How to Build an MVP for an Online Business
Most advice about minimum viable products assumes you are building software. An online business MVP is a different problem. Your product may be a service, a category of goods, a membership or a marketplace, and your first version will involve more manual work and fewer features than any product diagram suggests.
The test is not whether the software is impressive. It is whether a stranger can find you, pay you and receive what they paid for, while you learn enough to decide what to build next. This guide shows how to assemble that, what to buy rather than build, and what to leave out on purpose.
What an online business MVP must do
Write the transaction in one sentence before choosing any tool: who buys what, at what price, delivered how, paid through what.
For example: "Facility managers in Victoria Island buy a monthly cleaning-consumables bundle at ₦85,000, delivered every first Monday, paid by bank transfer against an invoice."
Everything in your MVP exists to make that sentence happen reliably. If a tool, page or feature does not help the transaction complete or help you learn from it, it waits. This single discipline keeps first versions small, and small is what keeps them cheap and fast.
An online business MVP is complete when a stranger who has never spoken to you can complete the sentence from start to finish, and when you can tell afterwards how much you made and whether they returned.
The five jobs your MVP stack has to cover
Every online business, whatever it sells, has to do these five things. Your MVP is whatever combination of tools does them at the lowest cost.
| Job | What it means | Cheapest credible option | When you need more |
|---|---|---|---|
| Attract | Someone who does not know you finds the offer | A landing page plus one channel (search, Instagram, a community) | When one channel stops producing enough orders |
| Order | The buyer states what they want | A form, a catalogue or a simple cart | When order errors or back-and-forth waste hours |
| Pay | Money reaches your account and is matched to the order | A payment link or bank transfer with confirmation | When reconciliation takes more than 30 minutes a day |
| Deliver | The promise is fulfilled and the buyer knows the status | Manual dispatch plus WhatsApp updates | When status questions dominate your inbox |
| Record | Customer, order and outcome are stored | A shared spreadsheet | When you cannot answer "who bought twice?" |
Build only what your current volume forces. Ten orders a week do not need an order management system. Two hundred do.
Manual, assisted, automated: choosing your stage
Online businesses mature through three stages. Most failed MVPs skipped straight to stage three.
Stage 1: Manual (weeks 1–6). You do everything by hand. Orders arrive by WhatsApp or a form, payment is by transfer, delivery is arranged personally, records live in a spreadsheet. Cost: near zero. Learning: maximum. Limit: roughly 10–30 orders a week depending on complexity.
Stage 2: Assisted (months 2–6). The website takes the order and the payment properly, automated messages confirm and update, and records flow into one place. You still handle exceptions personally. Cost: a real website build. Limit: a few hundred orders a month.
Stage 3: Automated (month 6 onward). Inventory, dispatch, invoicing, reminders and reporting run without you. Only genuine exceptions reach a human. This is where custom software and integrations earn their cost.
The decision rule: move to the next stage when a specific manual step is either costing you more than the automation would, or causing errors customers notice. Not before.
MVP stacks by business type
Indicative starting stacks. Each assumes Stage 1 or early Stage 2.
| Business type | Attract | Order and pay | Deliver | Record |
|---|---|---|---|---|
| Niche online store | Instagram plus 5 search-optimised pages | Simple storefront with card, transfer and USSD | Courier booked per order | Store orders export plus spreadsheet |
| Service business | Landing page with clear scope and price | Enquiry form plus payment link or invoice | Delivered by you, updates by WhatsApp | Spreadsheet or simple CRM |
| Subscription or membership | Landing page plus one community channel | Recurring payment setup or monthly invoice | Content or delivery on a schedule | Member list with renewal dates |
| Marketplace | Manual supply recruitment plus one demand channel | Orders by form; you pay suppliers manually | You coordinate both sides personally | Two spreadsheets: supply and demand |
| Digital products | Content or social audience | Payment link with automatic file delivery | Instant download or email | Payment records plus email list |
| Software tool | Direct outreach to 20 target users | Manual onboarding, invoice monthly | You run the process for them if needed | Spreadsheet of accounts and usage notes |
A marketplace deserves special care: the MVP is almost never a platform. It is you, matching supply and demand by hand until the pattern is obvious enough to encode.
What to include and what to leave out
A useful filter for every proposed feature: does it help the transaction complete, or does it only help it feel complete?
Include:
- One clear offer page with price, inclusions, delivery time and a refund or returns statement
- A working payment path that covers cards, bank transfer and, where relevant, USSD
- Order confirmation the customer can see, even if you send it manually
- A way for the customer to reach a human quickly, which in Nigeria usually means WhatsApp
- A basic record of every customer and order, including how they found you
- Mobile-first pages that load quickly on a mid-range Android phone
Leave out:
- Customer accounts and login, unless the product genuinely requires them
- Loyalty points, wish lists, coupons and referral schemes
- A custom checkout when a hosted payment page works
- A mobile app, unless the business cannot function without one
- A large catalogue; start with the 20 items that will produce 80% of orders
- A blog section you have no time to write
- An admin dashboard when a spreadsheet answers the same questions
- Multi-currency, multi-language and multi-branch features
Each exclusion saves build cost, reduces things that break, and shortens the time until you learn something.
