How much does custom software cost?

When an owner asks "how much does custom software cost," the honest first answer is a question back: what does it have to do, for whom, and what does it connect to? A booking form that emails a confirmation and a two-sided marketplace with payments, dashboards, and compliance are both "custom software," and they're an order of magnitude apart on price. The number tracks scope, not the fact that it's custom.

Six factors drive most of the variance between one quote and another:

1. Integrations

   Every third-party system the software talks to (a CRM, a payment processor, a calendar, a legacy database) adds build and maintenance cost. The initial build is only 30 to 40 percent of an integration's two-year total cost. Fewer, cleaner integrations mean a cheaper, more reliable build.

2. User roles and permissions

   One kind of user is cheap. Admins, staff, customers, and partners, each seeing different data with different permissions, multiply the screens, the logic, and the testing. Roles are one of the quietest but largest cost drivers.

3. Real-time and scale requirements

   Live updates, instant matching, notifications, and high-concurrency traffic all raise the engineering bar. If the software can process things in batches or near-real-time, it costs less than if everything must happen the instant it's requested.

4. Compliance and regulated data

   Health data (HIPAA), payment data, and other regulated categories add real cost and time (HIPAA can add 20 to 40 percent), and retrofitting compliance later costs several times more than building it in. If you're regulated, scope it in from day one.

5. Custom design

   A polished, bespoke interface typically runs 15 to 25 percent of the project. It's often worth it for customer-facing products and skippable for internal tools, where a clean, standard UI does the job for far less.

6. Who builds it

   Region and seniority swing the rate several times over, and specialist AI or security talent commands a premium. But rate is a weak proxy for cost: a fast senior on a fixed quote often beats a cheap team that takes three times as long or ships something you have to redo.

Here's the vocabulary you'll see throughout this guide, in plain English:

Fixed-price quote

A single agreed price for a defined scope, set before work starts. The risk of misjudging the effort sits with the builder, not you. Best when the scope can be written down clearly, which is most small business builds. It's the model behind milestone billing.

Time and materials

Billing by the hour or day for however long the work takes. Flexible for genuinely open-ended research, but the incentive runs backwards: slower work earns more. It's the model behind most agency budget overruns, because the meter keeps running whether or not anything ships.

Milestone billing

Splitting a fixed quote into working chunks, where you pay for each milestone when it ships and runs, and the final payment waits for your sign-off. It keeps a fixed-price build honest: you only release money against software you can actually use.

MVP

Minimum viable product: the smallest version of your software that delivers real value and lets you learn from real users. Scoping to an MVP first is the single biggest lever on cost, because it strips the build down to what actually has to exist on day one.

Total cost of ownership

The all-in cost over the life of the software, not just the build: hosting, maintenance, support, integrations, and the subscriptions it replaces or adds. Off-the-shelf tools look cheaper up front and often cost more over three years once per-seat fees and price hikes compound.

Cross-platform

Building one codebase (with React Native or Flutter) that ships to both the iOS and Android app stores, instead of two separate native apps. It typically cuts mobile cost 30 to 45 percent, which is why most small business apps should start here.

Published 2026 cost breakdowns from development firms converge on a similar shape, even though the exact edges differ vendor to vendor⁵⁶. Here are the three bands, framed for a small business buyer:

1. Simple build ($30k and under to ~$50k)

   A single-purpose tool: a booking or quoting form, an internal dashboard, a light customer portal, one or two integrations. Timelines run a few weeks. This is where AI-accelerated development bites hardest, and where a focused build can land in the low five figures or less. Most small businesses start here and are right to.

2. Medium build ($50k to $150k)

   A real application with several user roles, multiple integrations, a proper admin panel, and some workflow logic: a SaaS MVP, a marketplace, an operations platform. Timelines run one to a few months. This band covers the majority of serious small business software.

3. Complex platform ($150k to $500k+)

   Multi-sided platforms, real-time systems, heavy integrations, regulated data (health, finance), or anything with strict uptime and security requirements. Timelines run several months and up. Most small businesses never need this on day one, and shouldn't pay for it before an MVP proves the model.

A caveat worth stating plainly: these bands come from software vendors, who each price their own work, not from a neutral audited survey. Clutch, which aggregates real firm-reported project data, puts the platform-wide average software project at about $132,000, but the more useful number for a small business is its typical project band of roughly $10,000 to $50,000⁵. The average is dragged up by a handful of large enterprise builds. Yours probably isn't one of them.

The 2026 rate landscape, drawn from outsourcing surveys and marketplace data, looks roughly like this⁶⁷⁸:

Two things to hold in mind. First, specialist talent (AI, security, complex data work) carries a 40 to 60 percent premium, so a build that needs it costs more per hour by design⁸. Second, the lowest rate rarely wins on total cost. Coordination overhead, rework, and quality gaps close much of the gap between a $40 offshore hour and a $150 senior hour. The right comparison isn't rate against rate; it's the fixed quote and the shipped result against each other.

