Datasets / Generated

Sales orders

Two years of orders with regions, channels, discounts and refunds.

Ask a question of it Download CSV Jupyter notebook

11,316 rows10 columns Generated

Worked examples

Each of these is a real run: the pipeline picked the columns, shaped the rows and chose the form. Open the trace to see what it decided and why.

Net revenue by region, monthly

Three regions moving in different directions: the crossover is the answer, and it is only visible once the daily orders are rolled up to months.

how did net revenue trend by region over time? — 120 rows charted from 11,316, drawn as a line

Ask this yourself See the working

How it decided

Finding the data54 ms

Sales orders: 11,316 rows, 10 columns. order_date, region and net_amount answer this; the other six columns do not enter the chart.

Shaping the data13 ms

11,316 rows in, 120 out across 3 columns. Three regions moving in different directions: the crossover is the answer, and it is only visible once the daily orders are rolled up to months.

Choosing the chart14 ms

line. Time on x and one line per region. Grouped bars would put twelve months side by side and hide the crossing.

Applying defaults1 ms

Palette, spacing, axis titles and legend placement applied. 0 issues found.

Average net revenue per order, by channel

A per-order average, not a total: the channel with the most revenue and the channel with the most valuable order are usually not the same channel.

which channel brings the most revenue per order? — 6 rows charted from 11,316, drawn as a bar

Ask this yourself See the working

How it decided

Finding the data49 ms

Sales orders: 11,316 rows, 10 columns. channel and net_amount; order counts come from the grouping itself.

Shaping the data5 ms

11,316 rows in, 6 out across 2 columns. A per-order average, not a total: the channel with the most revenue and the channel with the most valuable order are usually not the same channel.

Choosing the chart7 ms

bar. A handful of named categories being compared on one measure is a bar chart, sorted so the ranking is the shape.

Applying defaults1 ms

Palette, spacing, axis titles and legend placement applied. 0 issues found.

Refund rate by product category

A rate, computed from the flag rather than counted twice: the mean of a boolean is the share that are true.

what share of orders get refunded, by category? — 5 rows charted from 11,316, drawn as a hbar

Ask this yourself See the working

How it decided

Finding the data54 ms

Sales orders: 11,316 rows, 10 columns. category and the refunded flag. Amounts are irrelevant to a rate.

Shaping the data10 ms

11,316 rows in, 5 out across 2 columns. A rate, computed from the flag rather than counted twice: the mean of a boolean is the share that are true.

Choosing the chart5 ms

hbar. Category names are long, so the bars run horizontally and the labels stay readable rather than rotating.

Applying defaults1 ms

Palette, spacing, axis titles and legend placement applied. 0 issues found.

Schema

10 columns, typed as the pipeline sees them — which is how it knows what can go on a time axis and what can be summed.

Column	Type	Example
order_id	number	100001
order_date	date	2024-01-01
region	text	APAC
channel	text	Paid search
category	text	Hardware
units	number	8
amount	number	544.41
discount	number	27.22
refunded	boolean	False
net_amount	number	517.19

First 8 rows

order_id	order_date	region	channel	category	units	amount	discount	refunded	net_amount
100001	2024-01-01	APAC	Paid search	Hardware	8	544.41	27.22	False	517.19
100002	2024-01-01	APAC	Referral	Services	3	2165.91	0.0	False	2165.91
100003	2024-01-01	North America	Partner	Training	6	1601.87	0.0	False	1601.87
100004	2024-01-01	North America	Organic	Support	2	154.86	0.0	False	154.86
100005	2024-01-01	LatAm	Email	Software	7	297.94	59.59	False	238.35
100006	2024-01-01	North America	Social	Support	5	99.78	9.98	False	89.8
100007	2024-01-01	North America	Paid search	Software	3	216.58	21.66	False	194.92
100008	2024-01-01	APAC	Referral	Training	6	335.2	0.0	False	335.2

Ask it something else

how did net revenue trend by region over time?which channel brings the most revenue per order?what share of orders get refunded, by category?show monthly revenue with discounts broken out

Start free