Brazilian modernist illustration of a Sao Paulo neighborhood, shops and parcel deliveries

SAMBA

The business.
The rhythm.
The next move.

THE BRIEF

A clearer view
of a growing business.

Samba wants to grow sales in 2022. The first step: give leadership a monthly view of orders, customers, sellers and spending.

Two dashboards bring the business together. The findings point toward a focused approach: understand the rebound, prioritize established demand, and look beyond attention-grabbing averages.

Historical case study · January 2021–January 2022

Delivered orders
49,544
Unique customers
48,022
Unique sellers
1,771
Average order value
$157.70

01 / THE BIG PICTURE

One business.
Two views.

Looker Studio connects the headline metrics with cities and categories. Tableau adds a view of customers, sellers, item quantity and monthly payments.

Samba Looker Studio executive dashboard showing full-period KPIs, city orders, category orders and monthly average order value
Looker Studio · Dashboard snapshot, January 2021–January 2022. Workbook-based calculations and metric definitions follow below.

02 / THE REBOUND

More orders.
Almost the same basket value.

January 2022 brought 6,900 orders, up 28.2% from December. Customers rose 27.9% and sellers 12.5%. Average order value barely moved: $152.22 to $152.23.

The monthly picture

Jan 2021–Jan 2022
Payments = total recorded order payment. Average order value (AOV) = payment total ÷ orders. All monetary values retain the source's $ notation; the workbook does not specify a currency code.
Monthly data
MonthOrdersCustomersSellersPayments ($)AOV ($)

03 / WHERE DEMAND LIVES

São Paulo
sets the pace.

The city leads with 7,020 orders. Across the wider state, 19,048 unique customers make São Paulo the largest customer base in the dataset.

The leading markets

Top 10 states
Whole-period rankings. Cities are grouped by city name and state. Customer counts are distinct within each state and should not be added across states.

04 / WHAT PEOPLE BUY

Big demand.
Small-sample surprises.

Bed, Table & Bath leads with 5,052 orders, followed by Sports & Leisure and Health & Beauty. Together, these three categories account for 26.2% of orders.

Order volume meets item quantity

Each point is a category. Orders use a logarithmic scale. Item quantity is the mean of the workbook's qty_item field, not the number of distinct products in a basket.
3.5

The highest quantity average.
Only two orders.

Hygiene Diapers records seven items across two orders. Its 3.5-item average leads the ranking, but the sample is too small to treat as a dependable merchandising opportunity.

05 / THE NEXT MOVE

Start where demand
is already visible.

01

Prioritize São Paulo

Use its established customer base as the starting point for a focused sales campaign.

02

Test the leading categories

Bed, Table & Bath, Sports & Leisure, and Health & Beauty are candidates. Check demand within the target market before choosing an offer.

03

Measure the result

Track orders and AOV together. Test campaigns against a comparison group before attributing growth to promotion.

The dashboards identify opportunities.
The next step is to test them.

These are proposed actions, not measured campaign effects or a sales forecast.

Behind the numbers

Analysis by Rafiq Naufal Kastara
Data preparation & metric definitions

Orders, Order History and Payments each contain 49,544 unique order IDs. The article joins them one-to-one with no unmatched orders. All records are marked Delivered and fall between January 2021 and January 2022.

Orders are distinct order IDs. Customers and sellers are distinct IDs within the selected period or group. Payment totals sum total_payment_value once per order. AOV divides that total by order count. Monthly unique counts are not additive across time.

Item quantity averages qty_item, which the data dictionary describes as quantity for the related product. It does not measure distinct items. Category names are lightly edited for readability; underlying groups are unchanged. Malformed geolocation fields are not used.

Monetary values use the dashboards' $ notation without currency conversion. The workbook does not identify an ISO currency code. Payments are not a measure of profit.

Dashboard snapshots & calculation differences

The Looker Studio workflow blends the tables and tidies category labels. The Tableau workflow uses inner joins, including the Advanced table. The images preserve those dashboard snapshots; the article's charts are recalculated from the saved workbook.

Some snapshot labels are imprecise: the monthly “average orders” line is average order value, the state view counts unique customers, and “unique items” is item quantity. The article names these measures directly.

Some lower-ranking values also differ. For example, the Looker image shows 2,388 Telephony orders; the workbook contains 2,308. House Comfort 2 has 21 items across 17 orders, an average of 1.24. The workbook, not a visually estimated bar length, supplies the interactive values.

The dashboard's AOV ranking groups by city name alone. This article separates same-named cities by state. That puts Trindade, PE first at $3,184.55, also from just one order. Monte Alegre do Sul remains the leader when grouping by city name alone.

The workbook is a saved historical export, not a live feed. The analysis describes patterns, not their causes. Campaign exposure, costs and experimental outcomes are not available.

Looker Studio dashboard

100%
Enlarged Looker Studio dashboard