SoftwareCrafting Logo

Decision Framework

Build vs Buy Software

Most things should be bought. The exceptions are worth knowing precisely.

Custom software is justified by three arguments, and "the tool does not do exactly what we want" is not one of them.

See the comparisonWhen we are the wrong choice
3 reasons
The only ones that justify building
40–60 seats
Where licence maths usually flips
Buy first
The default we recommend

Quick Answer

Updated August 22, 2026

Should we build custom software or buy an existing product?

Buy, unless one of three things is true: the process is a genuine competitive advantage, licence fees at your user count exceed the build cost within two to three years, or integration and workaround costs on the off-the-shelf tool already exceed building. A custom internal tool costs $18,000–$60,000 to build plus 10–15% a year to maintain, so the breakeven against a $50-per-seat product arrives somewhere around 40–60 seats. Below that, buying almost always wins, and the honest answer to "the tool does not do exactly what we want" is usually to change the process rather than the software.

Custom build

$18,000 – $60,000

Annual maintenance

10 – 15% of build

Rough breakeven

40 – 60 seats

Best for

  • Teams writing a build-versus-buy business case
  • Companies whose SaaS bill has grown past the point of comfort
  • Buyers who have been told custom is the answer and want to check

Not best for

  • Decisions already made where only justification is wanted
  • Regulated procurement with an approved-vendor constraint
  • Very early stage companies with no established process yet
Talk to a senior engineer

Side By Side

Buy off-the-shelf vs Configure and extend vs Build custom

Build vs Buy Software — side-by-side comparison
CriterionBuy off-the-shelfConfigure and extendBuild custom
Up-front costLowLow to moderate$18,000 – $60,000
Ongoing costPer-seat licenceLicence plus integration upkeep10 – 15% of build per year
Time to valueDays2 – 8 weeks10 – 24 weeks
Fit to your processApproximateGoodExact
Who improves it over timeThe vendor, freeThe vendor plus youOnly you, and only if funded
Lock-inVendorVendor plus your extensionsYour own codebase
Breaks if the vendor changes directionYesYes, worseNo

Each Option In Detail

What each one is actually good and bad at

Buy off-the-shelf

A product built for many customers, which you adapt your process to.

Per-seat or per-usage subscription

Strengths

  • Live in days, with no build risk at all
  • The vendor keeps improving it whether or not you invest
  • Support, security patching, and compliance are someone else’s problem
  • Cheapest option below roughly 40 seats in almost every category

Limits

  • Your process bends to the product rather than the reverse
  • Costs scale with headcount, sometimes uncomfortably
  • Vendor roadmap and pricing decisions are outside your control
  • Data export quality varies, and lock-in is real

Best for: Anything that is not a competitive advantage — which is most things

Configure and extend

An off-the-shelf core with custom integrations, fields, and automations around it.

Licence plus $5,000 – $25,000 of integration work

Strengths

  • Most of the fit of custom at a fraction of the cost
  • The vendor still maintains the hard parts
  • Faster than building and more flexible than plain purchase

Limits

  • Extensions break when the vendor changes their API or model
  • Accumulated customisation eventually becomes its own maintenance burden
  • It is easy to end up paying for both approaches and getting neither cleanly

Best for: When the product is 80% right and the missing 20% is well defined

Build custom

Software shaped exactly to your process, which you then own entirely.

$18,000 – $60,000 plus 10 – 15% a year

Strengths

  • Exact fit, and the process can be a genuine advantage rather than a compromise
  • No per-seat cost, so it gets cheaper per user as you grow
  • Full control over roadmap, data, and integrations
  • No vendor can change your pricing or discontinue your product

Limits

  • It only improves if you keep funding it, and most companies do not
  • You own security, compliance, and support forever
  • Ten to twenty-four weeks before it does anything at all
  • The maintenance line is permanent and frequently forgotten

Best for: Processes that are a competitive advantage, or where licence maths has flipped

Decision Guide

If this is true, choose this

Decision rules for building versus buying software
IfChooseWhy
The process is how you win against competitorsBuildBuying here means buying the same advantage your competitors can buy.
Licence fees exceed a build within two to three yearsBuildAt that point the purchase decision is the expensive one, not the build.
The tool is 80% right and the gap is well definedConfigure and extendYou get most of the fit without owning the hard parts.
You are under about 40 seatsBuyThe maths almost never works below that, however appealing custom sounds.
Nobody will own the software after launchBuyUnfunded custom software degrades into the spreadsheet it replaced.

