Applying AARRR in Practice

Dave McClure's AARRR framework (Acquisition, Activation, Retention, Revenue, Referral) forces you to examine the entire customer journey.

AARRR stage aggregation in Python

import pandas as pd

# Channel-level AARRR pipeline dataset
data = {
    "channel": ["Paid Social", "Organic Search", "Referral"],
    "acquisitions": [10000, 8000, 3000],
    "activations": [1800, 3200, 1650],      # D1 value moment
    "d30_retained": [270, 1120, 825],       # D30 retention
    "paid_converted": [120, 480, 450],      # Revenue stage
}
df_aarrr = pd.DataFrame(data)

# Calculate stage conversion rates
df_aarrr["activation_rate"] = df_aarrr["activations"] / df_aarrr["acquisitions"]
df_aarrr["retention_rate"] = df_aarrr["d30_retained"] / df_aarrr["activations"]
df_aarrr["pay_cvr_overall"] = df_aarrr["paid_converted"] / df_aarrr["acquisitions"]

print(df_aarrr[["channel", "activation_rate", "retention_rate", "pay_cvr_overall"]].to_string(index=False, formatters={
    "activation_rate": "{:.1%}".format,
    "retention_rate": "{:.1%}".format,
    "pay_cvr_overall": "{:.1%}".format,
}))

Takeaway

AARRR helps spot whether your acquisition engine is merely filling a leaky activation bucket.