Describe what you want in plain English. We’ll ask two or three questions that actually change the price, then show you a cost range, a timeline, and where every hour goes.
No paywall, no partial result, no consultation to unlock the number. Copy it, print it, or send it to whoever signs off on the budget.
Estimating software to the dollar means guessing. You get a range, and a list of what would push it toward either end.
When two native apps are overkill, or a web app would do the same job, it says so. We would rather be useful than expensive.
Screen count is a poor proxy for effort. Six things set the price, and they are why two quotes for the same idea can differ by 3x.
Every figure on the right comes out of the same pricing config the estimator uses.
"Add chat" is one line on a whiteboard and six weeks of engineering. Read receipts, attachments, delivery, notifications, moderation, and the offline case are all part of chat once real people use it, and none of them are optional. The biggest lever on a budget is how many things get built before launch rather than after it.
$6K – $11Kwhat "add chat" actually costs
An app with one user type is one app. An app with a customer, a technician, and an administrator is closer to three, because each role needs its own screens, its own permission rules, and its own test pass. Roles multiply design and QA faster than they multiply engineering.
$3K – $6Kper additional user type
Integration cost has almost nothing to do with your app and everything to do with the system on the other side. A modern documented API is a week. A twenty-year-old ERP with no test environment and one person left who understands it is a month, and the risk lands on the schedule rather than the scope.
1 week or 1 monthsame feature, different system
Offline is the most underestimated line item in mobile software. Storing data locally is easy. Deciding what happens when two people edit the same record on different devices, and one of them was in a basement for six hours, is a genuine engineering problem with a business decision inside it.
$10K – $19Koffline plus the sync it implies
Bluetooth devices, printers, scanners, and sensors add connection lifecycle management, pairing and permission flows, protocol handling, and failure recovery. The cost is set by the device rather than the app: a documented protocol is quick, and undocumented firmware is a research project with a deadline attached.
$5K – $9KBluetooth, before the surprises
HIPAA, SOC 2, and formal accessibility conformance change how software is built, not just what gets documented at the end. Encryption, audit trails, access review, and accessible components are far cheaper designed in than retrofitted, and the review cycle itself consumes calendar time nobody on the project controls.
+5% – 35%across the whole project
A language model reads your description. Ordinary code does the pricing. We keep those jobs separate because models are good at working out what you meant and bad at arithmetic they were asked to make up.
Write a paragraph about what the app does, who uses it, and what it has to connect to. A language model reads that and turns it into a structured spec.
We price your project under every possible answer, then ask only about the ones that move the total. Usually three to five. You can skip them.
A cost range, a timeline, hours broken down by workstream, what is driving the price, what nobody can know yet, and a smaller version if one makes sense.
The model never sees a rate, an hour figure, or a multiplier, so it cannot produce a price even if you ask it to. How the estimate is actually calculated.
Any calculator can show you a number. The useful part is seeing which assumptions it rests on, and which of those are still guesses.
Every estimate names what nobody can know yet and says which way each one swings. A vague description gets a wide range, because that is the honest answer.
Multipliers hit the line items, never the total afterward, so the rows always sum to the headline figure. You can check our math.
Every hour range lives in one config file with the reasoning written beside it. Nothing is buried in application code where it can quietly drift.
Separate native apps, a custom auth system, Kubernetes, sometimes a mobile app at all. When your requirements do not justify it, the result says so.
Change the platforms, the backend, the number of user types, any feature. The estimate updates as you go, because the pricing engine runs in your browser.
Type a description, read the result, close the tab, and no record of it exists on our side. Sharing is the only thing that saves anything, and it expires.
Most published app costs imply every project is six figures. Most are not. Almost everything we see falls into one of these three shapes.
$15,000 – $30,000
A focused first release
One workflow, done properly, for one kind of user. The right shape when you need to prove a market, replace a paper process, or put something real in front of users before committing further.
$30,000 – $60,000
A production application
Where most business applications land. Several connected workflows, more than one kind of user, an administrative side, and integration with something you already run.
$60,000 – $120,000
A complex build
Hardware integration, genuine offline operation, compliance obligations, or a heavy integration surface. Often better delivered in phases than as one release.
These assume a US-based team building software a business will depend on. They exclude the cost of keeping it running after launch, which is usually 15 to 20 percent of the build per year.
What people want to know before they trust a number on a website.
Still not sure?
Run your own project through it. Two minutes, and you can change any assumption we make and watch the number move.
Estimate my appThe plain answer, the real numbers, and the reasoning behind them.
What actually drives the price of a mobile application, line by line.
Which line items cross-platform shrinks, and which it does not touch.
App Store review, Apple’s release cadence, and when iOS-only is right.
Device fragmentation as a real QA cost, and why field work starts here.
How browser applications differ, and when you need neither an app nor one.
About two minutes, and the whole estimate is free.