Across December 2009 to December 2011, with a monthly trend that peaks every November.
PROJECT 04 / Retail Analytics · Tableau Dashboard
Online Retail Sales and Returns Analysis
An interactive Tableau Public dashboard of a UK online retailer’s transactions from December 2009 to December 2011: £20.81M in sales across 41,943 orders and 5,881 customers, with views of the monthly trend, international markets, best sellers and returns.
DATASETOnline Retail II, UCI Machine Learning Repository
SIZE1,067,371 transaction lines · 1 December 2009 to 9 December 2011
MEASURED EVIDENCE
What can be verified.
Metrics and outputs drawn from the project artifacts—not estimates added for presentation.
Placed by 5,881 customers, for 11,672,569 units sold.
Total sales divided by orders, shown as one of six headline figures.
Shown on the executive dashboard, with the ten most returned products ranked on the second.
VALIDATION
The headline figures agree with each other: £20.81M in sales over 41,943 orders gives the £496.24 average order value shown on the dashboard.
PLAIN-LANGUAGE INTERPRETATION
Sales are in pounds sterling as recorded by the retailer. Returns are counted in units, so one large cancelled order can move a product to the top of the returns ranking.
Why this project matters
More than a million transaction lines don’t tell a manager much on their own. The useful questions are simpler: how sales move through the year, which markets and products matter most, and what customers send back.
What was developed
An executive dashboard with six headline figures (sales, orders, customers, average order value, units sold and return rate), a monthly sales trend, the top international markets and the ten best selling products, filtered by year and country. A second dashboard ranks the ten most returned products. Buttons move between the two.
What it means
Sales climb every autumn and peak in November in both 2010 and 2011. Outside the UK, Ireland (EIRE in the data) is the largest market, and the best selling product is a cakestand at about £345K in sales. The overall return rate is 4.13%, and two products, a paper craft item and a medium ceramic storage jar, account for 80,995 and 74,494 returned units.
04 / WHERE THIS WORK APPLIES
From project to practical use.
Monthly sales reviews, stock planning ahead of the autumn peak, deciding which international markets to prioritise, and following up on the products customers return most.
05 / LIMITATION
What the project does not solve yet.
The data stops on 9 December 2011, so the sharp drop in the final month is a partial month, not a collapse in sales. The two most returned products are far ahead of every other product, which looks more like a few very large cancelled orders than a pattern; the dashboard doesn’t separate those out yet. It covers one retailer, so the figures don’t describe UK retail in general.
06 / NEXT ITERATION
Future advancement
Add customer level analysis such as RFM segmentation and repeat purchase rates, and separate one off bulk cancellations from everyday returns.
TECHNICAL TOOLKIT
Tableau PublicCalculated KPIsInteractive FiltersDashboard NavigationTime Series AnalysisProduct RankingReturns AnalysisData Visualization