How to Budget for Software Development in Nigeria
Custom software is the one technology purchase where the budget question is usually asked backwards. Owners ask "what does a custom system cost?" when the useful question is "what is the current way of working costing us, and which part of it is worth replacing first?"
Software budgets also behave differently from website and app budgets. The build is rarely a single event. Business systems grow in modules, they touch other systems, they require training and data migration, and they carry a genuine internal cost in staff time. This guide sets out a budgeting method for Nigerian businesses commissioning custom software, with indicative naira figures, a build-versus-buy framework and a worked example.
What a software budget actually covers
A custom software budget has six components. Vendor quotations typically address only the second.
- Problem definition. Process mapping, requirements, and the decision about what should not be automated at all.
- Build. Architecture, design, development, integrations, testing, deployment.
- Data migration. Getting years of spreadsheets, paper records and legacy exports into a structure the new system accepts.
- Change and training. Getting people to use it. This is the component most often budgeted at zero and most often responsible for failure.
- Run. Hosting, licences, support, monitoring, backups, small improvements.
- Internal time. Your staff in workshops, testing sessions and parallel running. Real money, rarely counted.
A budget that names all six is defensible to an owner or a board. A budget that names only the build will be exceeded.
Step 1: Cost the problem before costing the software
The size of a software budget should be justified by the size of the problem it removes. Quantify the current cost in four categories before you speak to any vendor.
| Cost of the current way of working | How to estimate it | Typical finding |
|---|---|---|
| Staff time on manual work | Hours per week spent on re-entry, reconciliation, chasing, reporting, multiplied by loaded salary cost | Often the largest single number |
| Errors and leakage | Value of stock discrepancies, mispriced invoices, missed renewals, duplicate payments over a year | Frequently underestimated |
| Delay | Revenue deferred because quotes, approvals or dispatch take days rather than hours | Hard to measure, easy to observe |
| Lost visibility | Decisions made late or badly because no one has current numbers | Usually described qualitatively |
If four staff each lose six hours a week to manual reconciliation, that is roughly 1,250 hours a year. Put your own cost per hour against it and you have the first line of the business case. A software budget of ₦4,000,000 is not expensive or cheap in the abstract; it is expensive or cheap relative to a number like that.
Step 2: Choose the budget model — build, buy or hybrid
The budget model matters more than the vendor. Three options, with different cash profiles.
| Option | Cash profile | Best when | Watch out for |
|---|---|---|---|
| Buy ready-made SaaS | Low upfront, recurring per user per month, usually in US dollars | Your process is standard; you can adapt to the tool | FX exposure, per-user costs as you grow, limited local support |
| Build custom | High upfront, lower recurring | Your process is a genuine advantage or genuinely unusual | Underestimating the run; vendor dependency |
| Hybrid | Moderate upfront, moderate recurring | Standard functions bought, differentiating functions built and integrated | Integration cost and ownership of the joins |
The hybrid model suits most Nigerian SMEs. Buy accounting, buy email, buy storage; build the ordering, dispatch, field-service or member-management system that is specific to how you actually operate. Budget the integration work between them as a named line, because it is real and it is recurring.
A simple decision rule: build where the process is a competitive difference or where no affordable tool fits Nigerian realities such as bank-transfer reconciliation, informal customer records or offline field work. Buy everywhere else.
Step 3: Size the build budget module by module
Custom software is priced by scope, and scope is easiest to control when expressed as modules. Indicative 2026 ranges for a web-based business system; actual quotations vary with complexity, vendor and exchange rate.
| Module | Typical contents | Indicative build cost |
|---|---|---|
| Core platform and user management | Accounts, roles, permissions, audit trail, settings | ₦800,000–₦2,500,000 |
| Customer or member records | Profiles, history, notes, search, documents | ₦600,000–₦2,000,000 |
| Sales, quotes and invoicing | Pricing rules, approvals, invoice generation, payment status | ₦1,000,000–₦4,000,000 |
| Inventory or asset tracking | Stock levels, multi-location, movements, reorder alerts | ₦1,200,000–₦5,000,000 |
| Operations or workflow | Job assignment, stages, SLAs, field capture | ₦1,500,000–₦6,000,000 |
| Reporting and dashboards | Standard reports, exports, role-based dashboards | ₦500,000–₦2,500,000 |
| Integrations | Payments, accounting, SMS or WhatsApp, logistics, bank statements | ₦400,000–₦1,500,000 each |
| Mobile or offline companion | Field app, offline capture, sync | ₦1,500,000–₦6,000,000 |
Whole-system totals commonly land between ₦2,000,000 and ₦30,000,000+, which is a wide band precisely because nobody buys every module at once. The budgeting discipline is to fund one or two modules properly rather than all of them thinly. A fully working sales module that the team uses daily is worth more than six half-finished modules nobody trusts.
