custom web app vs off-the-shelf software: how to decide
every business eventually faces some version of this question: do we buy software that exists, or do we build something tailored to exactly what we need?
the "build vs buy" decision comes up in different contexts — a booking system, an inventory tool, a client portal, a reporting dashboard, a workflow automation. the stakes are high: buying the wrong off-the-shelf software costs months of unmet expectations; building custom software costs real money up front. here's a framework for thinking through it.
start with what you actually need
the first question isn't build or buy — it's: what exactly do we need this software to do? a vague requirement ("we need a better client management system") is not enough to make this decision. you need to be specific:
- what processes does it need to support?
- who uses it and how often?
- what data does it need to manage?
- what other systems does it need to connect with?
- what does success look like in 12 months?
with specific requirements, you can actually evaluate whether off-the-shelf software meets them.
when to buy (use existing software)
the use case is common. if thousands of businesses like yours have the same need, there's almost certainly software built for it. crm (hubspot, salesforce, copper), project management (asana, notion, monday), accounting (quickbooks, freshbooks, xero), booking (acuity, calendly, jane) — these categories have mature solutions. buying almost always beats building here.
speed matters. off-the-shelf software is live today. custom development takes months. if you need a solution now, the build timeline alone is often disqualifying.
the budget is limited. a saas product at $50–$500/month is a very different risk profile from a $30,000 custom build. the saas can be switched if it doesn't work out; the custom build cannot be undone.
your core differentiation isn't in this function. if your competitive advantage isn't in how you manage your accounting or schedule appointments, using the same tools as your competitors for those functions is completely fine.
when to build custom
existing software doesn't handle your specific workflow. you've evaluated the relevant tools and they require workarounds, don't connect to your other systems, or force you to adapt your process to the software rather than the reverse. this is the clearest case for building.
the software is part of your product. if you're building a business whose value is partly delivered through software — a marketplace, a platform, a service with a client portal — the software is your product and needs to be custom.
you've validated the need and outgrown generic tools. many businesses start with off-the-shelf tools, grow, and find themselves with duct-taped workflows spanning five different apps. at some point, a custom tool that consolidates everything and fits the specific process becomes the more efficient solution.
security or data residency requirements. some industries (healthcare, government, financial services) have requirements that limit which software can be used. a custom build that runs on your own infrastructure may be the only compliant option.
the long-term cost of saas is significant. a tool at $800/month is $9,600/year. a custom tool built for $25,000 that meets your needs pays for itself in under three years, with no ongoing licence fees. this math doesn't always work, but when it does, it's compelling.
the hybrid approach: extend what exists
often the best answer is neither pure build nor pure buy. many off-the-shelf platforms have apis and extension capabilities:
- a crm with a custom integration that connects it to your proprietary workflow
- a booking platform with a custom client portal built on top of it
- a shopify store with a custom app handling specialized product configuration
this captures the speed and reliability of an existing platform while adding custom functionality where it matters. it's usually cheaper than starting from scratch and more flexible than pure saas.
the questions to ask
- does existing software handle 80%+ of our requirements without significant compromise?
- can we adapt our process to fit the software without losing important capability?
- what's the total cost of the saas over 3 years vs. the build cost?
- how fast do we need this?
- what's our tolerance for building and maintaining custom software?
the "build" decision requires honest answers to that last question. custom software requires ongoing maintenance — bug fixes, feature updates, dependency updates, security patches. it's not a one-time cost.
nanushi builds custom web applications and advises businesses on whether custom development is the right path for their specific situation. if you're at this decision point, we're happy to give you a straight assessment.