My first paying user told me I was charging too little. It took me 4 months to listen.

this is one of the most expensive lessons in SaaS and youre learning it in public, good on you. the thing you half-say thats worth making explicit: price is a signal before its a number. at 12 pounds you werent just leaving money on the table, you were actively repelling the exact customers you wanted. a serious business owner whod happily pay 120 sees 12 and thinks "toy, side project, gone in six months, not safe to build my workflow on", so cheap doesnt merely undercharge the good buyers, it filters them OUT and filters bargain-hunters IN, and bargain hunters churn the hardest and complain the most. your "i need more users not more revenue per user" logic was a false tradeoff, because the low price was suppressing the good users, not attracting them. the tell was right there in her email: someone getting two hours a day back offered you 10x unprompted. thats not a compliment to file away, thats the market screaming your price is wrong. when you finally raised it, did serious-customer conversion actually go UP, the way it usually does, or did you hit the growth slowdown you were afraid of?

The filtering effect is the part I wish more people talked about. I am mid-launch on a paid tool right now and the strongest pull the whole way has been to lead with how cheap it is, as if a low number were a courtesy to the buyer. Your point that the price is a signal about your own belief in the product reframes that: the user who said she would pay ten times more was not being generous, she was reading the value more accurately than the person who set the price. Did raising it change who showed up in support conversations immediately, or did the old low-price cohort linger for a while?

Underpricing is one of those things that’s invisible from inside the business but glaring from outside. A buyer or new customer sees it instantly because they’re comparing against the market, not against what you charged last year. What made the 4 months so long? Was it more the fear of losing existing customers, or just not believing the feedback?

£12 and let the new price argue with new visitors — you keep the goodwill and get a clean read in two weeks instead of agonising for a quarter. The other thing her "I'd pay ten times that" told you was your value metric, not just your price: she measured you in hours saved per day, and a flat monthly figure has to carry the entire quality signal on its own when the plan could be anchored on the outcome instead. One caution from doing this on a self-serve product — the churn effect of a rise usually surfaces in month two or three rather than week one, so retention cohorts are worth watching before you call the new price settled. With 1,200+ clients now, do you see enough spread in how heavily people use Genie 007 to justify tiering, and if so what unit would you trust a buyer to actually predict about themselves before they've used the product?

The signal I trust more than a customer telling you to charge more is nobody pushing back on price at all. When I ran Henson Group, if we weren't losing roughly one in five deals on price, we were underpriced and quietly collecting the accounts that churn hardest. Did your win rate actually drop after the increase, or just your lead volume?

That low price point definitely sends an unintended signal—when something is priced like an experiment, people treat it like a temporary toy rather than mission-critical infrastructure.

I'm running the opposite experiment, so this is a useful mirror for me. I sell one-off B2C wallpapers for $1 — pet photo in, styled phone lock screen out. Your point about price-as-signal is dead right for a B2B tool that saves two hours a day, where a low monthly fee reads as "toy". But at the consumer end, deliberation is the enemy: someone has to want a wallpaper of their cat more than they want the dollar in their pocket, and every extra dollar of thinking time kills that impulse. So I moved the signal-carrying job from the price to a free watermarked preview — the preview is what says "this is what you get", and the $1 never has to argue about quality. Different market, different lesson, but your core point stands: whatever number you charge, it's communicating something whether you mean it to or not. (Disclosure: I make FurWall, the $1 wallpaper thing.)

Spot on

We had the same delay — and ours wasn't psychological, it was structural. We'd anchored the price in the discovery call before we understood what the client was actually replacing. Once we reframed it as "what does your current process cost you per month," the number stopped feeling uncomfortable. Pricing resistance is almost always a value articulation problem, not a number problem.

For usage-based AI products, I think a low entry price can be different from low perceived value. I'm launching a video background-removal tool where compute cost grows with video length, so I chose pay-as-you-go credits with no subscription. The tension is credits versus minutes: credits price the infrastructure, while minutes communicate value to the buyer. Your story makes me think the pricing page should lead with the outcome and keep credits as implementation detail. Did customers respond more strongly to the two-hours-saved framing than to any particular feature?

