Software development costs in Switzerland.

The price range follows not the size of the idea but the number of people and systems that have to take part. A tool for one team is cheaper than an application for customers, which is cheaper than a platform where two parties meet. The frames below are market figures from published price overviews of Swiss development firms for 2026, rounded and deliberately wide:

• Internal tool or dashboard (CHF 30,000 to 60,000): one team, one role, few interfaces, no public use. The cheapest way into bespoke software — and usually the fastest to pay for itself, because saved working time is directly measurable.
• Business web application (CHF 60,000 to 120,000): several roles, customer access, a connection to accounting or CRM, permission management, audit logging. This is the most common category among Swiss SMEs.
• Multi-party platform (CHF 80,000 to 300,000): suppliers and buyers, payments, disputes, ratings, moderation. Here the effort sits not in the interface but in the rules behind it.
• Mobile app for iOS and Android (CHF 60,000 to 150,000): two platforms, store reviews, device differences, plus a server behind it. For most ventures a browser-based application is the cheaper and faster route to the same benefit.
• Validation prototype (with us, from CHF 10,000): one use case, real users, real payment — deliberately not a product. This is the only figure on the list that is a binding entry price from DLM Digital.

If you place your own application between two of these lines, you are usually right. If you think it sits well below them, you have generally overlooked one of the cost drivers in the next section — most often the interfaces. Websites and company sites follow different, considerably lower frames; those are set out on web design prices and in the guide what a website costs in Switzerland.

Four items decide the price of a software project, and none of them is the number of screens. Get these four under control and you have the budget under control.

• Integrations. Every connection to a third-party system — accounting, ERP, inventory, payment, shipping, calendar, identity provider — is its own small project with someone else's documentation, someone else's error messages and someone else's schedule. Two interfaces cost more than twice as much as one, because errors propagate between them. This is the item that blows up quotes most often.
• Roles and permissions. As soon as different user groups see and may do different things, it is not the number of features that grows but the number of combinations that have to be built, tested and documented. Five roles with a dozen permissions each produce a test matrix nobody works through on the side.
• Migrating existing data. Old data is incomplete, contradictory, duplicated and stored in formats nobody documented. Migration is rarely technically hard and almost always time-consuming, because every exception has to be decided individually — and only the client can decide. This is the item most regularly underestimated in quotes.
• Compliance and data protection. For Swiss applications, at minimum the revised Federal Act on Data Protection: a record of processing activities, a deletion policy, data processing agreements, choice of server location, right of access. Sensitive personal data — health, biometrics, criminal records — adds considerably more, and in licensed sectors such as finance the legal assessment comes before the first line of code.

Two further factors work indirectly but powerfully: the availability you require — an application that may go down at night is considerably cheaper than one that may not — and the number of decision-makers on the client side. Projects with five equally weighted opinions cost more than projects with one accountable person, regardless of the technology. The technical foundation itself — a headless CMS, say, or a framework such as Next.js — affects the price far less than sales conversations suggest.

Because two quotes against the same brief almost never describe the same project — the difference in price is usually a difference in scope that only becomes visible once you ask. Five causes explain almost every large gap.

• Different scope of delivery. Testing, acceptance, training, documentation, migration, a month of post-launch support: any of these can be missing. It is then missing from the price, not from the project — and reappears later as a change request.
• Different assumptions about your involvement. Who supplies content, test data, decisions and sign-offs changes the effort considerably. A low quote often assumes, without saying so, that your people will do a lot themselves.
• Different locations and rates. Swiss development sits at roughly CHF 150 to 250 an hour, offshore suppliers well below that. The difference is real and legitimate — but it requires someone on your side who writes requirements precisely and signs off results. That coordination effort appears in neither quote.
• Different treatment of risk. A fixed price contains a margin for everything that could go wrong. A time and materials quote does not — instead you carry the risk. Both figures are honest, they are simply not comparable.
• Deliberately low entry pricing. Some suppliers win on price and make their margin on change requests. You can spot it by a quote with no exclusion list and no named third-party systems.

The practical response is unspectacular: ask the cheapest supplier what the quote does not include, and the most expensive which item could be cut. The answers sort the field faster than any price comparison. How a clean selection process runs overall is set out under choosing a digital agency in Switzerland.

Quotes become comparable not through negotiation but through an identical brief and five mandatory figures you request from every supplier in the same format. It takes a few hours and is the most economical work in the whole procurement process.

