Fatolu Adedoyin Juliana

⛶ Expand

1 / 1

Retail analytics

ExcelSQLDashboard design

Coffee Shop Sales Performance Analysis

A coffee shop with 149,116 transactions across 6 months and three New York locations had no clear visibility into which products, time windows, and locations were driving profitability versus simply driving volume. The business was making operational and marketing decisions based on what felt busy — not what the data showed was actually generating revenue.

55.6%

Morning revenue share

The entire business runs on a 3-hour morning window. 88,288 of total revenue happens before noon. The business is paying full-day operational costs for a morning-dependent model.

3.40

Revenue per order — Premium Beans

Premium Beans generates 10x more per transaction than regular drinks — yet had only 336 transactions out of 149,116. The highest-value product in the portfolio is virtually invisible in the business strategy.

.31

Hot Chocolate revenue per order

Hot Chocolate earns 41% more per order than Brewed Chai Tea (.49) despite fewer total orders. The most popular product is not the most profitable — and nobody was acting on the gap.

.81

Lower Manhattan avg transaction

Lower Manhattan has the highest average spend per transaction across all three locations but the fewest total orders — signalling an untapped high-value customer base being underserved on volume.

The insight that changes the business

This coffee shop is not a coffee business — it is a morning business that stays open all day. 55.6% of 98,812 in annual revenue is generated before noon across three locations. Every staffing decision, every stocking schedule, every promotional strategy should be built around protecting and maximising that window first. One understaffed Monday morning or one ingredient running out at 9am costs more than an entire slow Tuesday afternoon. The afternoon gap — currently generating only 35% of revenue — represents a genuine growth opportunity for a targeted seasonal menu strategy, particularly iced beverages during warmer months.

What the business should do next

Immediately audit morning staffing and stocking protocols across all three locations. Introduce a targeted upsell training programme — staff should be recommending Hot Chocolate and Premium Beans at the point of ordering, not leaving customers to self-select. Test a 5–10% price increase on Brewed Chai Tea, which has strong habitual demand and low price sensitivity. Develop an afternoon seasonal menu to convert dead afternoon hours into a secondary revenue window. Lower Manhattan deserves a dedicated volume-driving campaign — the customers there spend more, they just come in less often.

⛶ Expand

1 / 1

Retail & e-commerce

ExcelRegional analysisRetail analytics

Adidas US Sales & Regional Performance Dashboard

Adidas US recorded extraordinary year-on-year growth — but leadership had no clear picture of which regions, retail partners, product categories, and sales channels were actually driving that growth versus which were benefiting from the rising tide. Without understanding the drivers, replicating the growth or defending the margins becomes impossible.

294%

Revenue growth 2020 → 2021

Revenue exploded from 82M to 17M in a single year. Growth at this scale needs a cause — understanding what drove it is worth more than any regional analysis.

42.3%

South region operating margin

The South has the highest margin of any region at 42.3% — significantly above the West's 33.2%. The West brings the most revenue but keeps the least profit per dollar earned.

39%

Men's Street Footwear revenue share

One product category carries 39% of total revenue. A single trend shift — consumer taste, supply chain disruption, competitor pricing — creates a revenue concentration risk that could be catastrophic.

39%

Online channel profit margin

Online is the most profitable channel at 39% margin versus In-store at 35.8% — yet In-store still accounts for the largest revenue share. The most efficient channel is not the most prioritised.

The insight that changes the strategy

West Gear outsells Foot Locker in total units — 625,262 versus 604,369 — but with significantly fewer transactions (2,374 vs 2,637). That means West Gear moves more units per engagement than any other retailer in the portfolio. It is the most productive retail partner by volume efficiency and it is being treated the same as every other partner. Meanwhile the South region quietly delivers the best margins in the business while receiving proportionally less strategic attention than the West, which looks better on the top line but retains far less per dollar of sales. The business is celebrating the wrong numbers.