The Honest Take

We build custom software and we will still tell you to buy

The most common conversation we have is talking someone out of building something. A custom CRM, a custom project tracker, a custom analytics dashboard — these are almost always worse and more expensive than the thing you can subscribe to today, because the vendor has spent years and thousands of customers on problems you have not thought of yet. The exceptions are real but narrow, and they turn on economics or competitive advantage rather than on the tool being annoying.

Do not hire us if any of these is true

  • A mature product already covers 80% or more of what you need
  • You are under about 40 seats and the licence cost is manageable
  • Nobody internally will own and fund the software after launch
  • The process you want to encode is not actually a competitive advantage

Still Deciding?

Send the brief and we will tell you which option fits

Including when that option is not us. A senior engineer reads every brief and replies with the lightest sensible next step rather than the largest one.

Prime Cares — doctor dashboard

Next.js dashboard with authentication and document verification.

Blood Link — donor matching platform

React, Express, and MongoDB platform with Maps-based matching.

Quick Brief

Start the conversation here

Tell us what you are weighing up. We will say which option fits and why, even when it is not us.

Your Name

Work Email

What do you need help with?

Get an honest recommendation

FAQ

Questions people ask about building versus buying software

When does building custom software make sense?

Three cases: the process is a competitive advantage, licence fees at your scale exceed a build within two to three years, or integration and workaround costs already exceed building. Outside those, buying is usually the better commercial decision.

What is the breakeven against a SaaS subscription?

Roughly 40–60 seats against a $50-per-seat product, once you include 10–15% annual maintenance on the custom build. Below that the subscription almost always wins; well above it, custom starts to pay.

What do people forget in the build case?

Maintenance and ownership. The vendor improves their product whether or not you spend anything. Your custom tool improves only when you fund it, and most companies stop funding it about a year after launch.

Is configure-and-extend a good middle ground?

Often, yes — when the product is 80% right and the gap is well defined. The risk is accumulating so much customisation that you end up maintaining a codebase anyway, while still paying the licence.

How do we avoid vendor lock-in when buying?

Check data export quality before signing, not after. Keep your own copy of critical data where you can, prefer products with real APIs, and avoid encoding core business logic inside a vendor’s automation builder.

Related Comparisons

Other decisions worth making before you commit

In-House vs Outsourcing Software DevelopmentThe comparison people run is salary versus invoice. That is not the comparison.Dedicated Team vs Project OutsourcingFixed scope buys certainty. Certainty is only available when scope is knowable.How to Choose a Software Development PartnerMost bad engagements were decided before the contract was signed.How to Compare Software Development AgenciesPortfolios are marketing. Compare on the things a bad agency cannot fake.

Once you have decided

Cost to Build a Custom CRMCost to Build a Web ApplicationHire Full Stack DevelopersHire SaaS DevelopersAll comparisons

Next steps

Where buyers go from here

The service pages behind this work, plus the tools that turn a decision into a scope.

related serviceFull Stack Web DevelopmentProfessional full-stack web applications using React, Next.js, Node.js, and PHP. Specializing in responsive, performant, and scalable solutions by SoftwareCrafting.Explorerelated serviceDatabase & Storage ArchitectureResilient data topologies for high-frequency reads, zero data loss, and infinite storage.Explorerelated serviceAPI Design & ArchitectureType-safe, schema-first API layers - from RESTful conventions to real-time RPC and graph queries by SoftwareCrafting.ExploretoolProject cost estimatorAnswer a few questions and get a budget band for your own project.ExploreproofCase studiesShipped SaaS, mobile, fintech, healthcare, logistics, and AI products.Explore
Ready when you are

Not sure
where to start?

Tell us your goal and we'll suggest the smallest, fastest way to get there. We reply in under 4 working hours.