Lighthouse

Level 6 · Customers who came back

From Jordan Reyes, Head of Marketing

“If we call a customer 'churned' after 180 days of silence, some of them must come back on their own. How many, and are their return orders worth anything? I need to know before I pay for a win-back campaign.”

Your goal

Use booked orders (status "completed" or "refunded"). A reactivation order is a booked order placed more than 180 days after the same customer's previous booked order (gap = (order_date - previous order_date).dt.days). Set result to a dict with "reactivation_orders" (int), "reactivated_customers" (customers with at least one reactivation order, int), "repeat_customers" (customers with 2 or more booked orders, int), "reactivated_pct" (reactivated × 100 / repeat customers, rounded to 1) and "net_sales": the net sales of the reactivation orders (Σ quantity × unit_price × (1 − discount_pct / 100) over lines with quantity > 0), rounded to 2.

Use the orders DataFramenot done yet
Use the order_items DataFramenot done yet
Store a dict in resultnot done yet

Your code

Python

Loading the editor…

Reference solution

ResultPreview of orders · press Run to see yours

Preview of orders
order_id	customer_id	order_date	status	channel
1001	125	2023-01-01 20:51:19	completed	web
1002	156	2023-01-02 07:58:46	completed	web
1003	22	2023-01-02 10:25:27	completed	marketplace
1004	48	2023-01-02 11:21:51	completed	marketplace

Mote checks your result

Press Run. Each row that matches the goal becomes a firefly in your lantern.

DataFrames

Loaded for you as pandas DataFrames, plus pd and np.

orders12,000 rows · 7 columns

order_idinteger
customer_idinteger
order_datetimestamp
statusvarchar
channelvarchar
shipping_feedecimal
coupon_codevarchar

First 5 rows of orders
order_id	customer_id	order_date	status	channel	shipping_fee	coupon_code
1001	125	2023-01-01 20:51:19	completed	web	0.00	NaN
1002	156	2023-01-02 07:58:46	completed	web	0.00	NaN
1003	22	2023-01-02 10:25:27	completed	marketplace	4.99	WELCOME10
1004	48	2023-01-02 11:21:51	completed	marketplace	4.99	WELCOME10
1005	82	2023-01-02 11:43:36	completed	web	0.00	WELCOME10

order_items30,338 rows · 6 columns

order_item_idinteger
order_idinteger
product_idinteger
quantityinteger
unit_pricedecimal
discount_pctdecimal

First 5 rows of order_items
order_item_id	order_id	product_id	quantity	unit_price
1	1001	121	9	31.99
2	1001	154	3	11.99
3	1002	154	4	11.99
4	1002	194	1	149.00
5	1002	193	2	19.99

customers2,018 rows · 9 columns

customer_idinteger
first_namevarchar
last_namevarchar
emailvarchar
cityvarchar
statevarchar
signup_datedate
segmentvarchar
referral_sourcevarchar

First 5 rows of customers
customer_id	first_name	last_name	email	city	state	signup_date	segment	referral_source
1	Owen	Hudson	owen.hudson@icloud.com	San Francisco	CA	2022-06-02	consumer	search
2	Douglas	Mendoza	dmendoza69@gmail.com	Atlanta	GA	2022-06-02	consumer	NaN
3	Henry	Brooks	hbrooks@lakesidecafe.com	Los Angeles	CA	2022-06-04	business	NaN
4	Brian	Graham	bgraham@bluecedarhomes.com	Seattle	WA	2022-06-06	business	pinterest
5	Kyle	Jimenez	kjimenez27@yahoo.com	Indianapolis	IN	2022-06-07	consumer	NaN

products120 rows · 8 columns

product_idinteger
namevarchar
categoryvarchar
subcategoryvarchar
unit_pricedecimal
unit_costdecimal
launched_ondate
is_activeboolean

First 5 rows of products
product_id	name	category	subcategory	unit_price	unit_cost	launched_on	is_active
101	Vale Velvet Loveseat	Furniture	Sofas	1659.00	NaN	2023-09-11	true
102	Alder Oak Coffee Table	Furniture	Tables	919.00	503.50	2024-05-16	true
103	Orchard Rattan Dining Chair	Furniture	Chairs	789.00	429.87	2023-05-27	true
104	Upland Oak Bookcase	Furniture	Storage	429.00	214.70	2022-12-20	true
105	Avery Leather Bed Frame	Furniture	Beds	449.00	232.70	2021-08-31	true

returns1,673 rows · 5 columns

return_idinteger
order_item_idinteger
returned_ondate
reasonvarchar
refund_amountdecimal