Spot on, Bill. This is such a common trap for early-stage builders. We often treat pricing as a math problem when it’s actually a positioning and psychology problem.

That low price point definitely sends an unintended signal—when something is priced like an experiment, people treat it like a temporary toy rather than mission-critical infrastructure.

Our biggest early blind spot was confusing 'accessibility' with 'value.' We assumed lowering the barrier to entry would speed up adoption, but it often just filtered for tire-kickers who demanded high-touch support while churning out the second things got tough. Raising prices shifts the entire dynamic because it filters for users who are actually invested in the outcome.

Out of curiosity, how did your conversion rates shift across the board after you finally ripped the band-aid off and raised it?

On the other side of this: I run FurWall, which turns pet photos into AI-styled phone wallpapers, and I deliberately priced it near the bottom (~$2 per wallpaper) because the purchase is pure impulse — someone sees their cat as a watercolor and wants it on their phone right now. What surprised me is that the real cost of pricing low isn't perceived quality, it's that $1-2 purchases don't get evaluated at all — so the free watermarked preview does all the trust-building before money changes hands. Curious if anyone else here deliberately sits at the bottom of the price curve, and how you make the product feel crafted rather than cheap.

One thing worth flagging: everything you've learned post-price-change comes from people who did convert at the new price — better conversations, lower churn, higher intent. That's real, but it's also a filtered sample. You have no visibility into whoever looked at the higher price, would've been a perfectly good long-term customer, and bounced silently because the number scared them off before they ever got far enough to discover the value.

At $12 you couldn't see this problem because almost nobody bounced on price. At the new price, some fraction of your "lost" traffic is probably people who were undersold by the pricing page rather than genuinely low-intent — and that's a much harder signal to catch, because it never becomes data. It just doesn't happen.

Worth testing for directly: has cart/signup abandonment specifically at the pricing page gone up since the change, and if so, do you have any way to talk to a few of those people? The customers who stayed are telling you the new price works. They can't tell you who it filtered out that shouldn't have been.

made the exact same mistake myself and now I am pivoting because the people that are showing up are taking their learning seriously and ADVANCING with all the right questions..

The four-months-to-listen part is the real lesson. A paying user telling you to charge more is the highest-signal feedback there is - they already voted with their wallet, so their price ceiling isn't hypothetical. The trap I fell into was anchoring on what the tool cost ME to build instead of what it saves the buyer. Did raising the price change WHO showed up, or just how much they paid? In my experience a higher price sometimes filters IN the more serious users rather than scaring people off.

This hits close to home as I get ready to lock in pricing for my app. think there's a real fear that raising prices will scare people off, when usually it just means you finally start attracting the right users. How did you arrive at the right increased pricing?

This is a great reminder right as I'm about to set my first price. I haven't launched paid yet, but I keep catching myself leaning toward "start cheap so it's an easy yes" — and now I'm wondering if that's exactly the trap you just described. My plan was tiered pricing based on the value delivered (in my case, recovered revenue for Shopify stores) rather than one flat low number, partly for the reason you mention: price as a signal of how seriously you take your own product. Curious — once you raised the price, did you lose any of your original low-price users, or did most just... stay?

Four months is a long time to sit on that sentence. I’d raise for new users first and grandfather the ones who already paid. Curious what you changed besides the number — copy, plan name, or just the price?

The part that stuck with me is that the price was filtering for a different kind of user, not just a different amount of money. £12 quietly told people this probably isn't serious, and the ones who needed it most read that signal and left.

Four months is also honest in a way most of these posts aren't. It's easy to say "just charge more" after the fact. Harder to admit you heard the answer and sat on it because raising the price felt like something you hadn't earned yet.

I'm pre-revenue and still figuring out what to charge, so the "I need more users, not more revenue per user" trap is one I can already feel myself walking into. Did the higher price actually slow signups at all, or did the volume stay roughly the same and just the quality shift?

This is really solid. How long did take you to get right?"

This may be a fake post.

Please check the website traffic on Semrush.

ZERO visitors.

