Data Science

Entytle Process – Part 1

Data Cleanup

We are often asked about the processing steps that Entytle customers’ ERP/CRM data undergoes, from the point at which it enters our system to the point at which Sales Opportunities are presented back to the customers. In this post we describe the first part of the process, during which the data is “cleaned up” and prepared for analysis. The analysis steps will be described in a later post.

For an in-depth explanation of why data cleansing matters, and how we do it, check out the first episode of the Entytle Podcast.

Step 1 Audit Customer Data

Entytle accepts customer data from different ERP/CRM systems for analysis. Customer data can be of different types. For analysis we principally use Purchase History and this is supplemented by Leads, Opportunities and Services. We do a preliminary audit to measure the quality of the data. The audit also helps us identify the Entytle modules which can be used by the Customer.

Step 2 Map Customer Data

Each customer has a unique data model for different types of data. Entytle maps the elements of the customer data to Entytle’s data model. Entytle requires at least 8 to 10 years worth of data. During the course of time a customer might have accumulated and stored data in different systems. For example, customers may have purchase history data from 1995 to 2000 in Excel and data from 2001 to 2017 in SAP. Entytle accepts data from both the sources and maps it to our standard data model.

Also Check Out:  The IoT for unconnected things

Step 3 Ingest Customer Data

Entytle provides a secure channel through which customers can share confidential data. Historical data is loaded during the first ingestion and then Entytle loads data updates from different data sources on a weekly basis.

Step 4 Standardize Customer Data

Similar data received in various formats from different data sources is transformed to a common format. During standardization, extraneous characters are removed from the data without loss of information. For example all dates are converted to a single format and all US state names are converted to their two letter abbreviation.

Step 5 Enrich Customer Data

Once the data is standardized, Entytle can append missing or incomplete data. Data can be enriched with proprietary, public and private third party (e.g. USPS ZipCode finder) data sources. Data enrichment includes, but is not limited to things like; purchasing customer’s address, name, and product details. For example, if we have customer name and street address but the ZIP code is missing, then Entytle populates the missing ZIP code. Contact details for the purchasing customer can be enriched and extracted using different customer data like Services or Activities history.

Don't miss any post

Subscribe now to be notified of new content