The Art and Science of Acquiring and Managing Data

By Mario Nowogrodzki, CPA, Mendelson Consulting – March 1, 2017
The Art and Science of Acquiring and Managing Data

Analytics is the application of computer technology, operational research and statistics to solve problems and answer questions in business and industry. A key component of that formula is data. 

The acquisition and management of data — for integration into accounting systems and analysis — is both an art and a science. And the purpose of that formula is to make informed decisions in business and indus­try (and life!). 

The Art: Integration with Accounting Systems 

Every organization and every situation inside an organization has unique infor­mational needs. And the acquisition and management of data can be a challenge — and sometimes an endless mission. We know that well-organized, meaningful data is important for any organization’s account­ing and business management system, and it is key for the survival of operations in an ever-evolving environment. 

The process of acquiring data begins with the who, how and why of business workflows and responsibilities. This is a must in order to understand what kind of data you need. Properly structured, organized data directly impacts the way an organization functions — meaning that data needs to be managed properly. 

One simple and common example of in­tegration is the import of expense transac­tions into the accounting system by an end user via download and automatic coding and posting of online banking transac­tions. This not only automates data entry for increased efficiency, but also improves accuracy by eliminating (or reducing) the human factor. Another application of integration is synchronization of online transactions into the order processing and accounting systems. This provides for the scalability of processing a high volume of orders without the limiting factor of human data entry. And a byproduct of this integration is the automated update of stock levels in the inventory system. 

Often, data has existing problems, but these issues are not found until integra­tion. You should often check data for integrity, completeness and accuracy — all controls we learn as CPAs. Also, computer systems are sensitive when it comes to data. Extra characters, spaces, leading zeros, or worse, duplication of records, can wreak havoc on integration. Often, data normalization is part of any successful integration project. Nor­malization is the process of organizing attributes and relations of data to reduce redundancy and improve data integrity. So any data management and integration project must begin by making data “good” — good enough so that the computer system behaves as designed. Most people understand the old adage of “garbage in, garbage out” — meaning that incorrect or poor-quality input will always result in faulty output. And such incorrect or incomplete information often results in failures in human decision-making. 

The Science: Data Mining 

Preparation of data prior to analysis is a key factor in any successful data mining effort. Data mining is the practice of examining what could be large amounts of data in order to generate meaningful information. 

Generally, data mining involves analyzing data from different perspectives and summa­rizing it into useful information — informa­tion that can be interpreted, by example, to help increase revenues, cut costs, and make other financial and management decisions. On a deeper, more-advanced level, data min­ing can also involve sorting through data to identify patterns and establish relationships. 

It is often very desirable to have a system and method for acquiring and managing data that provides real-time access to a centralized database including previously acquired as well as incoming data. However, typically the most interesting data originates and resides inside the computer applications that support the core business processes of an organiza­tion, such as the accounting program itself or an inventory management system. This data takes a variety of forms, including transaction records, master lists, and application and database logs. Because of their critical role in the business, these systems are often off-lim­its to computer processing and space-hungry data-mining activities. 

The best option is to transfer the data to a computing environment that is more data-mining friendly and less taxing on the systems that house and run an entity’s operations. This means a separate, high-ly-optimized database that can serve the purpose of simply just serving the data. Because it is a separate database from the core business operations, a benefit is that you can have additional data pieces in the data extraction process from the oper­ating system, including additional fields calculated from existing data for use in further analytics. 

Very often, the majority of efforts in data mining projects are spent in the acquisi­tion, preparation and management of data, and comparatively little time is spent on analysis. That is why much of the research on data mining focuses on providing increasingly sophisticated analysis and discovery algorithms. 

The Story: Analytics for Informed Decisions 

The purpose of acquiring and managing data is to make informed decisions in business and life. This can range from integrating external data into an accounting and busi­ness management system to the eventual production of reports for such decision making. But reports alone do not make the cut. It is the data gathering and interpreting that makes the story valid and meaningful. Data is the lyrics that make up the narrative that tells a story. And the story is what we use to make informed decisions. 


Mario Nowogrodzki

Mario Nowogrodzki, CPA, CITP, is founder and principal of Mendelson Consulting, an accounting technology firm that assists entities with planning, selecting and implementing business management systems.

This article appeared in the March/April 2017 issue of New Jersey CPA magazine. Read the full issue.