There is no free lunch, especially in app publishing

2009, mobile networks class. We were studying how TCP handles packets lost over 3G: the network detects the loss, replays the packet, and the transfer survives a radio link that drops things all the time. Someone asked the obvious question: why not just make the radio link reliable in the first place? The professor's answer became the soundtrack of that semester: there is no free lunch. Reliability over 3G is not free, it is paid for in retransmissions, latency and battery. The cost never disappears. It just moves to wherever you are not looking. I now run engineering at a no-code app builder. Every pricing decision I have seen from the inside since, his sentence explains better than any business book. Here is the version of it nobody puts on a landing page: you can build an app for free anywhere, and you can ship one for free nowhere. The asterisk every vendor knows by heart Search "free app builder". Every result promises an app for free. Every vendor behind those pages knows exactly where their free tier ends: right before the App Store and Google Play. Build for free, preview for free, then the store gates close and the pricing table appears. I will not do the pricing comparison here. We published a dated, sourced table of what each major free tier actually includes on the company blog: Can you really create an app for free?. Spoiler: the pattern is identical across the market, and one competitor even admits in its own comparison that store publishing always requires a paid plan. This post is about the side of the counter you cannot see from a landing page: what the word "published" costs the vendor. Because that is where the free lunch dies. What "published" costs us, forever A built app is a file. A published app is a commitment. From our machine room, that commitment looks like this. The store treadmill never stops. Google Play raises its target API requirement every year. Apple killed UIWebView in 2020. The legacy push protocol died in March 2021. The Play Data safety form arrived in July 2022, Apple privacy manifests in May 2024. Each event means rebuilding and resubmitting fleets of apps through two review pipelines. Skip one deadline and apps disappear from stores. This work is invisible, recurring, and nobody can bill it to "free". Runtime is real hardware. Our customers' apps call a backend every day. We run our own machines in racks in European datacenters. The CDN moves terabytes per day. Push delivery counts in millions of notifications per week. An engineer is on call all year; the status page is public (goodbarberstatus.com). None of this pauses when a customer pays zero. A free plan that included native publishing would mean carrying all of the above, indefinitely, for free. Nobody in this market does it. That is not stinginess. That is arithmetic. So who pays for your free plan? The lunch is never free; the bill just moves. On a typical free app builder tier, you pay with: - Branding. Your app advertises its builder. Removing the badge is a paid feature, everywhere. - The wall. The upgrade prompt lands at maximum investment, minimum leverage. That timing is not an accident, it is the business model. - A link instead of a listing. Free "apps" are often web links. The store fees never trigger, and neither does the store presence you came for. - Your time. The expensive one. Rebuilding elsewhere because platform one could not ship is the worst deal in no-code. And two line items belong to nobody's free plan because they are not the builder's to give: Apple charges 99 $ a year for a developer account, Google Play 25 $ once. A page that promises store publishing "for free" without those numbers is describing a demo. In defense of free tiers Said plainly, because my employer's competitors deserve fairness: free tiers are a great deal when your goal matches their content. Testing an idea on a real phone screen. Learning how a no-code editor thinks. Shipping a link to five colleagues. For all of that, use a free plan and pay nobody, including us. The problem is never the free tier. It is the landing page that lets you believe the lunch stays free all the way to the App Store. The menu with prices on it GoodBarber has no free plan, and now you know the reason is structural, not commercial. Our trade is different: the full platform, free for 30 days, no credit card, including a PWA you can actually publish during the trial (what free really includes). When it ends, nothing is charged, because we never took a card. You have seen the real product and the real prices, and you decide. No free lunch. But a menu with prices on it, and a table where you eat for 30 days before ordering. Same lesson as those 3G retransmissions: the cost never disappears, so honest design shows you where it sits. You know what to expect, and the price is not hidden. In 2009 we would have called that a well-designed protocol. Top comments (0)