Featured work

Retail · Excel

Reconciling three customer count methodologies for an investor-facing summary

Independent grading: 95/100(verify, opens in a new tab)

Problem

Chidi Okafor at Meridian Retail (simulated retailer) needed to explain to investors why three internal teams reported different customer counts. Marketing, operations, and finance each applied a different definition of 'customer', and the task was to replicate all three counting methodologies in Excel so the discrepancies could be explained clearly.

What I did

Each count maps to a different business rule, so I used a separate formula for each in a new sheet called Customer Count Validation. COUNTA on the customer_id column captures marketing's total of all signups with no filtering. For operations, COUNTIF on the is_duplicate column checks for the text string 'FALSE' rather than a boolean, which is the safer pattern when working with database exports. COUNTIFS then adds a second condition on total_orders greater than zero to arrive at finance's active customer figure. All three count cells carry the #,##0 number format so the table reads cleanly in a presentation context.

Finding

Marketing counted 2,331 total signups; operations counted 2,131 after removing 200 flagged duplicates; finance counted 2,105 active customers, a further 26 non-duplicate accounts that had registered but never placed an order.

Outcome

Chidi received a single summary table with live formulas and explanatory labels that pinpoints the source of each gap, giving a clear basis for the investor slide.

Next step

A useful next step would be to break the 26 zero-order non-duplicates down by acquisition channel, to see whether any single channel is driving registrations that never convert.

COUNTA, COUNTIF, COUNTIFS for multi-condition counting
PivotTable filter by boolean text field (is_duplicate = FALSE)
Sheet organisation and label-based summary table layout
Nigerian Naira number formatting with thousand separators

Retail · Excel

Data audit of a retail Excel export: sheet inventory, column mapping, and quality issue flagging

Independent grading: 91/100(verify, opens in a new tab)

Problem

Before any analysis could begin on the Meridian Retail (simulated retailer) dataset, Nura Ibrahim needed a documented baseline of what the file contained and what quality issues were present. The task was to explore Meridian_Data.xlsx, map its structure, and flag at least three genuine data quality concerns in a new Data Audit sheet.

What I did

The audit sheet lists all 16 columns in the Orders sheet and all 7 distinct order status values, verified by direct inspection. Three issues were identified and documented: the legacy_order_ref column is entirely empty across all 1,849 order rows; the dataset spans only Q4 2023, a contextual constraint worth flagging for any analysis that assumes a full year; and the Customers sheet contains 200 records where is_duplicate is TRUE, meaning deduplication is required before any customer count. Beyond the minimum three, I also noted inconsistent category formatting in the Order_Items sheet, where the same category appears in both slug format and display format.

Finding

The file has at least four documentable data quality issues: a fully empty legacy reference column, a Q4-only date scope, 200 duplicate customer records, and inconsistent category name formats across two conventions in Order_Items.

Outcome

The Data Audit sheet gives any analyst picking up this dataset a documented starting point, with the structural issues and deduplication requirement called out before any formula work begins.

Next step

Quantifying how many Order_Items rows carry the slug-format category label versus the display-format label would clarify whether a simple find-and-replace is enough or whether the inconsistency is spread unevenly across months or product lines.

Data audit documentation: column inventory and issue flagging
COUNTA, COUNTIF, COUNTIFS for multi-condition counting
Sheet organisation and label-based summary table layout
PivotTable construction with field grouping and value filtering

Retail · Excel

Monthly customer acquisition trend with cumulative growth combination chart

Independent grading: 92/100(verify, opens in a new tab)

Problem

Chidi Okafor at Meridian Retail (simulated retailer) needed a customer growth chart as the centrepiece of an investor presentation slide. The requirement was a combination chart showing monthly new customer counts alongside a running cumulative total, built from deduplicated customer data only.

What I did

The PivotTable on the Customer_growth sheet filters is_duplicate to FALSE before counting, which excludes the 200 flagged duplicate records and ensures the Grand Total of 2,131 reflects unique customers only. A cumulative column next to the pivot uses an expanding anchored range: the formula in the first month row starts with SUM($B$4:B4), and dragging it down expands only the end reference while the start stays locked at January, so each row sums from the beginning of the year through itself. The combination chart maps monthly counts to orange bars and the cumulative total to a navy line, applying Meridian brand colours to both series.

Finding

The deduplicated customer base reached a cumulative total of 2,131 unique customers across the tracked period, with monthly acquisition visible bar by bar and the overall growth trajectory shown on the line series.