• Working time included. Person-days or hours, broken down by concept, design, development, testing and project management. Two offers at CHF 80,000 can mean 250 hours or 450.
• An interface list with names. Not "connection to your systems" but "Bexio, Abacus, Stripe, Microsoft Entra ID" — together with who provides the credentials and the test environment.
• Ownership of the data migration. Who cleans the legacy data, who decides on exceptions, how many runs are planned and what an additional one costs.
• Acceptance and training. What counts as accepted, how many rounds of corrections are included, how long the warranty runs, which response times apply afterwards.
• Cost of the first year of operation. Hosting, licences, maintenance, support — as a separate figure, not a footnote. Only that sum plus the project price is the number to compare.

Add one question that reveals a surprising amount: "What is explicitly not included?" Suppliers who answer with a precise list have understood your project. Suppliers who dodge it have not thought it through yet — and will do so during delivery, at your expense. Also insist on a cost ceiling where work is billed by time, and on weekly rather than monthly transparency about consumption.

Operations commonly cost an order of magnitude of around 15 per cent of the development effort per year — a market figure from Swiss suppliers, not a promise from us, and an item almost every budget forgets at first. It breaks down into five parts.

• Security and updates. Libraries and runtimes receive fixes continuously. Leave them for two years and you pay to catch up as a project rather than as maintenance.
• Hosting and services. Servers, database, storage, payment provider, email delivery, mapping services. These costs grow with usage — for a successful application that is a good problem, but it is a problem that has to appear in the budget.
• Backups and recovery. Not only creating backups but actually rehearsing recovery at least once a year. An untested backup is a hope, not a procedure.
• Small adjustments. VAT rates, new forms, changed processes, one more field. Over a year those add up to more effort than expected, reliably.
• Support. Who takes reports, how fast is the response, what happens outside office hours. Every promised response time has a price, whoever meets it.

Never budget with the project price alone, then. Budget with the total cost over three to five years. Seen that way, expensive but low-maintenance solutions lose their sting — and cheap offers with no operating concept lose their appeal. That calculation is also the most honest basis for comparing custom software with an off-the-shelf product carrying per-user licence fees.

Bespoke software is the wrong decision in most cases — saying so costs us work, but it does not change the facts. There are five situations in which we advise against it.

• An off-the-shelf product covers 80 per cent. The missing 20 per cent is then almost never worth the tenfold cost of building your own, once maintenance, security and further development are counted. Check first whether the process can adapt to the software rather than the other way round.
• The process is not yet stable. Software sets processes in concrete. Encode a workflow that will change in six months anyway and you pay twice. Process first, tool second.
• Nobody in the company owns it. Every software project needs one person with decision-making authority and time. Without them the project slips regardless of how good the supplier is — and slips considerably.
• Demand is unproven. If the software is your product and nobody has paid for it yet, the order of events is wrong. Then a validation prototype comes first, not the full product.
• The budget does not cover operations. If you can barely finance the build, do not build. An application with no maintenance budget is a security risk after two years and a rewrite after four.

The honest order of assessment is therefore: off-the-shelf product, then off-the-shelf product with limited adjustment, then bespoke development. Custom software earns its price when the process it encodes genuinely is your competitive advantage, when standard products are already being bent with expensive workarounds, or when per-user licence fees exceed development costs over the years. That comparison is set out in full on custom software in Switzerland.

We build websites, web applications and validation prototypes from our studio at Gustav-Maurer-Strasse 23 in 8702 Zollikon, with our own development team. Our binding entry prices are CHF 10,000 for a validation prototype and CHF 3,000 to 25,000 for website and web application projects, each excluding VAT.

What we are not: a software house for large enterprise platforms with twenty developers. Our publicly visible work consists of websites, brand and e-commerce projects for Swiss SMEs, plus a web application that maps the workflow of a tax advisory practice from enquiry to booked appointment. If your venture sits in the six-figure platform range, we are the wrong supplier, and we say so in the first conversation rather than in the third month.

What we are well suited to: clearly bounded tools that solve a real bottleneck — calculators, configurators, booking and enquiry flows, customer portals, internal reporting. What such projects look like with us is shown on the pages for complex web solutions and for a bespoke calculator. Our project calculator gives a first rough estimate for your venture, and our configurator a non-binding preview.

And in case your real question is not development but the visibility that follows it: the ongoing marketing prices are published on SEO costs in Switzerland. Software nobody can find is the most expensive version of all.