The six-week build plan
- Week 1: Fix the transaction and the offer. Write the one-sentence transaction. Set the price. Define what is included and what is not. Write the delivery promise you can keep on a bad traffic day.
- Week 2: Choose buy-versus-build for each of the five jobs. Default to buying. A hosted store, a payment link, WhatsApp Business and a spreadsheet cover most stage-one needs.
- Week 3: Build the attract-and-order path. One page that explains the offer clearly and one route to place an order. Test it on a phone, on mobile data, with someone who has never seen it.
- Week 4: Wire payment and confirmation. Set up the gateway or account, test a live low-value transaction, and write the confirmation message the customer will receive. Confirm settlement timing with the provider.
- Week 5: Rehearse delivery. Fulfil five real orders, including one outside your city. Record the true cost and the true time.
- Week 6: Launch to one channel. Not everywhere. One channel you can serve properly, with a target number of orders and a date to review.
Then hold a weekly review for eight weeks: orders, contribution, complaints, time spent, repeat rate. Change one thing per week.
What an online business MVP costs in Nigeria
Indicative 2026 ranges; actual quotes vary with scope, vendor and exchange rate. Compare two or three written quotations on identical scope.
| Component | Indicative one-off | Indicative recurring |
|---|---|---|
| Domain name | ₦3,000–₦30,000 per year | Annual renewal |
| Shared hosting | ₦20,000–₦120,000 per year | Annual |
| Landing page | ₦80,000–₦400,000 | Minor updates |
| Basic business website (5–8 pages) | ₦150,000–₦500,000 | ₦20,000–₦80,000 per month maintenance |
| Simple online store | ₦400,000–₦1,500,000 | ₦30,000–₦120,000 per month |
| Custom web application (stage 3) | ₦1,500,000–₦10,000,000+ | ₦150,000–₦800,000+ per year hosting |
| Payment gateway setup | Usually free to set up | Per-transaction fees |
| Product photography and copy | ₦50,000–₦300,000 | Per new product line |
| Tool subscriptions (email, scheduling, design) | Often USD-priced | Review twice a year for rate movement |
A realistic total for a stage-one online business MVP in Nigeria is ₦150,000–₦1,500,000, weighted towards the low end if you buy rather than build and start with one channel.
What changes for an online business MVP in Nigeria
- WhatsApp is not optional. A significant share of buyers will want to confirm with a human before paying, and many will complete the order in chat. Treat WhatsApp Business as part of the MVP stack, with a catalogue, quick replies and labels.
- Payment plurality beats payment elegance. Cards alone will cost you orders. Support bank transfer with automatic confirmation, and USSD where your customers use it.
- Delivery is a product decision, not a logistics detail. Test with two or three couriers, measure by area, and price your delivery promise against real data rather than a published tariff.
- Trust signals matter more than design polish. A registered business name, a real address, a visible phone number, honest photographs of your own stock and a plain refund policy convert better than an elaborate homepage.
- Build for slow connections and mid-range phones. Heavy pages cost your customers data and cost you orders. Keep images compressed and pages light.
- Plan for power and connectivity interruptions. If one person runs the operation, a two-day outage is a two-day business closure. Arrange backup power and a second network before launch.
- Dollar-priced tools squeeze naira margins. Subscriptions, cloud hosting and AI API usage are exchange-rate exposed. Keep the list short and review it.
- Register and handle data properly. Payment gateways and courier accounts generally expect a CAC-registered business, and collecting customer details engages the Nigeria Data Protection Act 2023. Verify current requirements with the CAC and the Nigeria Data Protection Commission.
Example (hypothetical): a fabric-sourcing service in Aba
This is a hypothetical illustration, not a client result.
A trader wants to help small fashion businesses across Nigeria buy fabric from Aba without travelling.
- Transaction sentence. "Fashion businesses outside Abia buy 5–50 yards of specified fabric at a stated price plus a 12% sourcing fee, delivered by courier in 3–5 days, paid by transfer before sourcing."
- Stage 1 stack. A ₦180,000 landing page with photographs of available fabrics, a WhatsApp Business catalogue for daily stock, a payment link, courier booking per order, and a Google Sheet recording customer, fabric, yardage, cost, fee and delivery status.
- Deliberate exclusions. No cart, no accounts, no live inventory, no app. Stock changes hourly in the market, so a live catalogue would have been wrong within a day.
- First six weeks. Thirty-one orders, average order value ₦96,000. Three disputes over shade differences lead to a simple rule: a short video of the fabric before dispatch, sent on WhatsApp.
