Clustering for user segmentation

You

Marketing wants 'customer segments' for CatChow. I was just going to guess three: cheapskates, regulars, whales.

Alex Tutor

You can guess, or you can let the data tell you. That family of methods is called cluster analysis — finding groups of similar customers without predefined labels.

You

Isn't that the classification thing from logistic regression?

Alex Tutor

Close but backwards. Classification sorts objects into known classes you already named. Clustering finds the groups themselves — you don't even know how many exist yet.

Two situations come up. Either you know how many groups to expect, or you have no idea. The method you pick depends on which one you're in.

You

I have no idea how many segments CatChow customers actually split into.

Alex Tutor

Question 1

You have no idea how many customer segments CatChow actually has. Which family of methods fits this situation?

Choose one answerHierarchical clusteringk-means with k = 3Logistic regressionPaired t-test

Alex Tutor

Every clustering method starts the same way: turn 'how similar are these two customers' into a number — a distance.

You

Distance like kilometres between their houses?

Alex Tutor

Distance in the space of their metrics. Say customer A spends 100 UAH/month with 1 order, B spends 120 UAH with 1 order. They're close in that space.

Alex Tutor

d_{\text{Euclidean}} = \sqrt{\sum_i (x_i - y_i)^2}

Euclidean distance: the straight-line distance across all variables.

That's the Euclidean distance — the ordinary straight-line one. There's also Manhattan distance, which sums absolute differences instead of squaring them.

You

Named after the borough? Why would statisticians care about Manhattan traffic?

Alex Tutor

Because you can't cut diagonally through city blocks — you walk along the grid. It's less thrown off by one huge outlier variable than Euclidean distance is.

Alex Tutor

Question 2

Customer A spends 100 UAH/month with 1 order; customer B spends 120 UAH with 1 order. What is the Euclidean distance between A and B on (spend, orders)?

Your answer

Units: distance units (e.g. 20)

Alex Tutor

When you don't know the number of groups, use hierarchical clustering. It merges the two most similar customers, recomputes distances, then merges again.

You

It keeps merging... forever?

Alex Tutor

Until every customer is in one giant cluster. Along the way it draws a dendrogram — a tree you can cut at any height to get however many groups you want.

You

So the tree decides the number of clusters for me, after the fact?

Alex Tutor

You decide, by choosing where to cut — usually where branches are tall, meaning the merged groups were quite different from each other.

Alex Tutor

Question 3

What does hierarchical clustering produce that lets you choose the number of groups after the fact?

Choose one answerA dendrogramA scree plotA confusion matrixA correlation matrix

Alex Tutor

Now say you do know the number: leadership wants exactly 2 spend-based segments. Then the right tool is k-means.

Take four customers by (monthly spend, orders/month): A(100, 1), B(120, 1), C(500, 5), D(520, 5). k-means places 2 centroids — group representatives — and moves them until they settle.

Alex Tutor

Four customers, two spend clustersA and B (small spenders) sit far from C and D (big spenders); k-means groups them exactly along that gap.

Chart data

Monthly spend, UAH	Small spenders (A, B)	Big spenders (C, D)
100	1
120	1
500		5
520		5

You

So the centroid of {C, D} is just their average spend and average order count?

Alex Tutor

Exactly: (500 + 520) / 2 = 510 UAH, and (5 + 5) / 2 = 5 orders. That point becomes the new centroid for that cluster.

You

And every customer just checks which centroid is closer and jumps there if needed?

Alex Tutor

Right — assign to nearest centroid, recompute centroids as the mean of their members, repeat, until nobody moves anymore.

Alex Tutor

Question 4

Customers C(500, 5) and D(520, 5) are grouped by k-means. What is the centroid's average monthly spend for this cluster?

Your answer

Units: UAH (e.g. 510)

You

So after the centroids move, do we just... stop?

Alex Tutor

Not yet. With new centroid positions, some customers may now be closer to a different one, so we reassign and recompute again.

You

So it loops until nothing changes anymore?

Alex Tutor

Exactly — that's convergence. Try ordering the algorithm's steps yourself before we move on.

Alex Tutor

Question 5

Put the steps of the k-means algorithm in the right order.

Tap an item, then tap where it should go. You can also drag or use the arrows.

Alex Tutor

One trap: spend is measured in hundreds of hryvnias, orders in single digits. Distances would be dominated entirely by spend.

Alex Tutor

You

So orders/month would basically be ignored otherwise?

Alex Tutor

Almost. And one more thing: k-means depends on where centroids start. Run it several times — if you get different groups each time, your chosen k may be wrong.

You

How do I even pick k in the first place?

Alex Tutor

Business judgement first — how many segments can marketing actually act on? Then check how much clustering error drops as k grows, and stop once it flattens out.

Alex Tutor

Question 6

Because k-means uses distances, a variable measured in hundreds of hryvnias can dominate the clustering unless you standardise all variables first.

TrueFalse

Alex Tutor

So: hierarchical when you don't know k, k-means when you do, and always standardise your variables first.

You

So my 'cheapskates, regulars, whales' guess might not survive contact with real distances?

Alex Tutor

It might match perfectly, or the data might split on something you never guessed — frequency instead of spend, say. Let's write that up.

You

Before I run anything — remind me what to check first?

Alex Tutor

Question 7

Which of these are good practices before running k-means on CatChow customer data? Select all that apply.

Choose all that applyStandardise variables so spend doesn't drown out orders/monthPick k as large as possible so every customer gets their own personal clusterRun k-means multiple times with different starting centroids and compare resultsChoose k using business judgement or by checking where clustering error stops dropping much

Before we get to a messy real scenario, put the units problem in your own words first.

You

Uh oh, the dreaded free-response box.

Alex Tutor

You'll survive. Here's the setup.

You

Fire away.

Alex Tutor

Question 8Short answer · AI-checked task

In 2–3 sentences, explain why the same four CatChow customers could end up in different clusters depending on whether spend is measured in UAH or in thousands of UAH.

Write your answer and get a score with feedback from our AI reviewer.

Log in to get AI feedback

Good instinct. Now a messier, real situation — someone already ran a clustering and the results look off.

You

Let me guess: too many segments?

Alex Tutor

You've clearly been paying attention.

You

Show me the mess.

Alex Tutor

Question 9Case · AI-checked task

Write your recommendation for the growth marketer below.

A growth marketer ran k-means on CatChow's customer data with k = 6, reasoning 'more segments means more personalisation.' The resulting segments are tiny, several contain fewer than 10 customers, and re-running the algorithm produces noticeably different groupings each time. She wants your sign-off before building 6 separate email campaigns.

Write your answer and get a score with feedback from our AI reviewer.

Log in to get AI feedback

That's clustering. Next: factor analysis, for when it's not customers you want to group, but survey questions.

Alex Tutor

That’s the lesson. You answered every task — nicely done.

Next lesson: Factor analysis: surveys and NPS drivers