The Half-Life of Data

Author | Big Data, Data Cleansing, Data Quality, Global Data Quality | , , , , ,

Data is the new oil of the digital economy. Effectively used, data can help organizations to better understand customer needs and provide winning strategies to meet them. But the data you depend on to make these important decisions is always in flux.

In The Half-Life of Data, Bud Walker, Vice President of Enterprise Sales and Strategy for Melissa, breaks down the true nature of data and what it means for your business. From scientific fact to census info, to contact data and beyond, data will expire – and as it turns out, you can easily calculate when that will be.

For the full article, check out pages 19-20 in the latest addition of NoCOUG Journal. If you’re interested in joining NoCOUG Journal, click here.

Catch Bad Data Before it Wrecks Your Business

Blog Administrator | Data Cleansing | , , ,

Did you know that bad data can wreck your business? That’s
right, not updating, verifying and maintaining data is having a bigger impact
on your business than you realize. Bad data is costly and it provides inaccurate information. This ends
up affecting communication and sales strategies that, in the long run, affect and may potentially wreck your business.

Don’t let bad data wreck your business. Here are five things
you can do to stop it:

1. Data Cleansing

This step is a combination of the definition
of business rules and software designed to execute these rules. As a business,
you must identify the way data cleansing tools define rules to determine the
best option for particular data sets.

2. Address Data Quality

A key aspect of managing quality comes from
reviewing the business processes to see where location data is created,
modified and read–all with the intent of improving the quality of the address.

3. Address Standardization

Delivery accuracy saves money while
eliminating the rework and extra costs of failed delivery. This is why it is
crucial to standardize addresses. The best way to deal with this problem is to
treat each non-standard address as an exception and forcing the delivery agent
to deal with it. Another approach is to fix the problem earlier on by using
data tools to transform non-standard addresses into one that conforms with the
standard.

4. Data Enhancement

There are numerous ways data sets can be
enhanced. This includes adapting values to meet defined standards, applying
data correction and adding additional attributes.

5. Record Linkage and Matching

Do you know how many data sets contain
information about a specific individual? There are a lot of distributed resources
of information about customers and they each have a valuable piece of
information. But when the pieces of data are merged together they become an
incredibly insightful profile of the customer.

 The quality of your data can either make or
break your business. Don’t let bad data wreck your business. Call Melissa Data today to find out how to implement our solutions and get rid of bad data.

By Natalia Crawford

 

Melissa Data Completes the Data Quality Circle with Addition of Profiler Tool for SQL Server Integration Services

Blog Administrator | Address Quality, Address Validation, Address Verification, Analyzing Data, Analyzing Data Quality, Data Management, Data Profiling, Data Quality, Data Quality Components for SSIS, Data Warehouse, MDM, Press Release, SQL Server Integration Services | , , , , , ,

Flagship SSIS Developer Suite Now Enables Data Assessment and Continuous Monitoring Over Time; Webinar Adds Detail for SSIS Experts


Rancho Santa Margarita, CALIF – March 17, 2015 – Melissa Data, a leading provider of contact data quality and address management solutions, today announced its new Profiler tool added to the company’s flagship developer suite, Data Quality Components for SQL Server Integration Services (SSIS). Profiler completes the data quality circle by enabling users to analyze data records before they enter the data warehouse and continuously monitor level of data quality over time. Developers and database administrators (DBAs) benefit by identifying data quality issues for immediate attention, and by monitoring ongoing conformance to established data governance and business rules.

Register here to attend a Live Product Demo on Wednesday, March 18 from 11:00 am to 11:30 am PDT. This session will explore the ways you can use Profiler to identify problems in your data.

“Profiler is a smart, sharp tool that readily integrates into established business processes to improve overall and ongoing data quality. Users can discover database weaknesses such as duplicates or badly fielded data – and manage these issues before records enter the master data system,” said Bud Walker, director of data quality solutions, Melissa Data. “Profiler also enforces established data governance and business rules on incoming records at point-of-entry, essential for systems that support multiple methods of access. Continuous data monitoring means the process comes full circle, and data standardization is maintained even after records are merged into the data warehouse.”