Step 4: Budget the run, not just the build
Annual running costs for custom business software typically sit at 15–25% of the build, and they are not optional.
| Running cost | Indicative annual figure | Notes |
|---|---|---|
| Cloud or VPS hosting | ₦150,000–₦800,000+ | Scales with users, data volume and reporting load |
| Support and maintenance | 15–25% of build | Bug fixes, updates, security patches, small changes |
| Third-party licences and APIs | Variable, often USD | Mapping, messaging, document generation, analytics |
| Backups and disaster recovery | ₦50,000–₦400,000 | Offsite backups; test restores, do not assume them |
| Security review | Periodic | Particularly where customer or health data is held |
| Internal administration | Staff time | Someone owns users, permissions and data quality |
Two lines deserve emphasis for Nigerian businesses. First, backups: power and connectivity interruptions make tested restores worth paying for. Second, internal administration: every successful system has an internal owner, and if that role is unassigned the data decays within a year.
Step 5: Budget migration, training and adoption
The gap between a working system and a used system is funded here.
- Data audit and cleansing. Deduplicating customer records, standardising product codes and correcting stock counts. Usually the longest task and frequently done by your own staff.
- Migration. Importing legacy data, validating totals, running old and new in parallel for a short period.
- Training. Role-based sessions rather than one long demonstration, plus a short written guide staff can actually consult.
- Super-users. One or two people per department who receive extra training and answer everyday questions internally.
- Post-launch adjustment. Two to eight weeks of small changes as reality meets the design.
A practical allocation is 10–20% of the build budget for migration, training and adoption combined. Systems that fail in Nigerian SMEs usually fail here, not in the code.
Comparing a five-year total: custom versus SaaS
Budget decisions should be made on a multi-year view, because the two models have opposite shapes. The illustration below assumes 25 users and is indicative only.
| Year | Custom build (indicative) | SaaS at US$25 per user per month (indicative) |
|---|---|---|
| Year 1 | Build ₦6,000,000 plus run ₦900,000 | Subscriptions plus setup and configuration |
| Year 2 | Run ₦1,100,000 | Subscriptions, plus any per-user growth |
| Year 3 | Run ₦1,200,000 plus small changes | Subscriptions, plus price increases |
| Year 4 | Run ₦1,300,000 | Subscriptions |
| Year 5 | Run ₦1,400,000 plus a refresh | Subscriptions, plus migration cost if you switch |
Convert the SaaS column into naira at a rate you are comfortable defending, and repeat the calculation at a weaker rate. That second calculation is the real comparison for a Nigerian business, because subscription costs are usually USD-denominated while your revenue is not. Custom software converts a variable foreign-currency liability into a naira asset you own, at the cost of a larger upfront commitment and responsibility for maintenance.
Neither model wins universally. The point of the table is to stop the comparison being made on year-one cash alone.
What changes for software budgets in Nigeria
- Foreign-currency exposure runs through the run budget. Per-user SaaS, cloud hosting and many APIs are USD-priced. Budget renewals with headroom and review quarterly rather than annually.
- Bank-transfer reconciliation is a real feature. A large share of Nigerian B2B payment arrives by transfer with inconsistent references. Budget for matching, proof-of-payment upload and a manual reconciliation screen; imported foreign tools rarely handle this well.
- Offline and low-bandwidth behaviour. Field staff in warehouses, on sites and in transit will lose connectivity. Offline capture with later sync is an engineering cost, and omitting it pushes staff back to paper.
- Power realities. Servers hosted on the premises need inverters, generators and maintenance. Cloud hosting shifts that cost into a recurring, USD-influenced line. Budget one or the other deliberately.
- Data protection duties. The Nigeria Data Protection Act 2023 applies to systems holding personal data, and the Nigeria Data Protection Commission publishes current obligations. Budget for access controls, retention rules and, where needed, professional advice. Verify current requirements rather than assuming.
- Sector rules. Financial, health, education and logistics systems may attract additional oversight from bodies such as the CBN, NAFDAC or relevant regulators. Confirm applicability early, because compliance features are expensive to retrofit.
- Staff turnover and documentation. Budget for written documentation and repository ownership in your own name so that a developer leaving does not become a rebuild.
Example (hypothetical): a Kano manufacturer budgeting a sales and stock system
This is an illustrative scenario, not a Linestech client project.
A food-processing business in Kano sells through 40 distributors across the north. Orders come by phone and WhatsApp, stock is tracked in three spreadsheets, and month-end reconciliation takes a week. Cost of the current problem, estimated internally: roughly 1,500 staff hours a year plus recurring stock discrepancies.
