Aprenda Segmentation and Cohort Analysis Basics | Product Metrics and Data Exploration

Understanding how users interact with your product is crucial for making informed decisions. Two key techniques that help you gain these insights are user segmentation and cohort analysis. Segmentation means dividing your user base into groups based on shared characteristics or behaviors, such as how often they use your product or when they signed up. Cohort analysis, on the other hand, involves grouping users who share a common event within a specific timeframe—like signing up in the same month—and tracking their behavior over time.


              123456789101112131415161718192021222324
            
# Segment users into 'new', 'returning', and 'churned' based on activity data

users = [
    {"user_id": 1, "last_active_days_ago": 2, "signup_days_ago": 2},
    {"user_id": 2, "last_active_days_ago": 10, "signup_days_ago": 100},
    {"user_id": 3, "last_active_days_ago": 40, "signup_days_ago": 200},
    {"user_id": 4, "last_active_days_ago": 0, "signup_days_ago": 0},
    {"user_id": 5, "last_active_days_ago": 31, "signup_days_ago": 40},
]

segments = {"new": [], "returning": [], "churned": []}

for user in users:
    if user["signup_days_ago"] <= 7:
        segments["new"].append(user["user_id"])
    elif user["last_active_days_ago"] > 30:
        segments["churned"].append(user["user_id"])
    else:
        segments["returning"].append(user["user_id"])

print("New users:", segments["new"])
print("Returning users:", segments["returning"])
print("Churned users:", segments["churned"])

Segmenting users lets you tailor product features and marketing strategies to different groups. For instance, you might design onboarding flows specifically for new users, send re-engagement emails to those at risk of churning, or highlight advanced features to your most engaged users. This targeted approach increases the likelihood of meeting user needs and boosting engagement.


              12345678910111213141516
            
# Group users by signup month and count cohort sizes

import pandas as pd

data = {
    "user_id": [1, 2, 3, 4, 5, 6],
    "signup_date": [
        "2024-01-15", "2024-01-20", "2024-02-03", "2024-02-25", "2024-03-01", "2024-03-18"
    ]
}

df = pd.DataFrame(data)
df["signup_month"] = pd.to_datetime(df["signup_date"]).dt.to_period("M")

cohort_counts = df.groupby("signup_month")["user_id"].count()
print(cohort_counts)

Tudo estava claro?

Obrigado pelo seu feedback!

Seção 1. Capítulo 4

Pergunte à IA

Pergunte o que quiser ou experimente uma das perguntas sugeridas para iniciar nosso bate-papo

Deslize para mostrar o menu


              123456789101112131415161718192021222324
            
# Segment users into 'new', 'returning', and 'churned' based on activity data

users = [
    {"user_id": 1, "last_active_days_ago": 2, "signup_days_ago": 2},
    {"user_id": 2, "last_active_days_ago": 10, "signup_days_ago": 100},
    {"user_id": 3, "last_active_days_ago": 40, "signup_days_ago": 200},
    {"user_id": 4, "last_active_days_ago": 0, "signup_days_ago": 0},
    {"user_id": 5, "last_active_days_ago": 31, "signup_days_ago": 40},
]

segments = {"new": [], "returning": [], "churned": []}

for user in users:
    if user["signup_days_ago"] <= 7:
        segments["new"].append(user["user_id"])
    elif user["last_active_days_ago"] > 30:
        segments["churned"].append(user["user_id"])
    else:
        segments["returning"].append(user["user_id"])

print("New users:", segments["new"])
print("Returning users:", segments["returning"])
print("Churned users:", segments["churned"])


              12345678910111213141516
            
# Group users by signup month and count cohort sizes

import pandas as pd

data = {
    "user_id": [1, 2, 3, 4, 5, 6],
    "signup_date": [
        "2024-01-15", "2024-01-20", "2024-02-03", "2024-02-25", "2024-03-01", "2024-03-18"
    ]
}

df = pd.DataFrame(data)
df["signup_month"] = pd.to_datetime(df["signup_date"]).dt.to_period("M")

cohort_counts = df.groupby("signup_month")["user_id"].count()
print(cohort_counts)

Tudo estava claro?

Obrigado pelo seu feedback!

Seção 1. Capítulo 4

Segmentation and Cohort Analysis Basics

1. What is a user cohort in product analytics?

2. How can segmentation inform product feature development?

3. Which Python structure is suitable for grouping users by cohort?

Segmentation and Cohort Analysis Basics

1. What is a user cohort in product analytics?

2. How can segmentation inform product feature development?

3. Which Python structure is suitable for grouping users by cohort?