Retail Operations · Data Analytics

Nguyen Ngoc Phuong Trinh

Retail Operations Manager — MUJI Vietnam

Six years running retail floors, from supervisor assistant to store manager. I lead the daily operation of a store where footfall is low and every decision on stock, staffing and display has to earn its place. Today I read those same stores through data.

More about me See my analytics work

Supply Chain Performance Dashboard — see the full report ↓

Years in retail operations

Roles, assistant to manager

Store led end to end

Power BI

Certified, 2026

About

Six years on the shop floor, now reading the floor through data

Nguyen Ngoc Phuong Trinh

Retail Operations Manager
MUJI Vietnam

Emailtrinhnnp21@gmail.com
GitHub@trinhnnp21-hash
Based inHo Chi Minh City, Vietnam

I manage retail operations at MUJI Vietnam, where I have spent six years moving from supervisor assistant to store manager. My work is the unglamorous half of retail: stock that arrives late, shifts that run short, a display that quietly stops selling. Those are the things behind every number on a sales report.

I currently run MUJI Le Van Viet in District 9, Ho Chi Minh City — a store with modest footfall. That constraint shaped how I work: without volume to absorb mistakes, merchandising, staffing and inventory decisions have to be made on evidence rather than instinct.

That is what pushed me into analytics. I wanted to answer why rather than stop at how much, and to answer it with evidence solid enough that someone else is willing to act on it. The supply chain project below is the first piece I took end to end that way.

I am now studying three programmes on AI for data analytics — AI agents, Excel and Google Sheets. The goal is to make this kind of analysis run on its own, instead of being rebuilt by hand every month. See what is in progress →

Skills & tools

Retail operations

Store Operations
Sales Planning
Inventory & Stock Control
Workforce Scheduling
Visual Merchandising
Team Leadership

Analytics & reporting

Power BI Desktop
DAX
Power Query
Data Modeling
Star Schema
Time Intelligence
Excel

Analytical thinking

Business Framing
MECE Decomposition
Pareto Analysis
Diagnostic Analysis
Data Storytelling

Experience

2026 — Present

Store Manager

MUJI Le Van Viet — District 9, Ho Chi Minh City

Full ownership of a low-footfall store: commercial results, inventory, staffing and floor operations. Performance here comes from the quality of each decision rather than from traffic volume.
2023 — 2025

Senior Supervisor

MUJI Vietnam — Flagship Store

Coordinated daily operations at the flagship store, the largest in the network by both assortment and footfall.
2020 — 2022

Supervisor Assistant → Supervisor

MUJI Vietnam

Started on the sales floor and moved through merchandising, stock control and shift coordination.

Education & certification

Expected Oct 2026 In progress

AI for data analytics — three programmes

AI Agent for Data Analytics & Decision Making · AI-Powered Excel for Data Analytics · AI-Powered Google Sheet for Data Analytics

Each programme ends with its own capstone and certificate in mid-October 2026.
Aug 2026

Master Analytical Thinking & Data Analysis with Power BI

The Future Analyst Academy · ID MAD14Q7E3

Completed with the Supply Chain Performance Dashboard capstone.

Selected project

Supply Chain Performance Dashboard

A Power BI report on retail supply chain performance — delivery time, return rate, shipping mode and regional logistics — connecting operational numbers to the profit they end up producing.

Seven linked report pages: an Overview of revenue and profit, down into Supply Chain (Pareto of late orders by state, delivery performance by ship mode), Product (margin and return rate by sub-category), Customers (segments and high-value accounts) and Performance (KPIs by sales representative).

All metrics are built in DAX, including MoM/YoY comparisons, Perfect Order Rate, On-Time Rate and Profit Impact of Late Delivery. Pages are connected through drill-through and tooltip pages.

Power BI Desktop
DAX
Power Query
Star Schema
Time Intelligence
Pareto Analysis
Drill-through
Data Storytelling

DatasetRetail orders, 2014 – 2017
Orders9,994
Customers793
Report pages7 + 4 tooltip / drill-through
ScopeSales · Supply Chain · Product · Customer
ToolPower BI Desktop

Revenue

2,297,201

YoY ▼ 21.9%

Profit

286,397

Profit margin 12.5%

Return rate

8.0%

YoY ▲ 17.8%

On-time rate

54.2%

Late delivery 45.8%

Avg delivery days

34.6

Avg processing 830.6 hrs

Performance score

72.4%

Revenue achievement 78.1%

Gallery

Report pages

Click any image to view it full size.

Dashboard home page — Home — Power your decisions with data
Landing page with three headline KPIs — revenue, on-time delivery rate and profit margin — over a twelve-month revenue trend.

Business analysis dashboard — Business Overview
Orders, revenue, return rate and profit margin with MoM/YoY deltas; monthly revenue–cost–profit and a quarterly performance table.