The strongest signal isn't that one customer offered to pay ten times more. It's that she could put a number on the time saved: two hours a day. I build DictaFlow in a similar voice workflow space, and I've found that usage guides pricing better than compliments. I'd group customers by weekly actions or minutes saved, then compare retention and support needs before and after the price increase.

The signal worth watching after a price raise is not revenue per user, it is who shows up in your support queue. Every increase we ran at Henson Group changed the customer mix before it changed the P&L, and the cheapest tier was always where the highest-cost, lowest-patience accounts lived. Did your 12 pound cohort carry a different support load than the new one?

The price-as-quality-signal point is real, but there's a distribution layer underneath it that changed how I think about when pricing decisions actually matter.

With a small audience — say, 100 followers on X — a typical promotional post with a link gets about 500 impressions. After the algorithm's external-link penalty, call it 300 effective impressions. At ~1% CTR, that's 3 visitors to the product page. At a 0.5–1% cold conversion rate, the expected output is 0.015 to 0.03 sales per post.

At that level, you could double your price or cut it in half and the expected outcome barely moves. What you're measuring when you see zero sales isn't pricing signal — it's distribution signal.

Pricing as a quality filter — the mechanism you described where a higher price attracted better-fit users and lower churn — requires enough volume to observe the cohort difference. Your situation (1,200+ clients) is a scale where that lever is actually reachable. You raised the price and could see behavior change. That feedback loop needs a real sample size.

For founders earlier in the funnel: fix the denominator first. The pricing conversation is worth having at ~40–50 visitors/day. At 3 visitors/day, you're not running a pricing experiment — you're running on expected noise.

Had almost the same conversation with an early user. What finally convinced me was realizing the low price was quietly costing me trust: at a single-digit price people assumed the product was a toy and churned before setting it up properly. After raising prices the trial-to-paid rate barely moved, but retention got noticeably better, because the people who signed up actually intended to use the thing.

Four months is honestly not that slow. Most of us need to hear it from three or four users before we believe it.

We made the more extreme version of this call before we had a single user: no free tier at all, $9/mo from day one. It wasn't confidence, it was trying to skip the exact trap Eva_NomadOS describes above - free users generating support load with near-zero conversion. The harder part hasn't been picking the number, it's resisting the urge to bolt on a free tier the first time someone balks at paying anything, since 'nobody's signed up yet' and 'the price is wrong' feel identical from the inside, and only one of those is actually fixed by changing the price.

This hit home. I think a lot of first-time builders confuse “I need more users” with “I need to prove the value.” The pricing itself can actually be part of the positioning.
Curious — when you finally raised the price, did you change anything on the product/landing page, or did you literally just increase the price?

The price-as-filter part is the underrated one — when we raised ours, the questions got better and support got quieter. Same product, different people. The user who tells you to charge more is usually your best signal; it took me a few months to listen too.

This is interesting because founders often worry more about charging too much than too little. What convinced you that the customer was right: willingness to pay, the value delivered, or comparison with alternatives?

Same story here brother 😅
Took me months to realize people pay for value, not for hours.
Raising price actually got me better clients.

Great story. The pricing lesson hits home, it's so easy to undercharge when you're early.

My biggest mistake was keeping a super generous free plan for way too long because I was terrified of seeing zero signups. I ended up spending half my day answering support tickets for free users who complained about everything, while the few paying users never asked for anything. The second I killed the free plan and put up a paid trial, all the tire-kickers vanished and actual buyers showed up. Took me almost six months to realize free users almost never convert anyway.

The pricing reflects what kind of users you are targeting. It's easy to think that lower prices = more users willing to pay. But that's not always the case.

Lower prices tend to attract hobbyists or individuals who are simply trying out something new. These users might be easy to get, but they are also very easy to lose. Because most of them might have just started using the product on a whim.

Higher prices on the other hand, attract serious customers that actually NEED the product. These customers are a bit more difficult to land but when they buy, they buy because they plan on using the product for a long time -- their business depends on it.

If you're targeting businesses (fully or partially), I would experiment with increasing the prices.. or maybe introduce an enterprise plan. Serious businesses won't purchase something that looks "cheap". They need to know that your product is serious and you won't go out of business out of the blue.

