Choosing business software is one of the more expensive decisions a growing company makes, and one of the easiest to get wrong. The wrong choice costs money, but the bigger cost is trust: when staff are forced into a tool that does not fit their work, they quietly go back to paper and Excel.
This guide gives Ethiopian business owners and managers a practical way to choose, from defining the problem to signing with a vendor. You do not need a technical background to follow it.
1. Start with the problem, not the product
Before you look at a single demo, write down the three problems you want to solve, in plain words and with a measurable symptom. "We need an ERP" is not a problem. These are:
- "Closing the month takes us three weeks because sales, stock and accounts are in different files."
- "We lose track of which customers owe us money."
- "Our two branches cannot see each other's stock, so we turn away sales we could have made."
These statements become your test later: any system you consider must show it can fix them.
2. Map how work really flows
List the main flows in your business and who touches each one. For most companies they are order to cash (quote, sale, delivery, invoice, payment), procure to pay (request, purchase, receipt, supplier payment), stock movement, and payroll. Ask the people doing the work, not only the managers. The gap between the official process and the real one is where most software projects fail.
3. Decide which route fits: off-the-shelf, ERP, or custom
Most options fall into three groups. None is always best.
| Route | Best for | Strengths | Watch out for |
|---|---|---|---|
| Off-the-shelf tool (a standalone accounting or POS app) | Small teams with standard needs | Fast to start, low upfront cost | You adapt your process to the tool, and data sits in separate silos as you grow |
| ERP or business management system | Growing companies with several departments, stock, or sites | One database for finance, inventory, sales and more; add modules as you grow | Needs proper implementation and training; local support matters a great deal |
| Custom software | Processes unique to you, or a product you will sell | Fits exactly; you control the roadmap | Higher upfront cost and time; quality depends on the developer and their documentation |
Many companies end up combining them: an ERP as the core, with custom pieces only where their process is genuinely different.
4. Ethiopia-specific must-haves
Software built elsewhere can look impressive and still miss the basics that matter here.
- Tax compliance. Ethiopia's standard VAT rate is 15%. The system must handle VAT on sales and purchases and produce the reports you file from.
- E-invoicing readiness. In June 2026 the Ministry of Revenues issued Directive No. 1142/2026, creating the framework for approved electronic invoicing. Valid electronic invoices will need to be registered through the Ministry's system and carry an invoice registration number and QR code. As of September 2026 no nationwide start date has been published, so ask each vendor how and when they will support it.
- Ethiopian calendar and fiscal year. The calendar has 13 months and the fiscal year starts on Hamle 1 in early July. Reports and period closing should follow your real year.
- Birr and foreign currency. If you import, you need to record costs in the original currency and in birr, and see the real cost after exchange-rate movements.
- Local support and language. Who answers when something breaks on a Monday morning? Confirm named contacts, response times, and whether printed documents and training can be provided in the language your team uses.
- Data hosting and connectivity. Ask where data is stored, how it is backed up, and how the system behaves during power or internet interruptions.
5. Evaluate vendors with the right questions
Demos are designed to impress. Use these questions to look past them:
- Can you run our own real scenario, with our products, prices and taxes, not a sample company?
- Can we speak to two customers of similar size, and visit or call them?
- Who owns the data, and can we export everything at any time in a usable format?
- What is your plan and timeline for regulatory changes such as e-invoicing, and are updates included in the price?
- What happens if we want to leave, or if your company stops supporting the product?
- Who will actually do our implementation, and how many similar projects have they delivered?
6. Look at the total cost, not the licence price
The number on the quote is only part of the cost. Ask for a three-year view that includes:
- Software licence or subscription
- Implementation and data migration
- Customization and integrations
- Training for every role, including new hires later
- Hosting, backups and any hardware
- Support and updates after the first year
- Your own staff's time during the rollout
- A firm price quoted before anyone has studied how your business works.
- No customers you can contact, or only references the vendor selects and briefs.
- Vague answers about e-invoicing and tax updates.
- Your data is hard to export, or the contract says you do not own it.
- Pressure to sign quickly with a "special price".
A cheaper system that needs constant paid changes can cost more than a fair-priced one that fits from the start.
7. Score your shortlist
A simple scorecard keeps the decision fair when several people have opinions. Give each criterion a score from 1 to 5 for each vendor, and involve the people who will use the system daily.
| Criterion | What good looks like |
|---|---|
| Fit to your process | Runs your own scenario in the demo with little or no workaround |
| Local compliance | Handles VAT correctly, with a clear plan for e-invoicing |
| Local support | Named contacts, agreed response times, reachable in Ethiopia |
| Ease of use | A storekeeper or cashier can do a task after short training |
| Data ownership | Regular backups and export of all your data at any time |
| Three-year cost | Transparent, itemized, with no surprises for updates |
| Track record | References you can call from companies like yours |
| Room to grow | New modules, users and branches can be added without starting over |
8. Pilot before you commit
If you can, test with one department or one branch for four to eight weeks before rolling out to everyone. Agree the success measures in advance, such as fewer stock differences, faster month-end, or invoices issued the same day. A pilot exposes problems while they are still cheap to fix.
9. Plan for the people, not just the software
Name an owner inside your company who has the authority to make decisions. Train each role on the tasks they actually do. Choose a date after which the old spreadsheets and paper records stop being the official record. Without that date, teams tend to run two systems forever.
Red flags to watch for
Not sure which route fits your business?
Nur Solutions works with companies in Ethiopia and East Africa on ERP and business management systems, custom software, web and mobile applications, and IT consulting. Tell us about your business and we will give you a straight opinion.
Talk to us See our services