Data Quality

Data Cleansing

i/Lytics® Data Cleanse™ combines ISI’s advanced processing capabilities and unique multi-million word knowledgebases to provide a data management tool that quickly and accurately turns legacy account information from various sources into standardized, cleansed files that can be consolidated quickly and efficiently. It parses free-form data into distinct fields, compares those fields to words and phrases in ISI’s knowledgebases and correctly identifies them as male name, female name, surname, street name, organization, postal code, etc. It can also flag and correct variations, errors and misspellings, and can distinguish organizational records from individual records.

In addition to efficiently and accurately converting account information into usable customer records, Data Cleanse quickly and precisely processes free-form data to support real-time management of data entered at account opening, online or through a call center.

Individual Linking

Duplicate records within your database undermine the success of virtually every customer-related function your organization performs — from customer service to credit management to strategic planning. i/Lytics® Individual Link™ combines ISI’s advanced parsing capability with our unique ranking methodology to deliver unparalleled de-duplication accuracy when dealing with individual names.

Individual Link begins by parsing and standardizing name and address information and comparing data elements against ISI’s multi-million word knowledgebases to make sure they are properly identified and in the correct fields. Then, our unique ranking methodology compares two potential duplicates field-by-field to determine all the ways the records may be different and assigns the records a “rank” based on rules tuned to meet your organization’s business needs.

Because our methodology recognizes the specific ways two records vary, its results are much more accurate than those of aggregate comparison methodologies, e.g., weighted-field, scoring, and probabilistic systems, that simply add up match scores without considering the nature of the specific field differences between two records. It finds duplicates other methodologies miss and identifies which mis-matched duplicates should actually be unique customer records. Our assigned ranks also make it easy for clients to determine which ranks should be automatically combined, which identified as non-duplicates and which should be reviewed.

To further enhance your operational efficiency, not only does the accuracy of our ranking methodology reduce the number of potential matches requiring review, but our i/Lytics® Review Workbench™ system streamlines the review process, increasing reviewer productivity by 400% to 800% over other systems.

Corporate Linking

Identifying duplicate or linked corporate records is generally more difficult than individual records, because corporate records are frequently more complex, with multiple names, acronyms, and supplementary address information such as building or suite number.

i/Lytics® Corporate Link™ simplifies the comparison process by parsing and standardizing corporate records against our extensive knowledgebases of corporate name and address words to make sure data elements are properly identified and in the correct fields. Using ISI’s powerful, “intelligent” matching methodology, it then compares records on the basis of name and address; name only; address only; user-defined fields only; or any combination of those fields. It assigns a “rank” to each pair of potential duplicates based on the specific differences between the fields in each record. This “rank” enables clients to determine which potential match conditions represent “automatic matches,” which represent “non-matches” and which should be reviewed using the i/Lytics® Review Workbench™ system.

Eliminating duplicate records and identifying related corporate entities strengthens planning and decision making by providing a more realistic view of your customer base. It also helps you to determine more accurately the value of a customer’s overall relationship, informs better pricing and cross-selling strategies, and reduces the costs associated with redundant mailings and other communications.

Householding

i/Lytics® Household Link™ is a valuable customer-relationship identification tool that enables you to define “households” according to your own data and customer information needs. It groups customer records based on similarities in name, address, postal code and fields you define yourself, such as phone number or Social Security Number. It enables you not only to customize householding criteria, but also to determine the degree of variation to be permitted among household members. It assigns a “rank” to each pair of potential duplicates based on the specific differences between the fields in each record. This “rank” enables clients to determine which potential match conditions represent “automatic matches,” which represent “non-matches” and which should be reviewed using the i/Lytics® Review Workbench™ system.

To help you track household groups over time, Household Link’s Constant Key enables you to retain the same household grouping even if the group changes addresses or various members leave of join the group.

By identifying and retaining customer groupings, Household Link enhances online client service, strengthens target marketing and cross-sell initiatives, supports operational enhancements such as consolidated statements, and reduces the expense and waste related to duplicate mailings.

Online and Incremental Updates

i/Lytics® Change Management™ is an advanced data quality management tool that enables you to efficiently maintain and upgrade the integrity of your customer data when adding, changing, and deleting customer records either online or on an incremental batch basis. Driven by ISI’s powerful processing and proprietary knowledgebases, it enables users to interactively find customers, identify duplicates and households, enter updates and correct errors in real time.

i/Lytics Change Management’s sophisticated capabilities streamline and strengthen the data quality enhancement process. Its key creation capability provides users complete control over how the system identifies records to be compared. It can search based on name only, address only, name and address, birth date or any other user-specified fields. Its ability to match corporate as well as retail records enables it to meet database search and maintenance needs of companies with a mix of both customer types. In addition, when it identifies a set of duplicates, it can use rules-based “intelligent combine” and “best survivor logic” to automatically collapse the duplicates into a surviving record that retains the best information from both.

Integrated into any front-end or incremental batch back-end data entry application, Change Management ensures ongoing data quality and maintains a single customer profile each time a record is updated.

Back to top of page