If your product is strictly for hobbyists or for regular consumption, you can still raise prices, but the ceiling is probably not as high.

Gently against shenshenhq above: "they'd pay ten times that" is the most encouraging sentence you'll ever hear and close to the least reliable pricing evidence you'll ever get.

She was valuing something she already had and had already watched work for her. The person you need to price for is someone who hasn't used it, can't verify the two-hours-a-day claim, and is deciding from a landing page. Those are different acts. A delighted incumbent's number proves the ceiling exists; it says nothing about where the curve bends for people buying blind - which is why "I'd pay 10x" so seldom means 10x in practice. Your instinct not to just multiply by ten wasn't the mistake. Sitting at 12 for four months was.

Building on evanharland's point about who self-selects in, because I think there's a further consequence that takes about a year to show up. A low price doesn't only attract cheaper customers, it attracts customers whose underlying problem is small. Small-problem users generate more support per pound, churn sooner because the pain that would have held them was never that bad, and ask for features that pull the roadmap towards a product nobody will pay for. So the real cost of underpricing isn't the margin you left on each seat - it's that your roadmap quietly gets written by the people who needed you least.

This resonates — pricing confidence is the hard part, not the product. The "they'd pay ten times that" moment is worth more than any analytics. Curious how you finally picked the new number.

The useful signal here is not simply “charge more”; it is that price changed who self-selected into the product. I’d make the next test cohort-aware: keep existing users on their current plan, show new visitors a higher anchor, and compare activation, week-four retention, support load, and revenue per signup—not conversion alone. If the higher-priced cohort asks better questions and reaches value faster, that is positioning evidence; if they only convert less, the offer or proof may still be unclear. Did the price change alter the mix of use cases, or mainly the same users’ willingness to pay?

Yes man!! same problem with me can't decide a perfect price for my product. If I set it too low then client or customers would think that he/she is underestimating his/her own product and setting it too high could make it expensive

It’s funny how founders are often more scared of raising prices than customers are. Sometimes the low price can even make people question the quality. I’d be curious what happened to your sales after you finally changed the pricing

It’s funny how founders are often more scared of raising prices than customers are. Sometimes the low price can even make people question the quality. I’d be curious what happened to your sales after you finally changed the pricing

It’s funny how founders are often more scared of raising prices than customers are. Sometimes the low price can even make people question the quality. I’d be curious what happened to your sales after you finally changed the pricing

Funny how we trust users when they complain about pricing, but sometimes ignore them when they tell us the product is worth more 😅. I think a lot of founders underprice early because getting the first few customers feels more important than protecting margins. Did raising the price affect conversion much?

Funny how we trust users when they complain about pricing, but sometimes ignore them when they tell us the product is worth more 😅. I think a lot of founders underprice early because getting the first few customers feels more important than protecting margins. Did raising the price affect conversion much?

Really interesting point about pricing being a signal, not just a number. I think a lot of new founders (including me) focus too much on getting more users and forget that the price can affect how people perceive the value of the product.
Curious — when you raised the price, did you increase it gradually or did you make a big jump at once?

The pricing signal point is real, but I’d split it from funnel friction. If a user already says you save them two hours a day, raise price fast. If cold visitors are bouncing before activation, higher price will not fix an unclear path.

I like using a 7-day rule: one strong unsolicited value signal gets a pricing test within a week, one unexplained funnel drop gets a path audit before more acquisition.

@AmandaBrown I can turn those 20 customer interviews into a pricing segmentation matrix, define the two or three value ceilings, and give you the activation, retention, and willingness-to-pay thresholds for the next pricing test. The focused 20-minute session is $75. Book the Startup Advisory call here: https://calendly.com/dontae-threeum-nsuo/startup-advisory-call-20-min

The four month lag is the part that stuck with me. When a paying user says you are underpriced after getting real hours back, that is a cleaner willingness to pay signal than most early pricing experiments will ever give you. Waiting feels polite, but the market anchors hard on the first number you charge, and raising later costs more social friction than starting higher and offering a discount. Next time someone says that, I would treat it as data to act on within a week, not a compliment to file away.

