Nearly half of companies that buy an off-the-shelf ERP end up switching later because it didn't fit their operations. That's not an argument against packaged ERP — it's an argument for choosing with eyes open. This guide compares packaged ERP and custom software honestly: real 2026 pricing, five-year cost of ownership, the low-code middle path, and a framework to decide. We build custom software and we run our own product, so we'll tell you when buying beats building.
The two options
Off-the-shelf (packaged) ERP is pre-built software — SAP, Oracle NetSuite, Microsoft Dynamics 365, Odoo, Zoho. You license it, configure it within the vendor's limits, and use it. Custom ERP/software is built from scratch around your exact workflows; nothing is pre-made, and you own the code. The first gets you running fast within someone else's mould; the second fits you exactly but takes longer to build. The right answer depends entirely on how standard — or not — your operations are.
The real cost
Most comparisons make the same mistake: they compare sticker prices, like judging two cars by the price tag while ignoring five years of fuel and maintenance. Here's indicative 2026 packaged pricing, before customisation and consulting:
| Platform | Indicative price | Best fit |
|---|---|---|
| Odoo (Enterprise) | from ~$25/user/mo + impl. $5k–$50k | SMEs wanting flexibility on a budget |
| Zoho One | ~$37/user/mo | Small businesses, broad but shallow |
| Dynamics 365 / NetSuite | higher, quote-based | Mid-market, ecosystem buyers |
| SAP (enterprise tiers) | ~$270–$625/user/mo | Large, complex, global operations |
| Custom build | ~$50k–$400k+ once, no per-seat fee | Unique workflows; own the asset |
The hidden costs are where the truth lives. Customising packaged ERP to fit non-standard processes runs $50,000-$500,000 — and that customisation can break with vendor updates. Integrations add more, training adds more, vendor lock-in makes leaving expensive, and subscription prices rise every year. A "cheap" packaged ERP can quietly become the costly option.
Break-even, roughly: because custom software has no per-seat licence, it tends to reach cost parity with packaged ERP somewhere around 18-36 months — then pulls further ahead as headcount grows. The more users you'll add, the stronger the custom case. Always model 5-year total cost of ownership, not day-one price.
When packaged ERP wins
- Your processes are fairly standard — you're not doing anything a generic module can't handle.
- You have a smaller team (under ~50 users), where per-seat pricing stays affordable.
- You need to be live in weeks, not months.
- A platform like Odoo already covers 80%+ of your needs with room to configure.
For many small and mid-sized businesses, a well-implemented Odoo is the pragmatic starting point. Don't custom-build what you can configure in a fraction of the time.
When custom wins
- Your core workflows genuinely can't be adapted to any standard platform — and that difference is part of your edge.
- You're paying for many seats, where subscriptions eventually exceed the cost of owning a system.
- Packaged customisation keeps breaking on updates, or you're maintaining a tangle of workarounds.
- You need deep, proprietary integrations or strict compliance the market doesn't serve well (UAE FTA handling is a classic case).
- You want to own the asset and control the roadmap — or even licence it to others.
The low-code third path
It's not strictly build-or-buy any more. Low-code and no-code platforms let you assemble custom-fit workflows far faster and cheaper than traditional development — custom fit at closer to off-the-shelf speed. A large share of new business apps in 2026 are built this way. The trade-offs are real: platform limits, performance ceilings, and a different flavour of lock-in. But for the right problem, low-code is a genuine middle option worth evaluating alongside the other two — and a good partner will tell you when it fits.
A decision framework
Run your situation through four questions:
- How standard are your processes? Standard → packaged. Genuinely unique → custom (or low-code).
- How many users, growing how fast? Many and growing → custom's no-per-seat economics win over time.
- What's the 5-year total cost each way? Include licences, customisation, integrations, lock-in and annual increases.
- How much does fit and ownership matter? The more your operations are your advantage, the more custom makes sense.
Still deciding between building and buying more broadly? Our custom vs off-the-shelf guide zooms out, and the cost guide helps you budget a build. If custom is the answer, tell us about your operations — and if Odoo is the smarter move, we'll say so.