The business budgets in phases rather than commissioning everything.
| Phase | Scope | Indicative budget | Timing |
|---|---|---|---|
| Phase 1 | Core platform, distributor records, order capture, invoice generation | ₦4,200,000 | Months 1–4 |
| Phase 1 extras | Data cleansing and migration, training, two super-users | ₦700,000 | Months 3–5 |
| Phase 1 run | Cloud hosting, support retainer, SMS notifications | ₦1,050,000 per year | From month 4 |
| Phase 2 | Inventory across two warehouses, reorder alerts, dispatch notes | ₦3,000,000 | Months 7–10 |
| Phase 3 | Payment reconciliation against bank statements, distributor portal | ₦2,500,000 | Year 2 |
| Contingency | Held centrally at roughly 12% | ₦1,200,000 | Throughout |
Indicative figures for illustration only. Three budgeting choices carry the plan. Phase 1 stops at the point where daily work changes, so benefit arrives before the second invoice. Training and migration are funded explicitly rather than assumed. And reconciliation, the single most painful monthly task, is scheduled for Phase 3 only because it depends on clean order and invoice data produced by Phase 1.
How to present the budget for approval
Whether the approver is a managing director, a board or a family shareholder, the same structure works.
- State the problem in money. One slide, four numbers from Step 1.
- Present the model choice. Build, buy or hybrid, with the reason.
- Show the phased build. What each phase delivers and what it costs.
- Show the run cost separately, for five years. Approvers dislike surprises more than they dislike numbers.
- Show the comparison. Custom versus SaaS totals at two exchange-rate assumptions.
- Name the risks and the contingency. Data quality, adoption, integration dependencies, vendor continuity.
- State what happens if nothing is done. The cost of the current way of working, projected forward.
Budgets are approved when the approver can see the alternative to approving them.
Budgeting mistakes to avoid
- Funding every module thinly. Half-working systems get abandoned; one complete module gets used.
- Budgeting zero for training and migration. This is where most software investments are lost.
- Comparing custom build cost against SaaS year-one cost. Compare five-year totals, at more than one exchange rate.
- Forgetting internal staff time. Workshops, testing and parallel running are real costs against real operations.
- Omitting integration costs. Every join to accounting, payments or logistics has a build cost and a maintenance cost.
- No named internal owner. Without one, permissions, data quality and adoption drift within months.
- Ignoring exit and ownership. Source code, repository access, database exports and documentation belong in the contract and the budget.
- Treating the requirements document as free. Proper requirements work reduces build cost more than any negotiation on day rates.
Conclusion
Budgeting software well in Nigeria means costing the problem first, choosing deliberately between building, buying and a hybrid, sizing the build module by module, and funding migration, training and the five-year run alongside the code. Fund one module completely rather than several partially, hold 10–15% in contingency, compare custom and subscription totals at more than one exchange rate, and put ownership of code, data and documentation in the contract.
If you are preparing a software budget for approval, Linestech can help you scope the first module, produce an itemised estimate that separates build from run, and set out a phased plan that delivers a working result before the next budget cycle.
Frequently asked questions
How much should a Nigerian SME budget for its first custom business system?
A realistic first-phase budget for a single well-scoped module with migration, training and a year of running costs is roughly ₦3,000,000–₦7,000,000. Multi-module systems commonly run from ₦8,000,000 upwards. These are indicative 2026 figures; obtain two or three written quotations against one requirements document before fixing the number.
Is it cheaper to buy software than to build it?
In the first year, almost always. Over five years, the answer depends on user numbers and the exchange rate, because most SaaS is priced per user per month in US dollars. Run both totals at your current user count and at your expected count in three years, using two exchange-rate assumptions.
What percentage of the build should I set aside for support?
Plan 15–25% of the build cost per year. Systems with payments, many integrations or regulated data sit at the top of that band. Agree in writing what the retainer covers, what counts as a new change request and what response times apply.
Should the budget include a requirements document?
Yes. Paying for structured requirements work before development typically reduces total cost, because scope disputes and rework are more expensive than analysis. If a vendor offers to start building without it, treat that as a budgeting risk rather than a saving.
How do I budget for integrations with systems I already use?
Price each integration separately, and ask whether the other system has a documented API, an export file or neither. Documented APIs are cheapest; file-based exchange is workable; systems with no interface may require manual steps that carry a permanent staff cost you should also budget.
What contingency is sensible for custom software?
Ten to fifteen per cent where requirements are well documented, and up to 20% where processes are still being defined or data quality is unknown. Data migration is the most common source of overrun, so inspect your data before setting the figure.
Who should own the software budget internally?
One named person with authority to approve scope changes, ideally the manager who owns the process being automated rather than a general administrator. Shared ownership of a software budget reliably produces scope creep, because everyone can add and nobody can decline.
Sources and further reading
Figures, platform rules and regulations change. These are the primary references behind this article and the places to check before you act on it.
