Sales Analytics

Resolving customer identities with Numbers Station

Case Study
Industry
B2C Luxury
Data Stack
Salesforce
Salesforce
Snowflake
Snowflake
Tableau
Tableau
Numbers Station Transforms
SQL Transforms & Record matching

A B2C luxury goods company faced a challenge with silos in their customer data systems. The lack of integration between their different systems caused a disconnect in the customer journey, resulting in lost deals and frustrated customers. They turned to Numbers Station to resolve the identities of their customers and create a 360 customer dashboard. Using Numbers Station, the company saw an increase in customer satisfaction and an improvement in conversion rates, leading to increased revenue and growth opportunities.

Problem

A B2C luxury goods company faced a challenge with silos in their customer data. The core issue was that the company had two customer relationship management (CRM) systems (one for their physical stores and one for and online stores) with no way to connect the two sources of data. The lack of integration between the two systems caused a disconnect in the customer journey, resulting in lost deals and frustrated customers (e.g. marketing representatives reaching out to existing physical store customers). The company decided to invest in creating a 360 customer dashboard that would provide sales and marketing teams a holistic and centralized source of truth for their customer data.

Goal

Build a single source of truth for customer data by merging customers from different systems.

Challenge

The data coming from all the different physical stores was extremely messy. It consisted of json files with string fields, missing ids, misformatted dates and addresses, etc. The team of data analysts faced a major technical challenge to clean this data and join it with their online store CRM (Salesforce). In particular, there was no single identifier to join the customers on, and they had to rely on names and other messy fields to join customers. They tried implementing a set of rules in SQL which did not yield satisfactory matching quality due to noise in the data. Ultimately, they quickly realized that they needed more robust matching models and decided to try Numbers Station.

Solution

Using Numbers Station’s record matching product, the team of data analysts was able to build automated pipelines to map all physical store customers into their existing salesforce CRM. The Numbers Station pipeline was powered by cutting edge foundation models that significantly improved accuracy compared to their existing systems. The pipeline was deployed into the company’s Snowflake data warehouse and used daily to update customers’ 360 views and dashboards.

Results

The Numbers Station solution enabled data analysts to develop intelligent customer matching models, with significantly higher accuracy than their previous best rule-based model. This ultimately allowed the team to get a holistic view of their leads and customers and improve customer acquisition.

98%

Matching accuracy between physical and online stores’ customers

10x

Faster matching pipeline buildout compared to rule based solution

More sales analytics use cases

Analyze sales call transcripts

Extract common patterns found in sales call transcripts like mentions to competitors, pricing and pain points. Understand factors that lead to deal conversions and improve sales strategies.

Clean Salesforce data

Use Numbers Station to deduplicate customers and impute missing entries in Salesforce data. Build a holistic and accurate view of your sales pipeline to sell with confidence.

Analyze historical sales data

Create new views from historical sales data to understand what are high and low performing channels, forecast future sales trends and allocate resources more effectively.

Ready to automate data wrangling in your organization?

Start Your Free Trial