Nearly every company faces the challenge of accurately and quickly sorting and distributing the inbound US mail within its organization. And we have all seen the tubs and trays of inbound mail stacked and awaiting sortation and delivery. How do you determine if automation is an appropriate consideration for your organization? Read on!


The First Steps

The first steps may vary slightly in sequence by organization, but in general, they go as follows:


Document and Measure Your Current Process The old saying that you can't manage what you can't measure is still true. What does the current process involve in terms of head count and hours of labor? What actually gets accomplished? Is the mail rough sorted then fine sorted? How much additional sorting takes place? A time study is an important component to determine how many pieces per hour an individual is actually throwing. You may be surprised by these results, as often managers have relied upon the estimates used by the USPS for manual sortation such as 1,200 pieces per hour. An actual time study may reveal that your manual sortation is down around 300 or 400 pieces per hour! Why is that? Manually sorting US mail primarily calls for the recognition of a ZIP Code. If you have ever sorted inbound or interoffice mail, you know that usually much more information goes into the sorting decision. For instance, mail stop code, department name, person's name, building, room number, floor number... you get the picture.


So in a nut shell, sorting inbound mail is a complex decision process for a person or for an Optical Character Recognition (OCR) engine. By documenting and measuring the current process, you have assembled the building blocks of a cost benefit analysis. After all, if you automate a process, you really want to know the degree you have improved it including time, cost and accuracy of the sort. Manual sorting has a fairly high error rate compared to an automated solution. Once you have your current costs calculated, they could easily be loaded into a cost benefit analysis that will estimate the return on investment. The end goals should include faster and more accurate distribution of your company's document communications. Long term, they should also include a reduction of overall costs. In general, automation is best suited for medium- to high-volume operations.


Space Automation equipment requires space normally not much more than the previous cases but is not small. Consideration must also be given to the flow of mail once a machine process is put in place. The power and air supply will need to be planned along with consideration of noise levels. Over time, this equipment becomes much more quiet, although it still represents a machine process and should not be placed amongst office workers.


Budgeting Most of us will need to plan our capital expenditures a year in advance. This will require identification of the various vendor solutions and their associated costs. This budget request is much more credible having performed the cost benefit analysis. ·


Staffing Anyone considering automation equipment should also consider the staffing required to operate it. This equipment is now capable of providing detailed management reports of volumes and trend analysis that were not previously available. These reports can be extremely useful in identifying cost center concentrations for such purposes as budget allocation. Information will be available from an automated solution so you need to be prepared to generate and collect and use this data. Turnover is a challenge in most mailing operations, so I recommend careful selection of operators and trainees. Potential machine operators and database maintenance staff should be identified before any automation effort.


Address Format and Database In utilizing automation for inbound mail sortation, it is important to have some regularly occurring or predictable information for which the OCR to look. In general, a person's name is the least reliable way to sort mail through automation. Proper names tend to be close or the same in many cases. They are also often similar to building as well as street names. Department names are next because single departments often reside across several locations or actual mail stops. Mail stop codes are a reliable feature for OCR recognition because they are usually accompanied by the word "mail stop" or the letters "ms." Key words that appear on a predictable basis followed by some meaningful numbers are also acceptable and produce reliable recognition. Examples of these would be "Department #," "Agency #" or "Account #." Some organizations have unique ZIP Codes or a batch of unique ZIP+4 Codes assigned to them. If these ZIP+4 Codes are then exclusive to a mail stop, then they are an excellent way to sort the mail. PO Box numbers are also often used as a reliable sort feature and often correspond.


Technology today is capable of using all of these address features collectively or some of them selectively in order to produce a high confidence recognition match. The database utilized for matching is an important consideration, as you will need to plan a continual update to this when changes occur within your organization. There are a number of sources to include Human Resources, the IT or IS departments, Payroll or even the telephone directory. The important point to remember is that recognition rate is directly related to the accuracy and quality of the matching database.


Sort Schemes The current operations have some existing sort scheme by which the mail is separated. These are obviously limited by the requirement to sort the mail manually. In many cases, multiple sorts are required in order to get the mail down to a delivery point in a manual operation. Automation equipment has the advantage of on-board computer technology for the generation of sort schemes. Through the use of this technology, consideration can be given to sorting into the actual walk or delivery sequence. Mail can be sequenced by mail stop in a first pass and then sorted into delivery sequence in a second pass. In many cases, there is some sorting down stream by clerical staff that can also be eliminated through the use of sort scheme automation.


Internal and External Communication Efforts I cannot overemphasize the need for communications if you are considering an automation effort. The internal users or customers will need to be informed of your intentions and the addressing formats required for participating in the program or receiving the most benefit. They will then need to communicate with their correspondents the desired address components or formats. Careful consideration must be given to inventories and future print runs of letterhead and envelopes if some changes are being made for conformity purposes. I strongly encourage a newsletter to your internal customers as an ongoing reminder of these automation efforts.


Document and Measure the Results The benefits of having your organization receive their internal or inbound document communications faster are difficult to quantify, although significant. An example that comes to mind would be the opportunity to detect and out-sort mail containing checks. My guess is that the accounts receivable department would like to get these deposited as quickly as possible. Technology exists today that can detect and sort selectively on a number of criteria to include MICR detection, metal detection, length, height and thickness measurement. Some inbound mail streams are easily identified by some or all of these criteria. Reductions in labor hours and the elimination of tasks are easily measured. Downstream functions such as opening the envelopes may also be incorporated into an automation solution thereby eliminating that separate task. The important consideration is to be deliberate in documenting the changes produced by automation. Saved labor and time can evaporate unless consideration is given to it.


Historically, OCR equipment was originally designed for US mail automation programs and relied very heavily upon the uniform addressing requirements of the USPS. This equipment performs at very high speeds with high accuracy because of the consistency of addressing and the accuracy of the matching database. More recently very specialized OCR software has been developed for inbound and interoffice applications. These automation solutions are being utilized by large complex organizations to improve the process of inbound and interoffice mail sorting.


Terry Waldon is the inbound Sorting Product Manager for Bowe, Bell & Howell. For additional information or questions, please feel free to contact Terry via e-mail at