mta-credit-card-journeys.py

# Step 1: Convert all interactions to a list
journeys = campaign_data.groupby('customer_id')['channel'].aggregate(
    lambda x: x.tolist()).reset_index()

# Step 2: Add last interaction as 1 or 0 event representing activation
activation_results = campaign_data.drop_duplicates('customer_id', keep='last')[['customer_id', 'activation']]
journeys = pd.merge(journeys, activation_results, how='left', on='customer_id')

# Step 3: Add start and end states based on whether customer activated
journeys['path'] = np.where(
    journeys['activation'] == 0,
    journeys['channel'].apply(lambda x: ["Start"] + x + ["Null"]),
    journeys['channel'].apply(lambda x: ["Start"] + x + ["Activation"])
)

journeys = journeys[['customer_id', 'path']]

# Get overall activation rate
total_activations = journeys['path'].apply(lambda x: x[-1]).str.match('Activation').sum()
activation_rate = total_activations / journeys.shape[0]

can you open access to data as mentioned in
Campaign Logs — the goldmine of information
I have created a dataset that resembles what you would typically see in campaign logs.

https://pub.towardsai.net/discrete-time-markov-chains-identifying-winning-customer-journeys-in-a-cashback-campaign-39b62eb8a6fe