Web apps

For a custom web app in 2026, expect roughly $25,000 for something simple, $60,000 to $150,000 for a medium build, and $150,000 to $350,000 or more for a complex platform¹⁰. A SaaS MVP commonly starts around $40,000 to $60,000, and a marketplace with two sides, payments, and matching typically runs $40,000 to $80,000 and up¹¹. The cost lives in the moving parts: user roles, integrations, and anything real time. Our web app development service page covers how we scope and quote these.

Mobile apps

Mobile apps run about $15,000 to $50,000 for a simple app, $50,000 to $150,000 for a moderate one, and $150,000 or more for a complex build, with a median cross-platform MVP landing around $76,000⁹. The single biggest lever is the one most owners don't know to ask about:

Only three widely-cited studies actually measured output in a controlled setting, and they disagree in a way that tells the real story:

The pattern is consistent once you stop treating these as contradictions. AI compresses cost most on simple, new, well-defined work, and least (sometimes negatively) on gnarly existing systems that a senior already understands¹²¹³. Small business builds are overwhelmingly the first kind: a fresh, focused product with a clear scope. That's exactly where AI-accelerated development by senior engineers earns a real discount.

Two cautions keep the claim honest. Adoption is near-universal (84 percent of developers now use or plan to use AI tools), but trust is falling, and 66 percent say they spend more time fixing almost-right AI output¹⁴. That's why the model that works is AI acceleration with senior human review on everything that matters, not AI in place of engineers. The saving is real; the supervision is not optional.

The numbers have been grim for a long time, and they're remarkably stable across decades of research:

Standish's CHAOS data classes only 29 percent of projects as successful, with 52 percent challenged and 19 percent failed outright¹. McKinsey and the University of Oxford, studying 5,400 large projects, found an average 45 percent budget overrun and 56 percent less value than promised, with one in six becoming a company-threatening "black swan"². And CISQ put the 2022 cost of poor software quality in the US at $2.41 trillion³. Back in 2011, 75 percent of executives already expected their projects to fail from the start⁴.

Here's the part that matters for your budget: almost every one of those failures was billed by the hour, up front, with no accountability for the result. The fixes are structural. A written scope before any money moves. A fixed quote honored even if the builder misjudged the work. Milestone billing tied to software that actually ships and runs. And any new scope re-quoted in writing before it enters the build, so the original number never quietly inflates.

This isn't custom-good, SaaS-bad. For most functions (email, accounting, CRM, help desk), off-the-shelf is obviously right, and building your own would be a waste. The question is where the line sits, and total cost of ownership over three years is the way to find it, not the day-one price.

The hidden cost of the rented path is easy to underestimate. Companies waste around $21 million a year on average on licenses nobody uses, per-employee SaaS spend keeps climbing, and in one recent year SaaS prices rose 8.7 percent, more than double consumer inflation, with roughly three-quarters of vendors raising prices¹⁵¹⁶. When a tool is central to how you make money and every product on the market only sort of fits, that recurring, rising, un-owned cost is what custom is competing against, and often beats over a few years.

A practical rule: rent off-the-shelf until a workflow is both core to your business and genuinely poorly served by everything available, then build exactly that piece custom and keep renting the rest.

Pricing model is where most of the overrun risk lives, so it's worth getting right:

• Insist on a fixed quote against a written scope. Time-and-materials billing puts all the risk on you: if the work takes twice as long, you pay twice as much, and the vendor has no reason to hurry. A fixed quote moves that risk to the builder, who then has every incentive to scope carefully and ship efficiently.
• Split the quote into milestones. Pay for each working chunk when it ships and runs, not on a calendar. The final payment should wait until the software is live and you've signed off. You should never be far ahead of what's actually been delivered.
• Get new scope re-quoted in writing. Scope creep is the quiet killer. A healthy process quotes any addition before it enters the build, so the original number never silently inflates through change orders.
• Confirm you own everything on final payment. This one is a legal detail most owners miss. Under US copyright law, if the contract is silent, the developer keeps the copyright, not you. Demand a present-tense IP assignment so the code, repositories, infrastructure, and IP transfer to you in your name, with no license and no lock-in.

If you want to see exactly how a build runs from idea to launch, and why a fixed quote can be honored even when scope is misjudged, read our custom software build methodology, or the overview on our custom software development services page.

If you already know it's a web app or a mobile app, our web app development and mobile app development pages cover the specifics of each, including the cross-platform decision that saves 30 to 45 percent on mobile.

And if you'd rather just get a real number for your specific idea, our free 48-hour build plan turns a few sentences into a written scope, a milestone breakdown, and a fixed quote, with no sales call and no obligation.