What the business should do next

Increase investment in West Gear proportional to its output and investigate what operational factors allow it to move higher volume per transaction. Deliberately build Women's Apparel and Athletic categories — both currently underweight relative to their margin profiles — to reduce the 39% single-category concentration risk before a trend shift forces the issue. Conduct a deep-dive into the South region's margin performance and identify whether its model can be replicated in the Midwest, which currently has both the lowest revenue and among the lower margins. Accelerate the shift to online given its superior profitability — 39% margin versus 35.8% in-store represents millions in recoverable profit at scale.

⛶ Expand

1 / 1

Consumer electronics

ExcelDemand analysisCustomer segmentation

Gaming Console Sales & Demand Analysis

A gaming console company needed to understand where demand was genuinely strongest, which customer segments were driving adoption, and whether their channel strategy and marketing budget allocation matched actual buying behaviour across global regions. The surface-level numbers looked reasonable — but a deeper analysis revealed structural problems hiding beneath the growth metrics.

34M

True profit position

Revenue of 5.7M against product cost of 9.8M means the business is selling at a loss on every unit. The conversation must shift from growth to unit economics immediately.

.06

PlayStation 5 marketing ROI

PS5 generates .06 per of marketing spend — yet receives less budget than Nintendo Switch which returns only .12. Budget is misallocated away from the most efficient brand.

26.8%

Africa revenue share — ranked 1st

Africa is the highest-revenue region at 26.8% — not Europe as commonly assumed. This is an underserved market leading in performance while receiving proportionally less strategic investment.

29.6%

Teen segment — largest customer group

Teens are the largest revenue segment at 29.6% — while Hardcore Gamers, the most brand-loyal segment, represent only 13.9% of revenue and 11.6% of units sold.

The finding that reframes everything

The most important number in this entire dataset is not the revenue figure — it is the 4 million loss hiding beneath it. A business generating 5.7M in revenue while spending 9.8M on product cost is not a sales performance problem. It is a unit economics crisis. No amount of regional strategy, channel optimisation, or customer segmentation work changes the fundamental reality that the business model loses money on every console sold. Any strategic recommendation that ignores this is optimising the wrong problem. The growth conversation cannot happen until the cost structure conversation happens first.

What the business should do next

Urgently review the pricing model and cost structure — the business cannot sustain a 4M loss regardless of volume growth. Reallocate marketing budget toward PlayStation 5 which delivers 38% better marketing ROI than Nintendo Switch. Develop a dedicated Africa market strategy — it is your highest-performing region and it is currently being treated as an afterthought. Build a Hardcore Gamer retention programme — this segment is your most valuable long-term audience and it is dramatically underrepresented in your current revenue mix. Finally, address the return rate on Xbox Series X (5.01% average, highest of all brands) — at scale this compounds into a significant cost and brand credibility issue.

⛶ Expand

1 / 1

Healthcare analytics

ExcelPower QueryKPI dashboards

Lung Cancer Treatment & Survival Analytics Dashboard

Healthcare administrators and clinical teams needed to analyse 890,000 patient records across multiple countries to understand which patient profiles, treatment types, and risk factors were associated with better or worse survival outcomes across all four cancer stages. The goal was to move from uniform treatment protocols to evidence-based, patient-specific clinical decision-making.

890K

Patient records analysed

A comprehensive multi-country dataset analysed across 4 cancer stages, 4 treatment types, gender, smoking status, and asthma presence to identify the variables most predictive of survival outcome.

50/50

Gender distribution in survival outcomes

Male and female patients are distributed at near-equal rates across survival outcomes — but treatment effectiveness varies significantly by gender and cancer stage, meaning uniform protocols produce unequal results.

53%

Asthmatic patients in late-stage survival

Asthmatic patients represent 53% of late-stage cases in the survival analysis — indicating asthma as a measurable risk amplifier in advanced-stage lung cancer that clinical protocols should flag early.

