AI Readiness Checklist for Nigerian Businesses: Are You Ready to Adopt AI?
Readiness is the question most Nigerian businesses skip. They move from "AI is interesting" straight to "get a quote for a chatbot", and then discover that the chatbot has no product data to answer from, the sales team keeps its records in personal WhatsApp chats, and nobody has decided who is accountable when the bot gives a wrong price. The project stalls, the vendor is blamed, and AI gets a bad name inside the company.
This checklist exists to prevent that. It is an assessment, not an implementation plan: it tells you whether the ground is firm enough to build on, where the weak spots are, and what to fix first. Once you know you are ready, AI Implementation Checklistplement AI in a Nigerian Business gives the full step-by-step process.
What does AI readiness mean for a business?
AI readiness is the degree to which a business has the goals, data, systems, people, processes, budget, governance and infrastructure needed for an AI project to deliver value rather than stall. It is not about how advanced the technology is; it is about whether the organisation can absorb it. A readiness assessment answers three questions: should we start, where should we start, and what must we fix before starting.
Readiness is specific to the use case. A business may be fully ready to deploy an AI assistant that drafts marketing content and completely unready to automate order processing, because the second depends on data and systems the first does not. Score the checklist with one or two candidate use cases in mind.
How to use this checklist and score it
The checklist has eight areas with four items each, 32 items in total. For each item, score:
- 0 if the statement is not true for your business.
- 1 if it is partly true or true for some teams.
- 2 if it is clearly true.
Add up the totals per area and overall. The area scores tell you where the gaps are; the overall score tells you how to proceed. Involve at least two people in the scoring (for example the owner and the operations lead), because owners tend to score people and processes higher than the people who run them do.
Area 1: Business goals and use cases
- We can state, in one sentence, the business problem AI should solve (for example, "reply to WhatsApp enquiries within five minutes, day and night").
- We have identified one or two candidate processes that are high-volume, repetitive and measurable.
- We know the current cost or time of that process well enough to compare later (a baseline).
- Leadership agrees on what success would look like within six months.
Why it matters: AI projects without a defined problem become demonstrations. What Should a Nigerian Business Automate First?a scores low.
Area 2: Data readiness
- The information the AI would need (products, prices, policies, customer history, documents) exists in digital form, not only in people's heads or notebooks.
- That data is reasonably accurate and current, and someone is responsible for keeping it so.
- Customer and product records use consistent fields (for example, every product has a name, price, stock status and description).
- We know where personal data is stored and can control what an AI system is allowed to access.
Why it matters: data readiness is the most common failure point for Nigerian SMEs. An AI assistant cannot answer from a price list that lives in a manager's memory, and a forecasting tool cannot learn from sales recorded inconsistently across three spreadsheets.
Area 3: Systems and integration readiness
- Our core tools (CRM, accounting, inventory, website, WhatsApp) are cloud-based or otherwise accessible for integration, rather than isolated desktop software or paper.
- Our key systems have APIs or export functions, or we use platforms known to integrate (for example, common CRMs, payment gateways such as Paystack or Flutterwave, the WhatsApp Business Platform).
- We know who has administrative access to each system and can grant access to a developer safely.
- We have a place where an AI system's output can go (a dashboard, a CRM field, a WhatsApp inbox), not just a chat window.
Why it matters: an AI that cannot read from or write to your systems stays a toy. How to Connect AI to Your Business Databaseiness APIs describe what integration involves.
Area 4: People and skills readiness
- A named person owns the AI initiative and has time set aside for it.
- The staff who would use the AI's output have been consulted and are broadly willing.
- At least one person can document a process and evaluate a vendor's proposal (see AI Skills Nigerian Businesses Need).
- Leadership has decided what it will say about jobs and can say it honestly.
Why it matters: AI projects fail on adoption more often than on technology. A chatbot the sales team resents will be ignored; an automation nobody owns will break silently.
Area 5: Process readiness
- The target process is written down step by step, including exceptions and who handles them.
- The process is reasonably stable; it is not redesigned every month.
- We know which steps require human judgement and which are rule-based or repetitive.
- We have agreed which decisions AI may make alone and which need a person to approve.
Why it matters: automating a process nobody has documented locks in confusion. The "human in the loop" decision protects customers and the business.
Area 6: Budget and financial readiness
- We have a one-off budget for design and build, with a realistic range (see How Much Does AI Integration Cost in Nigeria?).
- We have a separate recurring budget for model usage, API calls and subscriptions, and we understand that these are typically priced in US dollars.
- We have tested what happens to the recurring cost if the naira weakens significantly.
- We have estimated the expected benefit (time saved, revenue gained, errors reduced) and it exceeds the cost within a period we accept.
