Search⌘ K
AI Features
Home
Courses
An Introduction to Entity Resolution in Python

Sell with Insights

Explore how to leverage resolved customer data to enhance business decisions by identifying upsell and cross-sell opportunities. Understand how to link resolved IDs with transactional data, analyze correlations between project size and service fees, and use customer-product relationships to target marketing campaigns. This lesson helps you apply entity resolution results to real-world sales insights and business growth strategies.

We'll cover the following...

This lesson is about how to put resolved data into action to improve decision-making. The restaurants data contains customer records with hidden duplicates. The cross_ref table encodes our resolution. It is time to use the resolved data and address the needs of the business.

Our cross_ref table covers only those original customer IDs with at least one match. We parse the values to strings and expand that table to cover all restaurants for joining with transactional data.

Python 3.8
restaurants = pd.read_csv('solvers_kitchen/restaurants.csv')
cross_ref['resolved_customer_id'] = 'cluster_' + cross_ref['resolved_customer_id'].astype(str)
cross_ref = restaurants[['customer_id']].merge(cross_ref, on='customer_id', how='left')
cross_ref.loc[cross_ref['resolved_customer_id'].isnull(), 'resolved_customer_id'] = cross_ref['customer_id']
print(cross_ref.sample(5, random_state=1))

Now, we add the resolved ID to every table having the original customer ID.

Python 3.8
restaurants = restaurants.merge(cross_ref, on='customer_id', how='left')
projects = projects.merge(cross_ref, on='customer_id', how='left')
contracts = contracts.merge(cross_ref, on='customer_id', how='left')
order_headers = order_headers.merge(cross_ref, on='customer_id', how='left')
print(restaurants.sample(2, random_state=1).T)

We are ready to utilize the resolved data to support the business with insights.

Upsell opportunities

...