Currently: Accountant
Four years across microfinance, consumer finance and digital assets. At a digital asset exchange I built the accounting function from nothing — chart of accounts, controls, statutory tax — then cut month-end reporting from a full working day to under an hour with Power Query.
What I do
Every engagement comes down to the same thing: make the numbers correct, then make producing them fast.
IFRS financial statements, management accounts, P&L, balance sheet and cash flow — built to survive an audit and readable by a non-finance founder.
Monthly reporting packs with commentary that says what changed and why, not just what the number was. Delivered on a calendar, not on request.
Budgeting, forecasting, variance analysis and scenario modelling that tells you where money is made and where it leaks.
Power Query models, pivot architectures and VBA macros that replace a day of manual assembly with a single refresh.
Data models and dashboards that give management weekly visibility instead of a monthly surprise.
Profiling, cleaning and transformation of messy source data — finding the problems, not just formatting the output.
PAYE administration, VAT computation and filing, Withholding Tax and statutory returns — built from nothing at a company that had none.
Multi-source reconciliation design, internal controls, SOP manuals and audit readiness. Including crypto wallets, exchanges and payment platforms.
Wallet, exchange and gift-card reconciliation with automated exception flagging. Most accountants have never done this. I do it monthly.
Selected work
Every file is downloadable. Open it, click into the formulas, and check the numbers yourself.
An anonymised extract from a digital asset exchange: 14,450 trades and ₦3.57bn of gross volume in one month. Every figure is a live formula, not a pasted value.
Demonstration projects on an anonymised extract. All customer, representative and account identifiers are replaced with surrogate codes, and settlement references are redacted. Monetary values are unchanged, so the analysis is real.
Business Intelligence
A single-refresh management reporting pack over 14,450 transactions and ₦3.57bn of gross volume. Replaced a manual month-end assembly that drew on the accounting system, bank statements and exchange exports. Every KPI is a live formula, so a new month drops in and all six sheets recalculate.
439 live formulas · 7 sheets · 0 errors
Download fileData Analytics
Source-to-report remediation of a 27-column extract. Found a ₦13.8m misstatement in the source summary — a formula pointing at the wrong cell — plus 90 phantom customers created by untrimmed IDs and 9 Bitcoin payouts with no blockchain hash. Ships with an audit-committee memo.
12 exceptions logged · ₦13.8m misstatement found
Download fileFinancial Analysis
Full Pareto across 36 products. Gift cards earn a 2.77% spread against 0.73% on crypto — 3.8× the margin on a smaller ticket — while crypto carries 52% of volume for 22% of profit. One product is 63% of total profit. Includes a scenario model for its disruption.
Two products = 81% of profit
Download fileManagement Reporting
Twenty representatives ranked on margin contribution rather than volume — because the two leaderboards are not the same list. The top performer by volume ranks tenth on margin. Segment mix explains most of the 78% margin spread, which changes how you'd set commission.
78% margin spread explained by mix, not skill
Download fileFinancial Modelling
Thirty days of daily performance with a 7-day rolling average, a weekday seasonality index, and a 31-day projection carrying a stated sensitivity band. Sunday runs 60% below Friday — consistently enough to be a staffing question rather than a demand mystery.
27% coefficient of variation · 60% weekend gap
Download fileInternal Controls
Three-way reconciliation of amount due against amount paid across every settlement account, with a materiality threshold. Every account agrees within tolerance — but 147 of 165 rounding variances fall against the business and one account settles 64% of all value. That is a policy gap and a continuity risk.
₦24,821 leakage quantified · 779 duplicates flagged
Download fileA Power BI report, an Excel dashboard and a set of SQL queries, each built on a separate practice or public dataset.
Excel DashboardPractice dataset
Six months of sales for a small food business — 220 orders, 15 products, five staff. An interactive pivot dashboard with slicers answers the owner's questions: best sellers, most profitable staff, best trading day, payment mix and customer rating by product. A row-by-row check found 15 Total_Sales formulas pointing at the wrong row, understating revenue by ₦15,896. Corrected, May — not January — is the strongest month. The ad-spend correlation had also picked up its own Grand Total row; fixed, it drops from 0.99 to about 0.41, too weak on six months to claim ads drive sales.
₦762,100 revenue · 43% margin · 15 formula errors fixed
Download filePower BIGuided learning project
A one-page Power BI report on a public survey of data professionals. A treemap of respondents by country cross-filters the whole page: favourite programming language, how hard it was to break into data, respondents by role, a participant count and average age, and salary and management satisfaction gauges on the survey's 0–10 scale.
