Among decisions that weigh most on a company's IT health, one stands out: choosing whether a new tool should be built custom, bought as off-the-shelf, or composed as a hybrid. Get it wrong and you pay for years in technical debt, team frustration, and consulting invoices that never end.
Over the past seven years we've worked with 200+ companies on this exact decision — both on our own products and pure consulting. The grid below has produced the fewest mistakes.
5 questions that clarify 90% of cases
1. Is the process already a competitive advantage, or do you just want to execute it well?
If the process is what differentiates you from competitors — how you compute prices, run production, build quotes — build it custom. Buying off-the-shelf levels you down. If the process is an industry standard (accounting, attendance, invoicing), buying off-the-shelf is almost always right.
2. Are there 3+ mature off-the-shelf products on the market?
If mature products exist (5+ years on the market, 1000+ customers, active community), building from scratch is highly risky. The off-the-shelf product almost certainly does 80% of what you need, and the vendor has invested hundreds of person-months you don't want to redo. If the market is fragmented or products are aimed at other geographies, the room for custom is real.
3. Do you have an internal team to maintain it?
Custom software doesn't end at go-live. It needs maintenance, updates, evolution. Without an internal team or a credible long-term partner, off-the-shelf is almost always better. Less control, but a vendor handles security patches, migrations, compatibility.
4. How much does your business change in 24 months?
Very stable businesses (traditional clinic, B&B, plumber) benefit from off-the-shelf: they change little, and the product evolves for them. Fast-evolving businesses (scaling startup, retailer pivoting models, manufacturer becoming service-based) need flexibility only custom delivers.
5. Do you have complex integration requirements?
If you have a stack of legacy systems to talk to in non-trivial ways (old ERPs, PLCs, barcode scanners, third parties without modern APIs), off-the-shelf often "can't" — and you end up doing custom or custom layers above an off-the-shelf core.
Three models that work in 2026
Model A — Pure off-the-shelf
You buy the product, accept its way of working, possibly tweak your processes to align. Predictable cost, fast time-to-value, right for commodity processes. Examples: standard ERP, basic warehouse management, payroll.
Model B — Pure custom
You build exactly what you need, from scratch or close to it. High cost, long time-to-value, maximum competitive value when process is strategic. Examples: a dynamic pricing platform for a retail chain, a MES for a manufacturer with a unique process.
Model C — Hybrid (most common choice)
Buy off-the-shelf as a base and build custom modules for distinctive parts on top. It's what we do with most grocery customers: off-the-shelf inpublish for commercial content, custom for ERP integrations and specific pricing logic. Optimal balance between control, cost and speed.
Common mistakes
- Building from scratch what already exists, because "we know our stuff". Almost always a failure.
- Buying off-the-shelf when the process is your edge: you level down to competitors.
- Underestimating custom maintenance cost: go-live is 30% of total cost over the first 5 years.
- Overestimating off-the-shelf "flexibility": the vendor always says "yes we can". Verify other customers actually do it.
Want an independent view on the decision you're about to make?
Let's talk
