Comparison · Software & technology

Existing software or custom development: buy or build?

The answer

If off-the-shelf software covers your need and the process is not what sets you apart, buy it: building it would reinvent what exists. A no-code layer or a custom integration makes sense when the market covers most but not all of it, or when several systems must talk to each other. Full development only holds up when nothing exists or when the software is your advantage — and if the problem stays manageable, changing nothing is sometimes the best decision.

Facts checked on · Updated · Market Switzerland

This analysis is available in French and English.

In brief

The trade-off
Buying is fast and costs a subscription for life, but you adapt to the software. Building fits how you work, but costs up front, takes time and needs maintenance every year.
Who it is for
Managers and team leads in Switzerland who need tooling for a business process.
The options
Keep the current way of working · Off-the-shelf software (SaaS) · A no-code layer · Existing software + custom integration · Custom development
Facts checked on
23 Oct 2026

A process costs you time or errors — spreadsheets, re-keying, emails — and you hesitate between buying existing software, configuring it yourself, or having something built that fits you.

The comparison

The comparison

Ratings from 0 to 4, from the point of view of someone in your situation — higher is better. The reasoning behind each rating is in the analysis.

The comparison
CriterionKeepExisting softwareNo-codeIntegrationBynilson serviceCustomBynilson service
Up-front costWhat you invest before first use.excellent (4/4)good (3/4)good (3/4)adequate (2/4)poor (0/4)The highest investment, before any use.
Recurring costSubscriptions, hosting, upkeep: what recurs every month.good (3/4)adequate (2/4)adequate (2/4)adequate (2/4)adequate (2/4)
Time to valueHow long before the problem is solved.excellent (4/4)excellent (4/4)good (3/4)adequate (2/4)limited (1/4)
Fit to the processHow closely the solution follows how you work.limited (1/4)The original problem remains.adequate (2/4)good (3/4)good (3/4)excellent (4/4)
RiskThe likelihood it fails, breaks or depends on one person.good (3/4)excellent (4/4)Proven by other customers; the vendor handles upkeep.adequate (2/4)Fragile as automations pile up.adequate (2/4)limited (1/4)
DifferentiationWhat the solution gives you that competitors lack.poor (0/4)limited (1/4)adequate (2/4)good (3/4)excellent (4/4)
ControlWho decides on changes, data and leaving.good (3/4)adequate (2/4)adequate (2/4)good (3/4)excellent (4/4)
ScalingCoping with more users, data and volume.poor (0/4)good (3/4)adequate (2/4)good (3/4)excellent (4/4)

The options

Keep the current way of working

Carry on with your current tools — spreadsheets, email, software already paid for — organised better.

For

  • No investment or migration
  • Nobody to train
  • You keep the flexibility of a tool everyone knows

Against

  • Lost time and errors continue
  • Does not keep up as the team or volumes grow
  • Data stays scattered
Best for
An annoying but manageable problem, few users.
Cost
The time lost every month.

Off-the-shelf software (SaaS)

Subscription software already used by other businesses for the same need.

For

  • Available now, proven by others
  • Updates, hosting and security handled by the vendor
  • Predictable cost, no up-front investment

Against

  • You adapt your process to the software
  • A per-user subscription, every month, forever
  • You remain responsible for the data you hand it
Best for
A common need that is not what sets you apart.
Cost
Price per user per month, plus onboarding.

A no-code layer

Configure a no-code tool or automations yourself on top of your existing software.

For

  • Close to your process without development
  • Quick to evolve for someone comfortable with it
  • Moderate investment

Against

  • Fragile as automations pile up
  • Depends on the person who built it
  • Limits on volume and complex logic
Best for
A need the market partly covers, with someone comfortable with these tools.
Cost
Tool subscription + set-up and upkeep time.

Existing software + custom integration

Bynilson service

Keep off-the-shelf software for the core and have what is missing built: connections, automations, a specific module.

For

  • You only pay for what is missing
  • Connects several systems no-code cannot hold together
  • The core stays maintained by the vendor

Against

  • A project to scope and pay for
  • Depends on the vendor’s APIs, which can change
  • Two suppliers to coordinate
Best for
Several systems to connect, or off-the-shelf software that covers almost everything.
Cost
Subscription + integration development + yearly upkeep.