First 5 rows of returns
return_id	order_item_id	returned_on	reason	refund_amount
1	14	2023-01-10	damaged	83.69
2	42	2023-01-10	wrong_item	19.99
3	105	2023-01-14	wrong_item	10.99
4	24	2023-01-16	changed_mind	26.99
5	126	2023-01-16	late	20.99

employees60 rows · 8 columns

employee_idinteger
first_namevarchar
last_namevarchar
departmentvarchar
titlevarchar
manager_idinteger
hire_datedate
salaryinteger

First 5 rows of employees
employee_id	first_name	last_name	department	title	manager_id	hire_date	salary
1	Catherine	Hale	Executive	Chief Executive Officer	NaN	2016-03-01	285000
2	Robert	Kim	Executive	Chief Operating Officer	1	2017-01-09	235000
3	Megan	Collins	Sales	Head of Sales	1	2018-04-16	172000
4	Chloe	Gutierrez	Sales	Account Executive	3	2020-01-21	72000
5	Frank	Bell	Sales	Account Executive	3	2023-05-21	73000

web_sessions40,000 rows · 7 columns

session_idinteger
customer_idinteger
started_attimestamp
devicevarchar
sourcevarchar
pages_viewedinteger
convertedboolean

First 5 rows of web_sessions
session_id	customer_id	started_at	device	source	pages_viewed	converted
1	NaN	2023-01-01 23:02:24	mobile	paid	1	false
2	18	2023-01-01 15:07:49	desktop	paid	1	false
3	NaN	2023-01-01 16:55:46	desktop	direct	7	false
4	NaN	2023-01-01 12:43:28	mobile	organic	3	false
5	NaN	2023-01-01 13:06:21	mobile	social	2	false

marketing_spend3,276 rows · 3 columns

spend_datedate
channelvarchar
amountdecimal

First 5 rows of marketing_spend
spend_date	channel	amount
2023-01-01	paid	492.99
2023-01-01	email	37.38
2023-01-01	social	200.53
2023-01-02	paid	709.13
2023-01-02	email	64.13

The lesson

Churn is not the end

Level 4 called a customer churned after 180 days without a booked order. That is a useful label for today’s list, but it is a snapshot. Look back over three years and many customers who crossed that line came back on their own. A reactivation is an order that ends such a long silence.

The previous order, row by row

Sort the booked orders by customer and date, and shift() within each customer moves every date down one row, so each order sees the customer’s previous order:

booked["prev"] = booked.groupby("customer_id")["order_date"].shift()
booked["gap"] = (booked["order_date"] - booked["prev"]).dt.days

The first order of each customer has no previous one, so its gap is NaN, and NaN > 180 is False: first orders are never reactivations, which is correct. The sort matters: shift() takes the previous row, not the previous date, so unsorted data gives nonsense gaps.

Three numbers, three denominators

Orders and customers answer different questions. One customer can reactivate twice, so there are more reactivation orders than reactivated customers. For the share, use the customers who could reactivate: you need at least two orders to have a gap, so the base is customers with two or more booked orders, not every customer.

What the return orders are worth

Use the same net sales as the rest of the track: valid lines only (quantity > 0), price paid after the coupon discount. Look the reactivation orders up in a per-order total:

order_net.reindex(react["order_id"]).sum()

Compare the result with the store as a whole. If reactivation orders are smaller than a typical order, a win-back discount eats a bigger share of each one.

Why this matters before a campaign

A win-back email sent to everyone who has been silent for 180 days will be followed by many returns. Some of them would have happened anyway, and the email will get the credit. The number you compute here is the baseline the campaign has to beat. The clean way to measure the campaign is to hold back a random group of churned customers and compare.

Common mistakes

Forgetting to sort before shift().
Counting cancelled orders as the “previous” order. A failed checkout does not end a silence.
Using every customer as the base. One-time buyers cannot reactivate.

Level 6 · Customers who came back

Your goal

Your code

Reference solution

Mote checks your result

What Jordan gets, and why

DataFrames

The lesson

Churn is not the end

The previous order, row by row

Three numbers, three denominators

What the return orders are worth

Why this matters before a campaign

Common mistakes

Lantern lit!

Level 6 · Customers who came back

The request

Your goal

Your code

Reference solution

Mote checks your result

What Jordan gets, and why

DataFrames

The lesson

Churn is not the end

The previous order, row by row

Three numbers, three denominators

What the return orders are worth

Why this matters before a campaign

Common mistakes

Cookie choices