Although the automated reading of addresses has achieved a very high level of accuracy in recent years, there is still a portion of addresses on mailpieces that is not sufficiently recognized and needs to be entered manually. However, today's advancements in recognition technology allow posts to successfully overcome even these residual bottlenecks in mail processing.

Foreign Mail
The main driver for investing in technology innovations is to meet customers' expectations: to timely deliver mail to the exact recipient and add value to the mailpiece by making it relevant to the addressee. One of the problems here is connected to processing foreign addresses. All mailpieces have to be sorted to discriminate between domestic and foreign mail. Foreign mailpieces are processed separately and are sent abroad. If a domestic address is interpreted as a foreign address, the mailpiece is sent abroad and will either be lost or, at best, reach the addressee with a significant delay. Today, recognition systems meet the strictest error rate requirements to differentiate between domestic and foreign mail.

Incomplete Addresses
Another example of an issue that can be solved using new technology is the processing of incomplete addresses, which is part of a bigger, undeliverable-as-addressed (UAA) problem. The accuracy of address information is often neglected, and mailpieces are sent with an incomplete or inaccurate address. These mailpieces require additional handling: if an address cannot be automatically resolved, the mailpiece is also sent for manual processing just to verify the conclusion that the address is unreadable and that it cannot be processed further. The latest Optical Character Recognition technologies (OCR) include an arsenal of sophisticated techniques enabling a conclusive judgment on the quality of address information (for example, technology can determine that an address is incomplete or inaccurate), thus eliminating the need to send a mailpiece for manual keying.

Simultaneously, modern systems can reduce the portion of mailpieces sent for manual processing, including those with addresses missing ZIP Codes. Until recently, OCR systems required the ZIP Code to recognize numbers first, and thus reduce the context to recognize the text part of an address. Today there are systems that can successfully and reliably read the text part of an address using a broad context, thus automatically resolving addresses that do not contain a ZIP Code.

Mail Redirection
Mail redirection is one of the most challenging postal situations and exploits one of the most sophisticated automated systems for mail processing. State-of-the-art recognition technology performs a variety of tasks on images of US letter mail that go beyond mere address reading. In addition to the destination address, this technology reads and matches personal/business names contained in the destination address block with names contained in the USPS Change of Address (COA) database. The capability to automatically read the addressee name eliminates the long and costly mail forwarding procedures, improves the speed and accuracy of mail forwarding, and substantially reduces expenditures on postal logistics. Moreover, advanced OCR also reads the return address, executes Address Change Service (ACS) determination and endorsement determination. Businesses commonly use the name and address data for many different purposes (e.g. customer relationship management, address validation, marketing, data quality, data validation, customer profiling, personal information, etc.).

While extracting information from the endorsement and ACS line, the technology enables further services for customers, allowing the utilization of an electronic address change service. The ACS service helps meet the needs of business mailers by providing a cost-effective, efficient means of obtaining accurate change-of-address (COA) information. This saves mailers time and money by reducing the amount of undeliverable-as-addressed (UAA) mailpieces and reducing labor needed to manually re-key addresses. Thus, the updated address data made available by OCR can be passed on to business customers for a fee, helping banks, insurance companies and other businesses ensure that their customer address databases are always up-to-date.

Other achievements that are worth mentioning include the capability to successfully detect overlapping mailpieces (so-called "double feed") even where the shift is very small, to ensure that the covered mailpieces are not lost or sent to a wrong destination. Technology can also successfully detect and recognize "Return to sender" stamps, which may be randomly located on mailpieces and are often of a very poor quality.

These new technology applications represent working systems with already proven efficiency and cost savings for the posts that have implemented them. Many deserve a closer look, as they have significant potential to further improve mail sorting efficiency and enable additional postal services. Not only can they read more addresses with higher accuracy, but they are also able to read address elements and data from a mailpiece that could not be reliably read before.

Dr. Tatiana Vazioulina has worked in product management and marketing at Parascript, LLC., an image analysis and pattern recognition company, for a decade. She can be reached a or 303-381-3106. Parascript can be found online at