458 days

Average treatment duration

Treatment duration remains near-consistent across both survived and deceased patient groups — suggesting that duration alone is not a reliable predictor of outcome and that treatment type and stage matter more.

The clinical insight

The data reveals that applying a uniform treatment protocol across all patient profiles is statistically inefficient at scale. Treatment effectiveness varies meaningfully by both gender and cancer stage — which means the same treatment delivers different outcomes depending on who receives it and when. Smoking status and asthma both show measurable influence on late-stage survival rates, yet these are risk factors that can be identified at intake. The opportunity is not in discovering new treatments — it is in using existing patient data to match the right treatment to the right patient profile at the right stage, rather than applying the same protocol to everyone and accepting the variance in outcomes as inevitable.

What clinical teams should do next

Implement patient risk stratification at intake — flagging smoking status and asthma presence as early intervention triggers rather than background information. Segment treatment planning by patient profile (gender, stage, risk factors) rather than applying uniform protocols. Use the treatment duration data to challenge the assumption that longer treatment correlates with better outcomes — the data does not support it. Allocate clinical resources toward Stage III and IV patients where treatment type differentiation has the greatest demonstrated impact on survival outcomes. Build a continuous feedback loop between treatment outcomes and intake data so that protocols can be refined over time with every new patient cohort.

‹

›

⛶ Expand

1 / 4

Statistical analysis — 3 phases

ExcelPythonANOVARegressionHypothesis testing

Global Space Exploration — 3-Phase Statistical Study

A three-phase statistical investigation into what actually determines mission success in global space exploration — testing whether budget, mission duration, mission type, and satellite category have statistically significant effects on outcomes. This project combined descriptive analysis, inferential statistics, and strategic recommendations to challenge assumptions that larger budgets and longer missions produce better results.

~0

R² — budget & duration vs success

Regression analysis confirmed that budget and duration explain virtually 0% of the variation in mission success (R² = 0.0001, p = 0.859). Resources alone do not predict outcomes.

p=0.30

ANOVA — success across eras

ANOVA found no statistically significant difference in success rates between the Early 2000s and Modern Era (F = 1.07 < F crit 3.84). Technological advancement has not changed outcome rates.

75%

Success rate — crewed & robotic missions

A two-sample t-test confirmed crewed and robotic missions succeed at statistically identical rates (p = 0.352 > 0.05). Mission type is not a determinant of success.

p=0.49

ANOVA — success across satellite types

Success rates are consistent across all five satellite categories — communication, research, spy, navigation, and weather. No satellite type carries significantly higher risk than another.

What the statistics proved

This study statistically dismantles the most common assumption in space mission planning — that bigger budgets and longer durations produce better outcomes. The data, tested across 3,000 missions using regression analysis, ANOVA, correlation analysis, and two-sample t-tests, shows that none of these variables significantly predict success. Budget correlation with success rate is r = -0.001. Duration correlation is r = -0.010. These are not weak relationships — they are essentially no relationship at all. What the data implies, but cannot directly measure, is that success is driven by factors that do not appear in the budget line: operational quality, planning rigour, international collaboration, and strategic decision-making. That is the story the statistics tell.

Strategic recommendations

Space agencies and mission planners should stop treating budget increases as a proxy for mission quality improvement — the statistical evidence does not support the relationship. Prioritise innovation, operational excellence, and pre-mission planning over raw resource allocation. Maintain a balanced portfolio of crewed and robotic missions with confidence — success rates are statistically identical, so the choice between mission types should be driven by objectives, not perceived risk differences. Diversify across satellite types freely — no category carries elevated failure risk. Foster international partnerships as a strategic priority, since the data on leading nations by mission count and success rate suggests that collaboration and knowledge transfer are more predictive of success than any resource variable measured.

Data Analyst.Strategic thinker.Results driven.

Data Analyst.
Strategic thinker.
Results driven.