Most ERP implementations fail. Especially at small companies. And it is almost never the software's fault — the culprit is misaligned expectations, underestimated hidden costs, and picking a system that was built for someone completely different. If you are considering ERP, this article will help you walk through that decision with your eyes open.
Do you actually need ERP?
ERP stands for Enterprise Resource Planning — a system that unifies different parts of your business in one place. Accounting, inventory, orders, invoicing, production, HR. It sounds attractive. But before you go down that road, answer this honestly: does your company show any of these symptoms?
- Spreadsheet jungle. You have five different Excel files for orders, inventory, and invoices that nobody keeps in sync.
- Manual data re-entry. An order from the online store gets typed manually into accounting. The warehouse worker gets a printed slip. The customer waits for an e-mail confirmation.
- Lost orders. You find out a job "fell through the cracks" only when an angry customer calls.
- Accounting lives in its own world. Invoicing and operational data do not talk to each other. Management cannot see margins in real time.
- Staff spend hours moving data instead of doing work that actually adds value.
If you agree with at least three of these, ERP — or at minimum an integrated business system — makes sense. If you agree with one or two, you might just need better use of the tools you already have.
Practical tip: Before you start evaluating software, map your processes first. Draw how an order travels from receipt to payment. Every manual step or handoff between systems is both a potential problem that ERP solves — and a potential failure point if the implementation is done poorly.
Types of ERP for small businesses
The market offers three broad categories:
Off-the-shelf (boxed) software
The Czech market is dominated by Pohoda (Stormware) and Money S3 (Cígler Software), with ABRA Flexi as another option. These are installed on your computer or server, work offline, and keep your data on your own infrastructure. They are popular because they are built natively for Czech tax law, VAT reporting, and accounting standards — no extra configuration needed.
For businesses up to roughly 20 employees, Pohoda and Money S3 are effectively the standard. They are affordable to acquire, accountants already know how to use them, and Czech-language support is readily available.
Cloud SaaS ERP
Odoo is the most flexible option in this category — open-source core, modular architecture, and extensible. Holded or Zoho Books are lighter alternatives. You pay a monthly subscription, access everything from a browser, and updates happen automatically.
The drawback: Czech localization is often incomplete. An accountant who knows Pohoda will need to re-learn everything. Czech-specific requirements (e-invoicing, VAT control statements) need additional configuration or add-ons that cost extra.
Custom-built system
Built specifically for your processes. No unnecessary modules, no UX compromises. Higher upfront investment, but it fits exactly how your company works — not how some software vendor imagines you work.
Comparison table: Pohoda vs. Odoo vs. custom
| Criterion | Pohoda | Odoo | Custom-built |
|---|---|---|---|
| Upfront cost | CZK 8–25k | 0 (open-source) | CZK 200–800k |
| Monthly cost | CZK 0–2k | EUR 5–40/user | maintenance per agreement |
| Czech compliance | native | requires configuration | depends on implementation |
| Flexibility | low | high | maximum |
| Accountant familiarity | very high | low | low |
| Implementation time | days to weeks | months | months to a year |
| API integration | limited | extensive | per specification |
| Suitability for manufacturing | partial | yes | yes |
| Vendor lock-in | medium | low | none |
Real-world example: Kovářík s.r.o. — 12 employees, custom metal component manufacturing — used Pohoda for accounting and Excel for job management. The problem: quotes, orders, production planning, and invoicing existed in four separate worlds. After a process analysis, BASAD delivered an integrated custom system: jobs, production planning, and invoicing in one interface, with Pohoda remaining as the accounting backend. Implementation took four months; employees were up to speed within a week.
The hidden costs nobody mentions
This is where most implementations fall apart. The license price is just the beginning.
Implementation
Odoo or any other complex system does not install itself in an afternoon. You need configuration, process mapping, integration with your bank, accounting software, and online store. With Odoo Community Edition (free), expect tens of hours of work from an implementation partner. These are costs the software vendor will not put in the price list.
Data migration
You have historical data in Excel, a legacy system, or an Access database. Moving it to the new ERP cleanly and consistently is work that takes time — and costs money. Poor data migration is one of the most common causes of implementation failure.
Training
One webinar is not enough. People need time before the system becomes second nature. For more complex ERP, expect a week of active training and several months where employee productivity is lower than it was before the implementation.