Why it matters: two-currency exposure is the financial trap specific to Nigerian AI projects. A monthly cost that is fine today can become uncomfortable after a currency movement.
Area 7: Governance, risk and compliance readiness
- We have, or will write before launch, a short AI policy covering approved tools, banned data and human review (AI Policy for Nigerian Businesses).
- We understand our obligations under the Nigeria Data Protection Act 2023 for the personal data the AI would process, and have verified current NDPC guidance for our sector.
- We know what the AI must never do (quote unverified prices, give medical or legal advice, promise delivery dates) and can enforce that in design.
- Someone is accountable for reviewing AI outputs and handling complaints or errors.
Why it matters: an AI mistake in front of a customer or regulator is the business's mistake. Governance readiness is what lets you deploy with confidence rather than hope. AI Governance for Nigerian Businesses.
Area 8: Infrastructure readiness
- Our office and key staff have reliable enough internet for cloud tools, with a backup (for example, a second provider or mobile data).
- Power interruptions do not routinely stop core systems for hours; critical devices have backup.
- Staff have devices (phones or laptops) capable of running the tools they will use.
- We have basic security in place: unique accounts, two-factor authentication on key systems, and a way to revoke access when staff leave.
Why it matters: an AI system that depends on a connection nobody can guarantee, or on shared passwords, will fail in ways that look like AI problems but are not.
Interpreting your score
| Overall score (out of 64) | Readiness level | Recommended next step |
|---|---|---|
| 0–15 | Foundations missing | Do not commission an AI project yet. Fix goals, data and processes first; use AI tools individually under a simple policy |
| 16–31 | Ready for a pilot | Choose one narrow use case in your strongest area; keep it small and measurable; budget for closing the weakest area during the pilot |
| 32–47 | Ready to implement | Proceed with a scoped project; use AI Implementation Checklist; address any area scoring under 4 in parallel |
| 48–64 | Ready to scale | Run several use cases; consider an AI roadmap (AI Adoption Strategy for Nigerian Businesses) and stronger governance |
Look at area scores too. Any area scoring 0–2 is a blocker regardless of the total. A business with a total of 40 but a data score of 1 should spend a month on data before building anything that depends on it.
What changes for Nigerian businesses
Readiness frameworks written for large foreign companies assume things Nigerian SMEs cannot: clean ERP data, stable currency, always-on power. This checklist is adjusted in five ways.
- Data is usually the weakest area. Many businesses run on WhatsApp threads, Excel and paper. The good news is that data readiness for a first project is narrow: you need the product list, prices and policies in one clean file, not a data warehouse.
- WhatsApp is a system. For most Nigerian SMEs, the WhatsApp Business App is where customers live. Integration readiness therefore includes understanding the difference between the app and the WhatsApp Business Platform (API), because only the latter can be connected to AI.
- Two-currency budgeting is mandatory. Build costs are quoted in naira; model usage, APIs and most tools are billed in US dollars. Readiness includes having modelled a weaker naira.
- Infrastructure is a real area, not a formality. Backup power and a second internet route belong on the list because they decide whether an automated system actually runs.
- Regulation is broader than it looks. The NDPA 2023 applies to most businesses that hold customer data; sector rules add more for finance, health and education. Verify current requirements with the relevant regulator; this checklist is not legal advice.
Example (hypothetical): a Kano building-materials distributor scores itself
Example (hypothetical): a Kano distributor of cement, roofing sheets and tiles, with 30 staff and sales to retailers across the North West, wants AI to handle price enquiries and stock checks on WhatsApp so that its three sales staff can focus on large orders.
The owner and operations manager score the checklist together:
- Goals: 7 of 8. The problem is clear and measurable (enquiry response time, sales staff hours).
- Data: 3 of 8. Prices change weekly and live in the owner's head and a WhatsApp broadcast; stock is counted on paper daily.
- Systems: 4 of 8. Accounting is cloud-based; there is no inventory system; sales use the WhatsApp Business App on personal phones.
- People: 6 of 8. The operations manager will own it; sales staff are keen because they are overwhelmed.
- Processes: 5 of 8. Enquiry handling is understood but not written down; credit decisions are judgement calls.
- Budget: 4 of 8. A build budget exists; recurring USD cost has not been modelled.
- Governance: 3 of 8. No policy; unclear who reviews bot answers.
- Infrastructure: 6 of 8. Generator and two internet providers; some staff phones are old.