The part that hit me hardest: "the price filters for a different kind of user." We kept our CSR generator API free at launch (no signup, no API key) — partly an SEO play, partly "more users first." But watching the user mix, the free tier attracts a lot of people who will never convert, while the serious users are the ones who'd happily pay for rate-limit headroom or an SLA. Same insight as yours, just from the free end of the spectrum: the pricing/access tier you pick doesn't just set revenue, it sets the quality of every interaction that follows.

Also — you're right that the unsolicited email was the strongest signal. Our rule of thumb now: unprompted feedback counts 10x, prompt-based feedback counts 1x. Took us a while to learn that too.

What stands out to me isn't that she was right, it's that the signal was unsolicited. She didn't respond to a survey or a churn-prevention email asking "would you pay more" — she emailed you unprompted, which is a much rarer and stronger signal than anything you'd get from asking directly (people routinely undersell their own willingness to pay when asked in the abstract).

Which makes the 4-month delay less about missing the signal and more about not having a system to act on signals that arrive outside your existing feedback loop. You had a "how do we get more users" process running, but no equivalent process for "someone just told me something structurally important, unprompted, at random."

Now that you've seen this once, do you have any way of surfacing that kind of signal faster next time — flagging unprompted, unusually specific feedback for a look within days rather than mentally filing it away? Or is the lesson more "you can't systematize this, you just have to be less attached to your existing plan when it shows up"?

She handed you the value in her own units, two hours a day, and you priced off how you felt instead. We did the same thing at Henson Group for years, pricing services off our cost rather than the client's saved headcount, and margin never showed up until we flipped it. The fix is not picking a bigger number, it is asking your next ten signups what the alternative costs them and pricing against that answer.

Mine was metering the wrong thing rather than pricing it wrong.

I run Lisar, a VPN service, and the plan ladder started out led by bandwidth: so many Mbps, so many GB a month, so many concurrent sessions. It felt rigorous. It is the structure that makes sense when you are the one paying for transit.

The problem is that it asks the buyer a question they cannot answer. Almost nobody knows how many gigabytes a month they use. So someone landing on the page has to guess, guessing feels like risk, and the safe response to risk is to take the cheapest tier or close the tab. The tiers were perfectly legible to me and illegible to them.

What people could actually self-select on was capability. Do I need a specific exit location. Do I want filtering at the DNS layer. Do I need routing a small team can share. Those map onto how someone describes their own situation in one sentence, with no arithmetic. As the capability differences started carrying more of the weight in how plans were presented, the conversation shifted from which number is enough to this is the one that does the thing I came for.

So the parallel to your story: your twelve pounds was sending a signal about quality. Mine was sending a signal about complexity. Both are the price communicating something you did not intend to say.

The part I would add to your framing is that raising a price is the easy version of this fix, because you can do it in an afternoon. Changing what you charge for means re-deciding what the product is segmented on, and that is slow and a bit humiliating, which is probably why it took me considerably longer than four months to admit.

Still not fully fixed either. The metered limits are still sitting right there on the page, because the underlying cost genuinely is bandwidth. The honest version is that your cost structure and your buyer's mental model do not have to agree, and when they disagree, the buyer's model should probably win the pricing page and yours should stay in the spreadsheet.

The AI-tool angle on this: when you built it in a weekend, your anchor for "what is this worth" gets even more distorted. The cost-to-produce feels like nearly zero, so you anchor the price to that instead of to the hours it saves the customer. Her "I'd pay ten times that" was the first clean signal that the product solved a £120 problem, not a £12 one — and those signals are rare, so the four months of ignoring it cost more than the missed revenue.

Also agree on price filtering who shows up: at the cheap tier you attract people testing whether AI can do the job. At the honest tier you attract people with the job that needs doing. Those are effectively two different products.

The four-month gap is the interesting bit to me, because the signal came from someone already paying, which is the least noisy source you had. One thing I'd want to know: when you raised the price, did you grandfather the existing £12 users or move everyone? I've seen the grandfathering choice quietly cap MRR for a year because the loudest advocates stay on the old plan and never test whether they'd have paid more. Also curious whether support volume per user went up or down after the change — my guess is down, since higher-intent buyers tend to read the docs first.