Custom development

Bynilson service

Software built for your process, which you own.

For

  • Exactly your way of working
  • The software and data are yours
  • Can become an advantage competitors lack

Against

  • The highest investment, before any use
  • Months of development, then maintenance every year
  • Risk of rebuilding what already existed
Best for
A need nothing covers, or software that is your product.
Cost
Development + 10–20% a year for upkeep (an assumption to replace with a quote).

The analysis

Start with what exists

Most of a business’s processes — invoicing, scheduling, CRM, project management, HR — are covered by off-the-shelf software that others have already paid to develop. Buying it costs a subscription; rebuilding it costs months and endless upkeep.

Before any talk of development, do the honest exercise: try two or three products on your real case, and write down what is missing. If the list is short, configuration or a small integration solves the problem.

When building holds up

Custom development is justified in two cases: no software does what you need, or the software is exactly what sets you apart — your product, or a process that makes your advantage. Either way, you need a budget to build it and to maintain it every year.

If you are unsure how to get it built — hiring or handing it to a provider — the Hire a developer or outsource? guide compares both, and the SaaS cost calculator gives orders of magnitude for a software product.

The middle ground: no-code and integration

Between buying and building there are two more modest routes. A no-code layer assembles existing tools and leaves you in charge; it becomes fragile as automations pile up and only one person knows how they hold together.

A custom integration keeps off-the-shelf software at the centre and has what is missing built: connecting accounting to the CRM, automating a re-keying step, adding a module. It is often the right answer when several systems must talk to each other.

Your data, whatever the option

Handing your data to subscription software does not relieve you: you must make sure the vendor can guarantee its security. Your customers can also ask you for the data they gave you in a common electronic format — check that the chosen software lets you export it.

Before signing, ask in writing how you get your data back when you leave. It is the question that costs most when forgotten.

Your recommendation

Your need, your recommendation

Five questions are enough for a first answer; the rest refine it. Every rule applied is shown under the result.

0 of 5 key questions

Off-the-shelf softwareKey question
This process for your businessKey question
Who will use itKey question
The problem todayKey question
Your budgetKey question

To refine

In-house skills
Other systems

What matters to you

How much the recurring cost matters
How much an exact fit to your process matters
How much controlling the tool and data matters

Recommendation

Answer 4 more key questions: without them, a recommendation would be a guess.

Missing

  • Off-the-shelf software
  • This process for your business
  • Who will use it
  • The problem today
  • Your budget

Decision rules v1.0.0

The costs

What each option costs over 1 to 5 years

Every amount is an assumption to replace with your offers and quotes. The calculation includes the cost of changing nothing.

Your time
CHF/h

Assumption — Loaded salary or the value of a billable hour.

Keep what you have
h/month

Assumption — Re-keying, searching, fixing errors.

Off-the-shelf software
users

Assumption

CHF/month

Assumption — Assumption — use the offer of the software you are considering.

CHF

Assumption

No-code
CHF/month

Assumption

h

Assumption

h/month

Assumption

Custom integration
CHF

Assumption — Assumption — replace it with a quote.

Custom development
CHF

Assumption — Assumption — only a specification and quotes replace it.

CHF/month

Assumption

%/year

Assumption — As a percentage of the development cost — a rule of thumb, not verified data.

  • KeepCHF 48’000
    One-off costs CHF 0 · Recurring costs CHF 48’000
  • Off-the-shelfCHF 19’000
    One-off costs CHF 1’000 · Recurring costs CHF 18’000
  • No-codeCHF 31’400
    One-off costs CHF 3’200 · Recurring costs CHF 28’200
  • IntegrationCHF 44’250
    One-off costs CHF 15’000 · Recurring costs CHF 29’250
  • CustomCHF 76’000
    One-off costs CHF 40’000 · Recurring costs CHF 36’000

Cheapest over 60 months Off-the-shelf

  • Off-the-shelf becomes cheaper than Keep in month 2.
  • No-code becomes cheaper than Keep in month 10.
  • Integration becomes cheaper than Keep in month 48.
Breakdown

Keep

Time lost
CHF 48’000
Total
CHF 48’000

Off-the-shelf

Subscriptions Assumption
CHF 18’000
Onboarding Assumption
CHF 1’000
Total
CHF 19’000

No-code

No-code subscription Assumption
CHF 9’000
Building the no-code layer Assumption
CHF 3’200
No-code upkeep
CHF 19’200
Total
CHF 31’400

Integration

Subscriptions Assumption
CHF 18’000
Integration development Assumption
CHF 15’000
Integration upkeep
CHF 11’250
Total
CHF 44’250

Custom

Development Assumption
CHF 40’000
Software upkeep
CHF 30’000
Hosting
CHF 6’000
Total
CHF 76’000

What this calculation shows

  • What each option costs over time, including the cost of changing nothing.
  • When a development costs less than the accumulated subscriptions.