Customization
No off-the-shelf software does exactly what you need. You will want an exception here, a special report there, integration with your specific shipping provider. In Odoo, customizations are done in Python and they cost money. In Pohoda, customization options are limited — you either accept what the software offers, or you look elsewhere.
Ongoing costs
Cloud SaaS looks cheap — EUR 15 per user per month. But for 10 users that is EUR 1,800 per year. Over five years you will have paid EUR 9,000 — for software you do not own and never will. Add support, updates, and any customization on top of that.
Practical tip: Ask vendors for total cost of ownership (TCO) over three and five years, not just the license price. If they refuse or cannot answer, that is a warning sign.
5 questions before you sign anything
1. Does the system handle local compliance without painful workarounds?
VAT returns, e-invoicing, control statements — this needs to work without friction. Foreign systems handle it through plugins, or not at all. Ask specifically: "How does the system process VAT control statements? Who configures it and what does it cost?"
2. How many users will actually use the system?
Not nominally — in practice. If you pay for 10 licenses but only three people use the system daily, you are paying for air. SaaS models charge per user — find out exactly who needs the system and to what extent.
3. What does integration with my online store, bank, and accounting software look like?
If the answers are not concrete — "we have a ready-made connector for Shoptet, WooCommerce, and your bank, here is the documentation" — or if you hear "that can be integrated" with no price or timeline, be careful.
4. What happens if the vendor goes bankrupt or raises prices?
For cloud solutions: can you export your data? In what format? Within what timeframe? What happens to your historical records? Read the contract, not just the marketing materials.
5. Can I speak with three reference customers from a similar industry?
Any serious vendor will provide references without hesitation. If they avoid it, or references are only big brand names without a real contact person, treat it as a warning.
Red flags when choosing an ERP vendor
These are concrete signals that should stop you:
- "We can have it running over the weekend." A weekend ERP implementation guarantees a disaster three months later.
- Fixed price without a process analysis. Nobody can quote an accurate price without understanding your processes. A fixed price without an analysis is either unrealistic or full of traps hidden in "extras."
- Vendor refuses to demo the system on your data. If they are selling you a system but cannot demonstrate it using your own business scenario, they are selling a promise, not software.
- Contract with no exit clause. If you cannot leave without a penalty fee, you are trapped.
- Nobody on the team understands local tax law. If the implementation team does not know your country's VAT specifics, accounting standards, and compliance requirements, you are paying for their education.
- "That can be added for an extra fee." If you hear this for every feature you need, the base product is insufficient and the real cost will be a multiple of the quoted price.
Real-world example: Zdravá Lahůdka — an 8-person B2C and B2B food retailer — signed a contract for a cloud ERP after a two-hour demo. License price: CZK 4,900 per month. Actual costs after one year: CZK 4,900 license + CZK 180,000 implementation + CZK 45,000 online store integration + CZK 28,000 training = over CZK 270,000. And the integration with their specific warehouse system never worked quite right. They are now migrating away from it.
When a custom-built system makes sense
A custom system is not for everyone — but for some companies it is the only sensible choice.
It makes sense when:
- Your business process is specific enough that off-the-shelf software requires so much customization that the cost approaches custom development anyway.
- You need deep integration with production machinery, external APIs, or specific data flows.
- You want a system that grows exactly with you — not one whose roadmap depends on decisions made by a vendor in Amsterdam.
- Your processes are a competitive advantage and you do not want to bend them to fit generic software.
The downside is obvious: higher upfront investment and reliance on the vendor who built the system. This is why it is essential to own the source code and have a clearly defined maintenance contract.
Summary: how to decide
| Situation | Recommendation |
|---|---|
| Small company, standard accounting, no complex processes | Pohoda or Money S3 |
| Growing company, needs integration and modularity | Odoo with a local implementation partner |
| Specific processes, need for exact fit | Custom-built system |
| Complex manufacturing or logistics | Consult before deciding |
| Company wants no vendor lock-in | Custom-built or Odoo (open-source) |
The most important piece of advice: do not underestimate the analysis phase before choosing. An hour-long consultation with an independent partner who helps you map your processes and compare options will pay for itself many times over. You are not looking for software — you are looking for a solution built for your company.
At BASAD Studios, we design and build business systems — from simple integrations of existing tools to full custom ERP solutions. If you are not sure what the right choice is for your company, get in touch or check out our business systems service.