This matches a pattern I keep seeing when looking at pricing/churn data: underpricing doesn't just leave revenue on the table, it quietly pollutes every metric downstream. A £12/mo signup and a £120/mo signup filter for completely different intent before they even open the product once — so your activation rate, engagement, and churn numbers end up averaging two totally different populations together. The real cost isn't the missed revenue on existing customers, it's the noise it adds to your data for months while you're trying to figure out what's actually working.

The price increase changed both economics and customer selection. I would now segment by the job being done and track activation, support load, retention, and expansion at each price, rather than judging the move on signup conversion alone. The best price is not the one that produces the most accounts; it is the one that attracts customers with enough urgency to realize the product's value.

Getting told you’re too cheap by the person actually paying you is probably the strongest pricing signal possible 😅 What happened to conversions after you finally raised the price?

The sneaky thing about pricing is that it doesn’t just affect conversion. It selects the customer.

Cheap can attract curiosity. Higher pricing can attract urgency. Sometimes “raise the price” is really a segmentation strategy wearing a dollar sign.

Four months is the part I recognise, though mine was ad spend rather than price. Campaigns that clearly weren't working stayed live for months because I hadn't admitted they weren't working, and the delay cost me more than the money did.

What broke the habit was writing the number down before anything went live - if cost per lead is over X by day four, it dies. Deciding while I'm still neutral is the only bit that works, and a price change takes the same shape: pick what you'd need to see, and by when, before you touch it.

One user offering ten times the price is one data point. She happened to be right.

The insight about price filtering customer type is the real story here. You weren't just leaving money on the table - you were running the wrong measurement system. Your £12 price was sending a signal you couldn't hear because it was too weak to override your conviction about needing volume. Her signal was stronger because it came from actual value experienced. Same product, two measurement systems running in parallel. When you switched to hers, it didn't just change revenue - it changed what customers asked for, used, and valued. That's the hard part: founders who can see the signal first often understand their product better than the founder does.

This is a really interesting perspective. I think one of the hardest things for early-stage founders is separating “we need more users” from “we need the right users.”

We're currently facing a similar question while building our SaaS product. We're still very early, so we're experimenting with how pricing, free trials, and perceived value affect activation and conversion.

One thing I'm learning is that pricing isn't just about revenue—it also communicates positioning and the value you believe you're creating.

Curious: when you eventually raised the price, did you test different price points, or did you make the jump based primarily on customer feedback?

Answering your closing question: the pricing mistake that takes longest to fix is the model, not the number. The number is an afternoon's change, which your £12 story proves. The model is architecture: it wires into billing code, decides who self-selects in, shapes your own incentives, and most founders inherit it from whatever the market leader does rather than choosing it.

We sell in a market where the default model is a percentage of the revenue the software touches, and the switching complaints in competitor reviews are all about the model, never about the number. You made the cheap kind of pricing mistake. The expensive kind is the one nobody emails you about.

The idea that price filters for a different type of customer is important.

It is easy to view pricing only as a revenue decision, but it also changes who tries the product, how seriously they use it, and the quality of feedback they provide.

How did you decide on the new price point after finally accepting that £12 was too low?

The part about the £12 price changing who took the product seriously is really interesting.

It’s easy to think of pricing as just a revenue lever, but it can also affect the type of customer you attract and the conversations you have with them.

What surprised you most after making the change?

This hits harder than it looks. Your customer's signal (£12 feels underpriced) wasn't just about revenue - it was about your measurement system.

At £12, you were measuring: user count, growth rate, market reach. The metrics that seemed to matter.

At the higher price, you started measuring: conversation quality, feature adoption, churn rate. The metrics that actually reveal whether someone found real value.

That 4-month gap between hearing the signal and acting on it is the hard part. Your brain had optimized around "we need volume," so even a clear signal about value got filed away as secondary data.

The person who paid more wasn't a different customer. She was the same person - but now your price was aligned with the actual value she was getting. Which made both of you measure and communicate about the same thing.

Price is a measurement system. It determines what signals you can see, and which ones you'll trust.