Publishing company solved its issues with internet presentation
ADQC cleansed and consolidated customer data from primary system in order to use them for the internet presentation of the printed directories.
Summary
ADQC cleansed and consolidated customer data from primary system in order to use them for the internet presentation of the printed directories.
Customer Profile
The publishing company was found in 1991. It is part of a significant publishing group, which associates directory publishers from nine countries. Since 1994 the company operates in Slovakia as well. The company publishes around 60 directories (4, 5 millions of copies) per year, including telephone directories, professional directories and local directories. Beside this it operates Czech and Slovak internet pages with actual database of 2, 3 millions of companies from Czech Republic and 100 000 companies from Slovakia. Since 1993 it provides direct marketing services for third parties (telemarketing, direct mail, database administration).
Business Benefits
The aim of the project was to cleanse and consolidate customer data from primary system in order to use them for the internet presentation of the printed directories. The solution was based on Ataccama Data Quality Center and covered client data cleansing (companies and addresses), consolidation and propagation of the cleansed data back to the primary system. The data contained both Czech and Slovak companies and addresses. The data was cleansed, checked against Czech and Slovak referential tables and unified into groups representing a single record based on defined rules. The cleansed data was propagated back to the primary system. After the initial cleansing, periodical data cleansing is performed. The consolidated client information is used for the following purposes:
- Internet presentation
- Consolidated data helps to eliminate numerous duplicities in the internet presentation of the printed directories
- Address identification (over 80% of addresses were identified) helps to assign geo-coordinates to each address and localize it precisely on the map
- The automated cleansing and consolidation reduces amount of the data for manual cleansing and deduplication
- The use of cleansed data raises customer satisfaction with the internet presentation and made it more usable for the end users
- Primary system
- The cleansed data that is propagated back to the primary system solves data non-quality directly at the source
- The precisely identified address serves for including the customers into relevant region according to their seat
- The consolidation of the client data done in the primary system helps to eliminate duplicities in the primary data
- Campaign preparation
- The consolidated data is used for periodical sales campaign preparation and prevents a situation when a client is contacted twice because of duplicate data or is part of a wrong regional campaign due to a bad address
- Client database
- The solution can be used as a base of the client database in the future, as other data sources can be plugged-in
Solution
The solution for client data cleansing and consolidation (including the propagation back to the primary system) was implemented within three months. The data was first taken out of the primary system into a separate database and cleansed and consolidated. The rules for client consolidation were set up based on user requirements and with respect to the local conditions and use of the data.
A representative record is chosen for each consolidated group of clients. The representative record is chosen from the records in the consolidated group based on the complex set of rules defined by the client. This record is used for the internet presentation.
Other part of the solution was focused on propagation of the data back to the primary system and unification of the data in the primary system. This process is strictly based on client requirements because not all the data elements can be returned back in order not to harm the data in the primary system.
The data that couldn’t be returned back is reported. Around 10 different reports are frequently generated and can serve as a basis for manual data cleansing.
See What's New