Supply chain analysis dashboard — Supply Chain
Pareto of late orders by state at a 70% threshold, with a regional table of on-time orders, return rate, perfect order rate and cost per order.

Delivery performance by ship mode — Delivery Performance
Delivery trend by ship mode, and delivery days against return rate across small, medium and large order sizes.

Product analysis dashboard — Product
Revenue and profit margin by sub-category, with return rate ranked directly against the late delivery rate of the same categories.

Customers analysis dashboard — Customers
793 customers across three segments: revenue per customer, average order value, repeat rate and return rate by segment.

Performance analysis dashboard — Performance
Scorecard by sales representative: revenue and profit achievement, perfect order rate, on-time rate and revenue ranking.

Drill-through & tooltip pages

Late orders by ship mode — Late orders · Ship mode

Orders by ship mode — Orders · Ship mode

Late orders by segment — Late orders · Segment

Profit by ship mode — Profit · Ship mode

What the data says

From numbers to decisions

Each conclusion below follows the same chain: evidence → finding → insight → recommendation. All figures come straight from the Power BI report.

Late delivery is systemic, not incidental

Evidence: Late delivery rate 45.8% · on-time rate 54.2% · average delivery 34.6 days · average processing 830.6 hours.
Finding: Almost half of all orders arrive late, and the processing stage before shipping accounts for most of the total lead time. The delay starts inside the operation, not in transit.
Insight: At this rate, lateness is a property of the current process rather than an exception. Tuning individual ship modes will not move the on-time rate far.
Recommendation: Set an SLA on the processing stage first, track perfect order rate weekly, and only optimise ship mode once processing lead time is stable.

Late orders concentrate in a handful of states

Evidence: Pareto of late orders at a 70% threshold: California (2,001 orders, 1,123 on time), New York, Texas, Pennsylvania and Illinois lead. The West region alone carries 3,203 orders at a 47.4% perfect order rate.
Finding: A small group of states produces most of the late orders, while the long tail contributes little.
Insight: The delivery problem is geographic. Spreading remediation evenly across the country dilutes it where it would not change the outcome.
Recommendation: Audit warehouse capacity and carrier performance in the states covering 70% of late orders before extending the effort anywhere else.

The most returned categories are also the latest delivered

Evidence: Return rate by sub-category: Machines 11.3%, Tables 9.0%, Binders 8.8%, Paper 8.6% — against late delivery rates of 43–49% in the same categories. Tables carries a negative margin (−0.09); Bookcases and Supplies sit near zero.
Finding: The sub-categories with the highest return rates are also among the latest delivered, and several of them are already losing money at the margin.
Insight: This is a behavioural correlation, not a definitional one: late delivery raises the chance of a return, and returns eat straight into margins that are already thin.
Recommendation: Address Tables and Machines first — either tighten the delivery SLA for these categories or reprice them to absorb the cost of returns.

Revenue is falling while the customer base stays

Evidence: Revenue 2,297,201 (YoY ▼ 21.9%) · revenue per customer $2,897 (YoY ▼ 33.5%) · AOV $459 · repeat customer rate 98.5%. Consumer is 51.58% of customers, but Corporate carries the highest return rate at 8.5%.
Finding: Revenue per customer is falling faster than total revenue. Customers are not leaving; each one is simply spending less.
Insight: With a 98.5% repeat rate, the risk sits in the experience of existing customers rather than in a shortage of new ones. Poor delivery is eroding the value of each relationship.
Recommendation: Re-measure revenue per customer after the on-time rate improves, rather than spending the budget on acquisition.

Data limitations and their direction: the dataset covers orders from 2014–2017 and contains no warehouse operating costs, returns handling costs or inventory carrying costs. Profit and margin figures above are therefore overstated. The data stops on 30 December 2017, so the final period's MoM movements are truncated and should be read as a signal, not a conclusion.

Credential

Certificate of Completion

Master Analytical Thinking & Data Analysis with Power BI — The Future Analyst Academy, issued 15 August 2026.

Click the image to view full size · ID MAD14Q7E3 · 08/15/2026

In progress

Next capstones — expected October 2026

Three programmes running in parallel, each ending with its own capstone and certificate in mid-October 2026. This section will be updated as results come in.

Expected Oct 2026

AI Agent for Data Analytics & Decision Making

Building an AI agent that runs the full analysis loop — raw file, cleaning, descriptive and diagnostic analysis, report — to shorten the distance between data and decision instead of rebuilding it by hand each period.

AI Agent
Workflow Automation
Decision Support
Prompt Design

Expected Oct 2026

AI-Powered Excel for Data Analytics

Bringing AI into Excel to cut down data preparation and reporting time — the closest fit to daily store operations, where most numbers still live in a spreadsheet.

Excel
AI Formulas
Data Prep
Reporting

Expected Oct 2026

AI-Powered Google Sheet for Data Analytics

Analysis in Google Sheets with AI assistance, aimed at reports that can be shared with decision makers immediately and stay current without anyone installing software.