8 visuals · 1 interactive page
Download fileSQL
Seven queries against a sales orders table: which cities we sell to, orders ranked by unit price, the top five, orders from Lagos, Kano and Kaduna, a ₦10,000–₦20,000 price band, a pattern search for oil products, and a headline summary of order count, revenue and average unit price.
-- Headline summary
SELECT
COUNT(order_id) AS total_orders,
SUM(unit_price * quantity) AS total_revenue,
AVG(unit_price) AS average_unit_price
FROM orders;7 queries · commented and runnable
Download fileNo projects in this category yet.
Capability, demonstrated
Two charts drawn from the work above. The first shows why every total gets rebuilt from its source rows before anyone reads it.
Fifteen Total_Sales formulas pointed at the wrong row. Fixing them moved the best month from January to May.
| Month 2024 | Recorded | Corrected |
|---|---|---|
| Jan | ₦158,104 | ₦154,100 |
| Feb | ₦99,200 | ₦103,500 |
| Mar | ₦96,100 | ₦96,600 |
| Apr | ₦113,500 | ₦107,800 |
| May | ₦145,400 | ₦160,400 |
| Jun | ₦133,900 | ₦139,700 |
| Total | ₦746,204 | ₦762,100 |
Minutes to produce the monthly pack at a digital asset exchange, before and after the Power Query model went live.
Lagos, Nigeria
About me
I'm an accountant and financial analyst with over four years across microfinance, consumer lending, multi-branch healthcare and digital assets. I started inside regulated institutions where the process was rigid and the deadlines were absolute, and I now work in a company that had no accounting structure at all when I arrived.
That second kind of work is where I do my best. At my current employer, a digital asset and gift card exchange, I built the finance function from a blank page — the chart of accounts, the reporting framework, the internal controls, the statutory tax processes. Then I automated the month-end close, cutting the reporting pack from a full working day of manual assembly to under an hour with Power Query.
Building a chart of accounts from nothing teaches you why every control exists. It also teaches you that most reporting problems are not reporting problems — they are capture problems that only become visible at month end. So I tend to work backwards from the report to the point where the data is created, and fix it there.
What I'm most useful for: a finance function that has outgrown its spreadsheets, a month-end that takes too long, a reconciliation nobody trusts, or a management pack that reports the wrong number because a formula points at the wrong cell. I've found and corrected exactly that — a ₦13.8m misstatement sitting in a summary block nobody had reconciled to source.
I'm completing the ICAN (ACA) qualification, expected 2027, and I'm open to remote roles and consulting engagements.
Digital asset & gift card exchange · Lagos
Hospital Assist Nigeria
Rosabon Financial Services
LAPO Microfinance Bank
Edexcel University, Benin Republic
Capabilities
Grouped by discipline with years of hands-on use. No percentages — a number like "Excel 95%" invites the question "95% of what?"
Writing
Ten pieces on problems I actually ran into this year — a misstated summary, a month-end that took a day, a reconciliation with no audit trail. Written from the work, not from theory.
A formula pointed one cell too far. Nobody had reconciled the summary to the transaction rows in months. Here is how to catch it in ten minutes.
The manual steps that eat the time are almost always the same four. A walkthrough of replacing them with one Power Query refresh.
On-chain settlement, exchange statements and gift-card logs do not arrive in a format your ledger accepts. A practical method that survives an audit.
The top performer on our volume leaderboard ranked tenth on margin. What that does to commission design, and how to fix the measure.
What to decide first, what to leave flexible, and the three decisions that are painful to reverse later.
An untrimmed customer ID created 90 customers that did not exist, and quietly distorted every per-customer metric downstream.
What is due, when, and the automated schedules that stop you missing it — written for a company that has no finance function yet.
The point at which a spreadsheet stops being the right tool, and what moving to a proper model actually buys you.
What you can honestly extract from a short history, what you cannot, and how to present a projection without implying false precision.
Not the hardest task. Not the most annoying one. The one that runs most often and fails most expensively.
Available for work
Open to remote roles and consulting engagements. Based in Lagos on WAT (UTC+1), with comfortable overlap across GMT, CET and US Eastern hours. Happy to be engaged directly, as a contractor, or through an Employer of Record such as Deel or Remote.com — which lets companies hire in Nigeria without setting up a local entity.