visible one. Visible does not mean optimal; it means published.
The fourth mistake produces the cluttered three-tier pricing page where each tier is a feature ladder organized by build order, not by customer. The result is that an actual customer cannot tell which tier they belong in — they have to read every bullet, decide which features they need, and self-select. Most do not bother. They pick the cheapest, or they leave.
The right way to design tiers is by user segment. Each tier should map to a different kind of customer with a different value model: starter for the early-stage user who needs the core cheap, growth for the median customer who values reliability and integrations, scale for the heavy user with budget who values support, audit, and limits. When tiers are segment-shaped, customers self-select correctly without having to read every line.
The fifth mistake is cosmetic but real. Founders default to round numbers — $10, $20, $50, $99 — because they look clean. Pricing tests across multiple SaaS companies have repeatedly shown that ugly numbers ($14, $29, $79) often outperform their round-number neighbors on conversion. The cognitive effect is that ugly numbers feel calculated, like the founder ran the numbers and arrived at exactly $29 because that is what the value supported. Round numbers feel like they were picked by gut — which they usually were. Even the difference between $99 and $100 is meaningful: $99 reads as “under a hundred,” while $100 reads as “triple digits.” This is one of the cheapest pricing fixes available, and most indie SaaS pricing pages skip it.
Strip away cost, competitors, and comfort, and one question remains: what would my best-fit customer pay if my product delivered the value I claim? That number is the only one that matters. The right way to find it is not to guess. It is to ask. Five fifteen-minute conversations with target customers will produce a willingness-to-pay distribution more useful than a hundred hours of competitor analysis.
The script is borrowed from Peter van Westendorp’s 1976 Price Sensitivity Meter and adapted for one-on-one conversations: ask at what price the product would feel expensive but considerable; at what price it would feel suspiciously cheap; at what price it would be flatly out of reach. The intersection of those three answers across a small sample reveals a defensible price band. Set your starting price at or slightly above the median of the “expensive but considering” answers — the slight upward adjustment corrects for the systematic underpricing bias documented across pricing research, where interview subjects anchor low because they imagine writing the check personally.
The single most useful pricing heuristic for solo founders is to price two to three times whatever number first felt right. This is not a joke; it is a corrective for the systematic underpricing bias above. The asymmetry of the bet is the whole argument. If you priced too high, a few prospects will tell you so on sales calls and you can adjust down — the data arrives quickly and unambiguously. If you priced too low, you will get plenty of customers, you will feel validated, and you will never know that you left half your revenue on the table. The signal of mispricing-low is silent; the signal of mispricing-high is loud. Optimize for the loud signal.
For any B2B SaaS aimed at customers who can expense the tool, a useful mental floor is $99 per month. Below that, the math stops working at solo-founder volumes. A $19 tool requires roughly five times the customers of a $99 tool to produce the same MRR — five times the support load, five times the churn replacement work, five times the marketing reach. Anything below $20 is hobbyist territory; the math of $19 SaaS is built for venture-backed companies with a thousand customers and a marketing team. The math of $99 SaaS is built for one-person operations with a hundred customers and a content strategy. The exception is a high-volume B2C product where reach is the moat — if you have a viral loop and the unit economics are real at small scale, $9 or $19 can work. Most solo founders are not building that.
The tier structure that consistently outperforms in SaaS is three tiers mapped to three customer segments. Free belongs in there only if your funnel is genuinely lead-gen and the free tier produces qualified prospects that convert — otherwise it is a charity. The three paid tiers: Starter for the early-stage user who needs the core cheap, Growth for the median customer who wants reliability, integrations, and team features, and Scale for the upmarket customer with budget who values priority support, audit logs, SSO, and higher limits. Each tier should be designed for a real customer archetype, not for a feature ladder.
The middle tier — Growth — is the one most customers should buy, and pricing pages should anchor there. Starter exists to make Growth look like the obvious next step; Scale exists to make Growth look like the reasonable middle. Three-tier structures consistently outperform two-tier or four-tier structures: two tiers force a binary, four tiers produce decision fatigue.