Profiler leverages sophisticated parsing technology to identify, extract, and understand data, and offers users three levels of data analysis. General formatting determines if data such as names, emails and postal codes are input as expected; content analysis applies reference data to determine consistency of expected content and field analysis determines the presence of duplicates.

Profiler brings data quality analysis to data contained in individual columns and incorporates every available general profiling count on the market today; sophisticated matching capabilities output both fuzzy and exact match counts. Regular expressions (regexes) and error thresholds can be customized for full-fledged monitoring. In addition to being available as a tool within Melissa Data’s Data Quality Components for SSIS, Profiler is also available as an API that can be integrated into custom applications or OEM solutions.

Request a free trial of Data Quality Components for SSIS or the Profiler API.
Call 1-800-MELISSA (635-4772) for more information.

News Release Library





Data Profiling: Pushing Metadata Boundaries

Blog Administrator | Analyzing Data, Analyzing Data Quality, Data Management, Data Profiling, Data Quality, Golden Record | , , , , ,

By
Joseph Vertido
Data Quality Analyst/MVP Channel Manager

Two truths about data: Data is always changing. Data will always have problems. The two truths become one reality–bad data. Elusive by nature, bad data manifests itself in ways we wouldn’t consider and conceals itself where we least expect it. Compromised data integrity can be saved with a comprehensive understanding of the structure and contents of data. Enter Data Profiling.

Throw off the mantle of complacency and take an aggressive approach to data quality, leaving no opening for data contamination. How? Profiling.

More truths: Profiling is knowledge. Knowledge is understanding. That understanding extends to discovering what the problems are and what needs to be done to fix it.

Armed with Metadata

Metadata is data about your data. The analysis of gathered metadata with Profiling exposes all the possible issues to its structure and contents, giving you the information–knowledge and understanding–needed to implement Data Quality Regimens.

Here are only a few of the main types of Generic Profiling Metadata and the purpose of each:

  • Column StructureMaximum/Minimum Lengths and Inferred Data Type – These types of metadata provides information on proper table formatting for a target database. It is considered problematic, for example, when an incoming table has values which exceed the maximum allowed length.
  • Missing InformationNULLs and Blanks – Missing data can be synonymous to bad data. This applies for example where an Address Line is Blank or Null, which in most cases is considered a required element.
  • DuplicationUnique and Distinct Counts – This allows for the indication of duplicate records. De-duplication is a standard practice in Data Quality and is commonly considered problematic. Ideally, there should only be a single golden record representation for each entity in the data.

Other equally important types of Generic Profiling Metadata include Statistics for trends data; Patterns (ReqEx) allow for identifying deviations from formatting rules; Ranges (Date, Time, String and Numbers); Spaces (Leading/Training Spaces and Max Spaces between Words); Casing and Character Sets (Upper/Lower Casing and Foreign, Alpha Numeric, Non UTF-8) Frequencies for an overview of the distribution of records for report generation on demographics and more.


Metadata Revolution & New Face of Profiling

Right now the most powerful profiling tool for gathering Metadata is the Melissa Data Profiler Component for SSIS, which is used at the Data Flow level, allowing you to profile any data type that SSIS can connect with, unlike the stock Microsoft Profiling Component, which is only for SQL Server databases.

More importantly the Melissa Data Profiler offers over 100 types of Metadata including all the Generic Profiling Metadata mentioned here.

The innovative Melissa Data’s Profiler Component gathers Data Driven Metadata, which goes beyond the standard set of profiling categories. By combining our extensive knowledge on Contact Data, this allows us to get information not simply based on rules, norms, and proper formatting. Rather, it provides metadata with the aid of a back-end knowledge base. We can gather unique types of metadata such as postal code, State and Postal Code Mismatch, Invalid Country, Email Metadata, Phone and Names.


Take Control

The secret to possessing good data goes back to a simple truth: understanding and knowledge of your data through profiling. The release of Melissa Data’s Profiler for SSIS allows you to take control of your data through use of knowledge base driven metadata. The truth shall set you free!

For more information on our profiling solutions, please visit our website