- What the MVP taught. The real bottleneck was not ordering; it was confirming shade and availability. The stage-two build therefore prioritised a daily-updated availability list with video, not a checkout.
- Stage 2. A ₦900,000 store with per-fabric availability, deposit payments, courier integration and an order status page, funded by six weeks of trading rather than by savings.
The MVP earned its keep by pointing the second version at the right problem.
What to measure in the first 90 days
Five numbers, reviewed weekly, are enough.
- Orders per week, by channel. Tells you which channel to double down on.
- Contribution per order. Price minus cost of goods, gateway fee, packaging and delivery subsidy. Revenue without this number is decoration.
- Repeat rate within 60 days. The strongest predictor of whether the business will survive advertising costs later.
- Time per order. Your capacity ceiling, and the signal for what to automate first.
- Reasons for refusal and complaint. Keep a written list. It becomes your product roadmap.
If contribution is negative, fix price or cost before spending on growth. If repeat rate is near zero, fix the product or the segment before spending on anything.
When to rebuild and when to extend
Extend the MVP when the model is working and the constraint is capacity. Rebuild when the model has changed.
- Extend if orders are growing, the transaction sentence is unchanged, and the pain is manual work. Add order management, automated confirmations, inventory or invoicing to what exists.
- Rebuild if the customer segment changed, the pricing model changed (for example from one-off to subscription), or the tool you started with cannot support the transaction you now run.
- Watch the exit cost. Before choosing any platform, confirm that you can export customers, orders and products. A cheap tool you cannot leave is expensive.
- Keep ownership. Domain, hosting, payment gateway, courier accounts and source code should be registered in the business name, not a developer's personal account.
Mistakes to avoid
- Building software before the transaction works manually. Every unsolved operational problem becomes an expensive feature request later.
- Launching on four channels at once. You will not learn which one works, and service quality will drop across all of them.
- Buying stock or capacity before the first orders. Sell first, then source, for as long as the product allows.
- Over-specifying version one. A 40-feature scope document for a business with no customers is a budget in search of a purpose.
- Choosing tools you cannot leave. Check data export before you commit, not when you are migrating under pressure.
- Ignoring the cost of your own time. If one order takes two hours, the business does not work at any volume, no matter how good the margin looks.
- Treating the MVP as permanent. It is scaffolding. Budget for the second version once the numbers justify it.
- Skipping the customer record. Orders scattered across chats mean no reminders, no reactivation and no ability to prove repeat behaviour.
Conclusion
An MVP for an online business is a commercial experiment with just enough technology attached. Define the transaction, cover the five jobs of attract, order, pay, deliver and record with the cheapest credible tools, stay manual longer than feels comfortable, launch to one channel and let real numbers decide the second version. Done this way, a Nigerian online business can be trading within six weeks for a fraction of what a full build would cost, and the build that follows will be aimed at a problem you have proved exists.
If your manual version is working and you are ready to turn it into a proper storefront, booking system or business application, Linestech helps Nigerian founders build the next version without over-building it. Share your transaction sentence and your numbers, and we can advise on the smallest build that will carry your growth.
Frequently asked questions
How is an MVP for an online business different from an MVP for an app?
An app MVP centres on a software core loop and has to work on a device and, often, in an app store. An online business MVP centres on a commercial transaction and may involve little custom software at all. Many online business MVPs are a landing page, a payment link, WhatsApp and a spreadsheet.
Can I use no-code tools for an online business MVP?
Yes, and for stage one it is usually the right choice. Hosted storefronts, form builders, scheduling tools and payment links cover most needs. Check pricing in naira terms, confirm that local payment methods are supported, and verify that you can export your data later.
How long should I stay in the manual stage?
Until a specific manual step is measurably costing you more than automating it, or is causing errors customers notice. For most Nigerian online businesses that is six to twelve weeks of real trading, or the point where orders exceed what one person can handle carefully.
Do I need a website at all to start?
Not always, but you will want one quickly. Social profiles and WhatsApp can carry the first orders. A website adds credibility, captures search demand, holds your customer data and gives you a payment path you control.
What should the MVP budget be as a share of my capital?
A useful guide is no more than a third of your available capital on the build, leaving the rest for stock, delivery, marketing and the first months of operating. Businesses that spend everything on the website have nothing left to bring customers to it.
How do I know the MVP has succeeded?
When strangers buy through a repeatable channel, contribution per order is positive, a reasonable share of buyers return, and you can name precisely which manual step to automate next. Success is clarity plus cash, not a finished product.
Should the MVP include an admin dashboard?
Rarely at stage one. A spreadsheet answers the same questions for the first few hundred orders. Build a dashboard when you know exactly which decisions it must support, which you only learn by running the business without one.
Who should build the MVP: a freelancer or an agency?
For a stage-one setup, a freelancer or a self-assembled stack is usually enough. Bring in an agency when the build involves payments, inventory, multiple roles or integrations where a failure costs money, or when you need design, testing and support handled alongside development.
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.