Packaging is which features are in which tier. Pricing is what each tier costs. Founders often conflate them and end up with a single mediocre answer to two questions. The useful insight: you can have one pricing tier and still vary your effective price by usage or by add-on. A single “Pro” tier at $49/month with metered usage above a threshold gives you simple pricing and variable revenue. Decoupling packaging from pricing opens design space that the “three tiers, three prices” default closes off.
Usage-based pricing — charge per API call, per gigabyte, per AI token — is the fastest-growing pricing model in SaaS for several years running, according to OpenView’s annual benchmarks. It works because it aligns price with value, and for infrastructure and AI products where COGS scales with usage anyway, it is often the natural fit. Where it kills you is in customer psychology. Usage-based pricing produces variable bills, which produce sticker shock, which produces churn. A customer who used the product heavily one month and got a $400 bill instead of the $99 they expected will cancel before the next cycle — even if the value they got was worth $400. The fix is transparent usage dashboards, proactive emails when a customer is about to cross a threshold, and configurable spending caps. If you are not willing to build that operational layer, do not ship usage-based pricing. For deeper tradeoffs on the AI side, where usage pricing dominates, our piece on the true cost of running an AI SaaS walks through the margin math when COGS scales with usage.
Eventually, every solo SaaS founder raises prices. The question is what to do with existing customers, and the answer that almost always wins is to grandfather them: existing customers keep their old price; new signups pay the new price. This builds enormous goodwill at low cost — the marginal revenue you would have gained by forcing the new price on existing users is usually outweighed by the churn it causes. The exception is when the gap becomes structural after two or three rounds of raises. At that point, a planned migration with thirty days of notice and an option to lock in a year at the old price is appropriate. The founders who get into trouble are the ones who raise prices on existing customers without warning, in a footer of a routine email. That move turns a small percentage of churn into a public incident every time.
Imagine a productivity tool aimed at solo founders — it takes a stack of customer interviews and produces a structured insight summary. The founder ships it at $19 per month because that feels comfortable. After three months they have 80 paying customers and $1,520 in MRR. Decent traction; not enough to live on.
Run the math at $79. Suppose conversion drops by half — a worst-case assumption from any pricing study. They now have 40 paying customers at $79, which is $3,160 in MRR — twice the revenue from half the customers, with half the support load and half the churn replacement work. Even if conversion drops by 60% (32 customers at $79), MRR is $2,528, still 66% higher than the $19 number. The hidden tax of the $19 price is not just lost revenue per customer; it is the operational tax of needing twice the customers to support the same MRR. For a one-person company, that is the difference between working full-time on the product and working full-time as the customer-service-and-marketing arm of the product. The SaaS pricing experiments playbook covers the cohort-substitution test design needed to actually validate these numbers.
The hardest part of pricing is psychological, not analytical. Founders know intellectually that they should price higher; they cannot bring themselves to do it. The fix is not more spreadsheets. The fix is to accept that the first price will be wrong, the second price will be wrong, and the third price will be roughly right — and to schedule those iterations rather than agonize over getting the first one perfect.
Set the first price at twice your gut. Run a cohort-substitution test in three months. Raise again in six months if the test passes. By the end of year one, you will have a price that fits the value, supported by data, with grandfathered early customers feeling lucky. Pricing is a recurring discipline, and getting it wrong twice is part of getting it right the third time. Do not obsess over the first number; obsess over the iteration cadence. The rest of the framework — positioning, models, mechanics, audits — lives in the complete guide to SaaS pricing; the broader playbook for solo founders is in the solo founder pricing playbook; and the SaaS pricing calculator can model the revenue math for your specific scenario. If you have not yet decided what your product is or who it is for, pricing is downstream of that decision — read what is SaaS positioning and the complete guide to SaaS positioning first.
The stack, prompts, pricing, and mistakes to avoid — for solo founders building with AI.