Churn prediction and prevention in Digital Marketing - Time & Space Complexity

Time Complexity: Churn prediction and prevention

O(n)

Understanding Time Complexity

When predicting and preventing customer churn, we want to know how the time to analyze data grows as we get more customers.

We ask: How does the work needed change when the number of customers increases?

Scenario Under Consideration

Analyze the time complexity of the following churn prediction process.

for each customer in customer_list:
    features = extract_features(customer)
    prediction = model.predict(features)
    if prediction == 'likely_to_churn':
        send_prevention_offer(customer)
    

This code checks each customer, predicts if they might leave, and sends an offer to keep them if needed.

Identify Repeating Operations

Look for repeated steps that take most time.

• Primary operation: Looping through each customer to predict churn.
• How many times: Once for every customer in the list.

How Execution Grows With Input

As the number of customers grows, the work grows in a similar way.

Input Size (n)	Approx. Operations
10	About 10 predictions and checks
100	About 100 predictions and checks
1000	About 1000 predictions and checks

Pattern observation: The work increases directly with the number of customers.

Final Time Complexity

Time Complexity: O(n)

This means if you double the customers, the time to predict churn roughly doubles too.

Common Mistake

[X] Wrong: "Predicting churn for many customers takes the same time no matter how many customers there are."

[OK] Correct: Each customer needs a separate prediction, so more customers mean more work.

Interview Connect

Understanding how prediction time grows helps you explain how your solution handles more customers smoothly and efficiently.

Self-Check

"What if the model prediction step itself took longer as more customers are processed? How would that affect the overall time complexity?"