Outcome

Chidi received a presentation-ready chart that shows both the pace of monthly acquisition and the cumulative scale of the customer base in one view, suitable for the investor slide.

Next step

Splitting the monthly acquisition counts by acquisition channel would show which channels drove the largest cohorts and whether any channel's contribution is declining over time.

PivotTable filter by boolean text field (is_duplicate = FALSE)
Running cumulative sum with expanding anchored range (SUM with mixed references)
Combination chart (bar plus line) with dual data series
PivotTable construction with field grouping and value filtering
Date grouping by month in PivotTable

Retail · Excel

Gross vs net October revenue calculation for finance cross-check

Independent grading: 95/100(verify, opens in a new tab)

Problem

Nura Ibrahim at Meridian Retail (simulated retailer) needed to cross-check the October 2023 revenue figure against the finance system. Specifically, she wanted both the gross total across all order statuses and the net total after stripping out cancelled, returned, and refunded orders, along with the gap between them.

What I did

Both figures required SUMIFS against the Orders sheet, filtering on order_date for October 2023 and on the total_naira column. The net revenue formula adds exclusion conditions for three status values: cancelled, returned, and refunded. Rather than hardcoding the gap, I wrote it as a live subtraction formula referencing the two SUMIFS cells, so any update to the source data flows through automatically. All three output cells carry the Nigerian Naira format with thousand separators as specified.

Finding

Gross October revenue was NGN 56,876,792.48 and net revenue after removing cancelled, returned, and refunded orders was NGN 38,450,382.35, a gap of NGN 18,426,410.13.

Outcome

Nura received a verified, formula-driven summary she could compare directly against the finance system figure, with the revenue lost to non-completed orders visible in a single cell.

Next step

Breaking the NGN 18,426,410.13 gap down by the three excluded statuses individually would show whether cancellations, returns, or refunds are the largest drag on net revenue.

SUMIFS with date-range and multi-criteria filtering
Gap calculation with live subtraction formula
Nigerian Naira number formatting with thousand separators
Sheet organisation and label-based summary table layout

Retail · Excel

Q4 revenue by order status with stacked bar chart for cancellation trend review

Independent grading: 95/100(verify, opens in a new tab)

Problem

Nura Ibrahim at Meridian Retail (simulated retailer) wanted to see whether the October cancellation pattern persisted across the full Q4 2023 period or was a one-off. The task was to build a PivotTable breaking revenue by order status and month for October through December, then visualise the split as a stacked bar chart.

What I did

The PivotTable uses Orders as its source, with order status on rows, order_date grouped by month on columns, and sum of total_naira as the value field. Grouping the date field by month inside the PivotTable keeps the layout dynamic rather than requiring manual column setup for each month. All seven statuses, including cancelled, confirmed, delivered, pending, refunded, returned, and shipped, appear as row labels with a Grand Total row. The stacked bar chart sits below the table and references the Status Breakdown sheet directly, so the chart updates if the pivot refreshes.

Finding

The PivotTable confirmed NGN 56,876,792 in October gross revenue and showed the full status distribution across all three Q4 months, giving Nura a direct month-by-month view of how cancellations and other non-delivered statuses compare.

Outcome

Nura can now look at each status row across October, November, and December in a single table and read the stacked chart to judge visually whether cancellation revenue share is rising or falling across the quarter.

Next step

Calculating each status as a percentage of that month's gross total would make it easier to compare relative cancellation rates across months where overall order volumes differ.

PivotTable construction with field grouping and value filtering
Date grouping by month in PivotTable
Stacked bar chart referencing PivotTable data
Nigerian Naira number formatting with thousand separators

Samuel Adeniyi

About

Skills

Featured work

Reconciling three customer count methodologies for an investor-facing summary

Problem

What I did

Finding

Outcome

Next step

Data audit of a retail Excel export: sheet inventory, column mapping, and quality issue flagging

Problem

What I did

Finding

Outcome

Next step

Monthly customer acquisition trend with cumulative growth combination chart

Problem

What I did

Finding

Outcome

Next step

Gross vs net October revenue calculation for finance cross-check

Problem

What I did

Finding

Outcome

Next step

Q4 revenue by order status with stacked bar chart for cancellation trend review

Problem

What I did

Finding

Outcome

Next step

Additional projects

Customer Analytics

Dashboard Development

Data Analysis

Data Quality

Financial Reporting