The build vs buy decision comes down to one question: is your business process standard enough to fit inside someone else's software? If yes, buy — platforms like Salesforce, NetSuite, and Shopify exist because most businesses run similar workflows. If your process IS your competitive advantage — the thing that makes your company win — then forcing it into a platform designed for the average case costs more in workarounds, integrations, and lost efficiency than building software that matches it exactly.
What are the 7 questions that determine build vs buy?
- Does an off-the-shelf product handle 80%+ of our requirements without customization?
- Is our process standardized across the industry, or unique to how we operate?
- What is the total cost of ownership for the platform over 3-5 years (not just license fees)?
- How much are we currently spending on workarounds — spreadsheets, manual processes, middleware?
- Does the platform's data model match our data model, or will we fight it constantly?
- Can we accept the platform's release schedule and roadmap, or do we need control over when features ship?
- Is there a compliance or data residency requirement that eliminates hosted platforms?
If the answer to question 1 is yes and questions 2-7 do not raise red flags, buy. If the answer to question 1 is no — or if questions 3-7 reveal hidden costs that close the gap — build.
When does buying off-the-shelf make sense?
Condition | Why Buy | Example |
|---|---|---|
Standard workflow | Platform handles it out of the box | Basic CRM, project management, accounting |
Low user count (<20) | License costs stay manageable | Small sales team on HubSpot |
Speed to deploy matters | Weeks vs months | MVP validation, seasonal launch |
Ecosystem integrations needed | Pre-built connectors save $50K+ | Marketing stack with 10+ tools |
Vendor has deep domain expertise | Platform knows your industry better than a dev team would | Healthcare EMR, legal case management |
Buying works when the platform was designed for a business like yours. The closer your process maps to the platform's assumptions, the more value you extract from its features, updates, and ecosystem.
When does building custom software make sense?
Condition | Why Build | Example |
|---|---|---|
Process IS the advantage | Software should match your process, not the other way around | Custom manufacturing workflow, proprietary pricing engine |
Platform fit is <60% | More time customizing than using | Salesforce with 40+ custom fields and 15 custom objects |
TCO exceeds build cost by year 3 | License + admin + consultant + apps > build + maintain | $150K/year SaaS vs $200K build + $30K/year maintain |
Data model mismatch | Platform forces compromises on how data is structured | Multi-entity financial reporting, complex product catalogs |
Compliance requires data control | Hosted platforms cannot meet data residency or audit requirements | HIPAA, ITAR, government contractors |
Vendor lock-in risk | Platform vendor can change pricing, features, or terms unilaterally | Oracle, Salesforce price increases |
Building works when the gap between what the platform does and what you need creates ongoing costs — in admin time, consultant hours, or manual workarounds. The custom build eliminates those recurring costs permanently.
What does build vs buy actually cost over 5 years?
Scenario | Buy (5-year TCO) | Build (5-year TCO) | Winner |
|---|---|---|---|
Standard CRM, 10 users | $150,000–$250,000 | $200,000–$300,000 | Buy |
Complex CRM, 50 users, custom workflows | $500,000–$900,000 | $250,000–$400,000 | Build |
Standard ecommerce, <$5M GMV | $200,000–$350,000 | $250,000–$400,000 | Buy |
B2B ecommerce, custom pricing, ERP sync | $400,000–$700,000 | $250,000–$450,000 | Build |
Standard accounting/ERP, single entity | $300,000–$500,000 | $350,000–$500,000 | Buy (or tie) |
Manufacturing ERP, non-standard processes | $500,000–$2,000,000 | $250,000–$500,000 | Build |
The pattern is consistent. When the workflow is standard, buying wins because the vendor amortizes development cost across thousands of customers. When the workflow is non-standard, buying loses because the customization cost is borne entirely by you — and it recurs every year.
What hidden costs do most build-vs-buy analyses miss?
Customization cost on bought platforms is the most overlooked expense. Salesforce implementations average $150,000-$500,000 in consulting fees for customization alone — on top of licensing. If you are customizing more than 30% of the platform, you are paying twice: once for the platform you are not using, and once for the custom work.
Integration tax grows every year. Each new tool your platform connects to requires middleware, API maintenance, and monitoring. A company with 8 integrations typically spends $20,000-$50,000 per year maintaining connections that break when any vendor pushes an update.
Vendor lock-in has a real dollar cost. Migrating away from Salesforce, SAP, or Oracle takes 6-18 months and costs $100,000-$500,000+. That migration cost is implicit in every year you stay — it is the exit fee you have already committed to paying.
Custom software has its own hidden costs. The biggest is key-person dependency — if the original development team is unavailable, onboarding a new team takes 4-8 weeks and costs $15,000-$30,000 in ramp-up. Mitigate this with documentation, code standards, and a maintenance contract.
The cost of doing nothing is the one nobody calculates. If your team spends 10 hours per week on manual workarounds that software would eliminate, that is $25,000-$50,000 per year in labor — every year the decision is delayed.
What is the right process for making the build vs buy decision?
- Map your current process exactly — every step, every handoff, every exception. Do not simplify it.
- Demo 3 platforms. For each, document what fits, what requires customization, and what is impossible.
- Calculate true 3-year TCO for the best platform option: license + admin + consultants + apps + integrations + training.
- Get a scoped estimate for custom build + 3 years of maintenance.
- Compare the two numbers. If they are within 20%, build — because custom software appreciates (you add features) while platform costs only inflate.
In projects where clients came to us after a failed platform implementation — CRM rebuilds after Salesforce, ERP rebuilds after SAP Business One, ecommerce rebuilds after Shopify Plus — the common pattern is the same. The platform worked for the first 12-18 months. Then the business outgrew it, customization costs accelerated, and by month 30 the total spend exceeded what a custom build would have cost. The earlier the build decision is made, the less is wasted on platform workarounds.
Build vs buy is a math problem, not a philosophy question. Run the numbers over 3 years, include every cost — not just license fees — and the answer is usually clear.
Need a team to build this for your business?