What it does not prove

  • That the chosen option will solve the problem: the calculation assumes it; a project can fail.
  • A guaranteed saving: every amount is an assumption to replace with your offers.
  • Team adoption: unused software saves nothing.

Typical cases

  • Small team

    Four people, a common need the market covers, hours lost every week, no development budget.

    Off-the-shelf software. The need is common and does not set you apart: buy, configure, and keep your budget for your core business.

    Why →
  • Internal process

    A team of about twenty, a partly covered need, someone comfortable with tools, no development budget.

    A no-code layer. Assemble existing tools to cover what is missing — and document what was built.

    Why →
  • Systems to connect

    Several programs that do not talk to each other, a blocking problem, nobody technical in-house, a budget of a few thousand francs.

    Existing software with a custom integration. Have the connections built, not new software.

    Why →
  • The software is the product

    Nothing exists on the market, the software is what you sell, many users, a technical team, a substantial budget.

    Custom development. Here the software is your advantage: it must belong to you.

    Why →
  • Manageable problem

    Few users, a need the market covers, an annoying but not serious problem.

    Keep the current way of working. The cost of change exceeds the gain; tidy your tools and come back if the problem grows.

    Why →

Limits of this page

  • This page names no vendor: it compares categories of solutions. Every price in the calculation is an editorial assumption to replace with your offers and quotes.
  • The yearly upkeep rate for development (10–20% of the initial cost) is a rule of thumb, not verified data.
  • The calculation assumes every option solves the problem: it prices neither a failed project nor team adoption.
  • We have not evaluated any specific software for your need.

What we do not know

The most uncertain part: what the problem really costs you today. Few businesses measure it; the calculation asks for your lost hours, and the decision often hinges on that one number.

Frequently asked questions

  • Is it better to buy software or have it built?

    Buy if off-the-shelf software covers your need and the process does not set you apart. Build only if nothing exists, or if the software is your competitive advantage.

  • How much does maintaining custom software cost?

    A rule of thumb puts yearly upkeep at 10–20% of the development cost — an assumption, not verified data. Ask for it to be priced in every quote.

  • Can no-code replace development?

    For a need the market partly covers, often. It becomes fragile as automations pile up or when only one person knows how they work.

  • Who is responsible for data in subscription software?

    You are. Swiss data protection law asks you to make sure the provider can guarantee its security; also check how to export your data if you leave.

Sources and method

Every fact on this page points to a source read on the date shown. Prices and availability are re-checked at least every 90 and 120 days.

  1. A business that hands its clients’ data to booking software stays responsible: it must make sure the provider (processor) can guarantee data security (art. 9 FADP).

    verified · read on

    Text read on the source
    Le responsable du traitement doit en particulier s’assurer que le sous-traitant est en mesure de garantir la sécurité des données.
  2. In Switzerland, a person can request, free of charge, the personal data they provided, in a common electronic format, when processing is automated and based on consent or a contract (art. 28 FADP).

    verified · read on

    Text read on the source
    La personne concernée peut demander au responsable du traitement qu’il lui remette sous un format électronique couramment utilisé les données personnelles la concernant qu’elle lui a communiquées

How this page is made

Research and writing
Danilson Ramos
Review
no external review
Decision rules
v1.0.0
What this page adds
A comparison that includes keeping what you have and no-code, published decision rules that rule out development when the market covers the need, and a one-to-five-year calculation that also prices doing nothing.

No vendor mentioned pays to appear here, and no link on this page is an affiliate link. What Bynilson sells is labelled as such and plays no part in the recommendation rules.

Research and writing assisted by AI tools; every fact was read on its source on the date shown, and the decision rules are written and tested by hand.

Read the method →