Many research firms have made their 2010 IT predictions, which range from smarter customer service to real-time analysis. A common thread in many of these predictions is the use of advanced analytics in everyday decisions.
When considering the rise of analytics in IT developments, it is easy to get caught up in the potential value and neglect the methods behind the numbers. Analytics can provide a multitude of benefits to companies, but only if the information being analyzed is accurate. Existing or purchased consumer data is frequently the basis for this analysis, and this data can be wrought with errors. The accuracy of the underlying customer data is critical in order for advanced analytics systems to provide any business value.
Before moving forward with new analytics efforts, it is important for organizations to pause to make sure that existing systems produce accurate data that can support the desired advancements.
Why Does Data Matter?
When it comes to advanced analytics, data quality is a prerequisite to success. Duplicate accounts, misspelled names, incorrect geographical information or inaccurate purchase history can throw off results. These errors drain businesses resources and efficiency by promoting ineffective marketing, unnecessary customer service data entry and incorrect customer order fulfillment.
These types of data quality issues require consistent rework but can also dramatically impact the return on investment for an often expensive analytics technology. More often than not, incorrect customer data results in bad analysis that cannot be used. Even worse, bad data may lead to poor business decisions - the exact opposite goal for an analytics project.
Who Struggles with Data Quality?
With a renewed IT emphasis on analytics, one may be surprised to hear that only half of businesses have an adequate level of confidence in the quality of their data, according to a 2009 Experian QAS survey. This lack of trust is primarily attributed to an internal lack of manual resources, an inadequate data strategy and failings of existing technology to maintain data quality.
This statistic shows that businesses are not alone in their struggle to achieve accurate data. Even the U.S. government has its problems with data confidence. In a recent article, Jeff Kelly describes how poor data management played a key role in allowing the December 2009 attempted terrorist attack. A simple misspelling of Umar Farouk Abdulmutallab’s name in the visa database allowed him to move freely through airport security. Had safeguards been put in place to ensure the accuracy of data, Abdulmutallab may have been subject to additional screening or prevented from boarding the flight to Detroit.
How Do You Improve Business Data Quality?
Businesses should improve data confidence in order to ensure the success of infrastructure changes. A first step toward achieving this goal is to remove duplicate accounts. Customer history is often spread between multiple accounts in one or more databases. This frequently occurs when external lists are imported for marketing purposes, formatting techniques are inconsistent, or names are misspelled.
One simple, low investment way to eliminate duplicates is to review customer contact data and correct common errors. Address, email and phone data are unique identifiers for most customer files and can therefore be leveraged to identify duplicates. Accurate address data in particular can help organizations combine household records to enhance the overall customer value.
Errors can also be quickly discovered by a database administrator. It is easy to spot patterns of errors in a small database sample, such as certain contact fields are formatted incorrectly or left empty altogether.
Once businesses identify and eliminate inaccuracies, they need to identify the source to prevent future errors from entering the database. Depending on the type of business, these files could come in through purchased lists or staff or customer data entry. If most of the data entry errors originate from the same entry point, it is easy for businesses to put safeguards in place prior to entry.
Lists can be scrubbed on the back-end before entry to the database, and front-end verification techniques can be integrated into call centers and customer-facing websites. Once contact data is accurately and properly formatted, database administrators can use the unique contact identifiers described above to remove duplicate accounts. These scrubs can be performed manually or with advanced phonetic deduplication software. The size of the database often dictates which matching technique should be used.
By ensuring that data is properly formatted and duplicates are removed, businesses can guarantee that each customer only has one entry in a central database. This singular, clean customer view allows for accurate advanced analysis.
Move Forward with Confidence
If research predictions hold true, companies will be investing heavily in advanced analytics for real-time decision-making and other business functions. Many of these new ideas require additional IT resources, which are already strapped thin for most businesses.
Stakeholders need to ensure the accuracy of data before committing to extensive new projects that require a great deal of personnel and budget. Without accurate data to power these systems, most businesses will spin their wheels and never achieve the desired result.
By simply cleaning contact data and removing duplicates, businesses can be confident that the additional investments and increased efficiencies promised by advanced analytics technologies can be achieved without additional headaches.
Register or login for access to this item and much more
All Information Management content is archived after seven days.
Community members receive:
- All recent and archived articles
- Conference offers and updates
- A full menu of enewsletter options
- Web seminars, white papers, ebooks
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access