Total: 38 of 64, "ready to implement", but data and governance are blockers. The distributor decides on a four-week preparation phase: a single price and stock sheet updated by one person each morning, a simple daily stock count entered digitally, a one-page AI policy, and a rule that the bot quotes prices only from that sheet and never handles credit. Only then does it brief developers on the WhatsApp assistant, using How to Build an AI WhatsApp Chatbot in Nigeria. The figures in this example are illustrative, not a Linestech client result.
What it costs to close readiness gaps
Closing gaps is usually cheaper than a failed AI project. Indicative 2026 ranges; actual costs vary with scope, vendor and exchange rate.
| Gap | Typical fix | Indicative cost |
|---|---|---|
| Goals unclear | Half-day workshop with leadership; use-case scoring | Internal time; ₦100,000–₦500,000 with an external facilitator |
| Data not digital or inconsistent | Data clean-up, single source of truth, simple inventory or CRM tool | ₦100,000–₦1,000,000 in staff time and tools; software subscriptions extra |
| Systems not connectable | Move to cloud tools with APIs; WhatsApp Business Platform onboarding | Tool subscriptions (often USD); integration work ₦300,000–₦2,000,000 |
| Skills missing | Training and champion programme (AI Training for Nigerian Employees) | Staff time; ₦150,000–₦1,500,000 external |
| Processes undocumented | Process mapping sessions | Internal time; ₦100,000–₦400,000 facilitated |
| Governance absent | AI policy, data map, review roles | Internal time; ₦50,000–₦300,000 for policy support |
| Infrastructure weak | Backup power, second ISP, device upgrades, 2FA | Highly variable; often the best-value spend on the list |
Mistakes to avoid
- Scoring alone. Owners overestimate people and process readiness. Involve the people who do the work.
- Treating the total as the answer. A high total with a blocked area still fails. Fix any area under 3 first.
- Assessing the company instead of the use case. Readiness is different for content drafting and for order automation. Score with a specific use case in mind.
- Buying tools to "become ready". A new CRM does not create data discipline; it moves the mess. Fix the process, then choose the tool.
- Ignoring the recurring USD cost. Budget readiness means having modelled the monthly cost under a weaker naira, not just approving a build quote.
- Skipping governance because the project is small. A small chatbot can still quote a wrong price or leak a phone number. Write the one-page policy before launch.
- Waiting for perfect readiness. A score in the pilot range is a green light for a narrow, measurable first project. Readiness improves fastest through a real pilot.
Conclusion
AI readiness for a Nigerian business comes down to eight areas: goals, data, systems, people, processes, budget, governance and infrastructure. Score the 32 items honestly with the people who do the work, treat any weak area as a blocker regardless of the total, and use the overall score to decide between fixing foundations, running a narrow pilot or implementing at scale. Most businesses that score in the pilot range should start something small within a month; readiness improves faster through a measured first project than through further planning.
If your assessment shows you are ready for a pilot, or shows gaps in data and systems you want help closing before you build, Linestech can help Nigerian businesses scope, integrate and deploy AI systems that fit their actual readiness.
Frequently asked questions
How long does an AI readiness assessment take?
Using this checklist, a small or medium business can complete the scoring in a two-hour session with the owner and the operations lead, plus a few days to verify data and system facts. A more formal assessment by an external consultant typically takes one to two weeks and produces a written report and gap plan; see AI Consulting Cost in Nigeria.
Can a business that runs on Excel and WhatsApp be ready for AI?
Yes, for the right use case. Content drafting, enquiry handling from a clean price sheet, and summarising conversations need little infrastructure. Order automation and forecasting need structured data. The checklist will show a low data and systems score; the fix is a single clean sheet and, where relevant, the WhatsApp Business Platform rather than the app.
Do we need a data scientist to be AI-ready?
No. For the AI systems most Nigerian SMEs adopt (assistants, chatbots, document tools, workflow automation), readiness depends on clean business data, documented processes and someone who can manage a vendor. Data science skills matter for custom forecasting and analytics, which are usually outsourced.
What is the minimum readiness for a WhatsApp AI chatbot?
A clear scope (what the bot answers and what it hands to a human), an accurate product, price and policy sheet the bot answers from, access to the WhatsApp Business Platform, a person who reviews conversations weekly, a policy on customer data, and a recurring USD budget for messaging and model usage. Everything else can be built up over time.
Should we assess readiness before or after getting quotes from AI developers?
Before. Quotes are more accurate and comparable when you can describe your data, systems and process clearly. Developers also price uncertainty; a business that arrives with a readiness score and a gap plan typically gets tighter scopes and fewer surprises during the project.
How often should we repeat the readiness assessment?
Repeat it before each new AI use case and at least once a year. Readiness changes as you clean data, move systems to the cloud and train staff, and it can fall when key people leave or a system is replaced. A quick re-score takes under an hour once the first assessment is done.
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.
