Background

The TexShare Coordinating Council (formerly the Advisory Board) requested the ILL Working Group to analyze the benefits of using Ariel in TexShare. In the spring of 1998 Nancy Paine, Inter-Library Service Librarian, University of Texas at Austin, and Keiko Horton, Coordinator of Interlibrary Loan Services, University of Houston Libraries, surveyed TexShare libraries about their use of Ariel. At the Second Annual Interlibrary Loan Workshop, held June 5, 1998 in Dallas, Texas, Keiko Horton presented a paper entitled "Ariel Use in TexShare: Is It Cost Effective?" This paper analyzed the results of the Ariel survey and addressed the issue of cost effectiveness.

The goal of this report is to compare delivery options and provide recommendations for document delivery among TexShare libraries.

Summary of Ariel Survey

There were 74 usable survey responses, indicating that 47.3% of the respondents were currently using Ariel in their ILL operations. These were predominantly (84%) state colleges and universities. Just over half of the Ariel users photocopied documents before scanning. Most (59%) of the Ariel workstations were stand-alones. More than three-quarters (78%) were using outmoded, "slow" scanners.

For most Ariel users, the volume of documents transmitted via Ariel was low--66 percent transmit fewer than 100 documents per month. Similarly the volume of documents received via Ariel is low--66 percent receive fewer than 100 documents per month.

Most often a combination of both classified and student staff were responsible for the Ariel operation, although in all but one library all levels of staff were involved in Ariel in varying degrees. Student assistants and classified staff were most likely to scan documents; paraprofessionals were most likely to train other staff and maintain the software.

Two-thirds of the respondents used Ariel as the primary method of document delivery. Most Ariel users evaluated it favorably. The most common disadvantage recognized was the slow response rate to the need for resends, and the most common reason for resends was missing pages.

The main reasons libraries gave for not using Ariel were:

• Low volume of ILL
• High cost
• Not enough staff
• Difficulty in using Ariel (e.g., slow speed, low-quality copies, insufficient technical support)

New members often asked for the same kind of financial assistance given to the original 52 TexShare members. The high initial start up cost was the biggest concern.

Cost Effectiveness of Ariel

To address the cost effectiveness of Ariel, Nancy Paine and Keiko Horton prepared a cost-comparison table (see attached). The table compared the costs of supplying a 10-page document using U.S. mail, TExpress courier, fax, and Ariel. Adjustments were made for volume of activity.

The calculations were based on the following assumptions (made in May 1998):

• Hourly staff rate of $6.50
• Computer workstation $2458
• Cost of Ariel software $895
• Cost of HP 4Plus scanner $953
• Cost of Fujitsu 3096G scanner $4350
• Fax machine $2375
• Copier equipment cost not included; photocopy cost estimated at $.06 per page
• Three-year life span for equipment

The study revealed that Ariel was more cost effective for high-volume operations than for low-volume operations. No matter which method of document transmission was used, with the exception of U.S. mail, the average total cost per 10-page document decreased as volume increased. The total cost per document using U.S. mail stays about the same at $1.80 no matter what the volume of copies mailed. For low-volume operations, dividing the startup costs for Ariel by the number of documents transmitted results in a very high per document cost.

As was stated earlier, more than half the libraries responding to the survey indicated they photocopied documents before scanning on Ariel. This was not surprising since so many of the surveyed libraries were using extremely slow scanners which require more staff time and are tedious to use. However, the cost comparison shows that it actually costs more to photocopy first than to scan directly from the item even with a slow scanner. Using the slow scanner without copying first takes more staff time even though it is less costly. Traditionally ILL operations do not have to count all the pennies they spend, but they are frequently short-staffed. So the expense of copying first yields a better workflow for the ILL unit, even though the per unit cost is higher for the library. These problems are eliminated with a fast scanner, although the equipment cost may be too high for all but high-volume operations.

It should be noted that slow scanners that work with Ariel (e.g., HP Scan Jet Plus, 3c, 4p, 5p, IIc, IIex, IIp, IIIp) are no longer made. Libraries contemplating using Ariel should check the RLG Web Site for a list of RLG-supported scanners (http://www.rlg.org/ariscan.html). Libraries that have been happiest with Ariel have purchased RLG-recommended equipment, including a computer with more than the recommended memory and a fast scanner.

Recommendations to the TexShare Coordinating Council

Several advantages of using Ariel for document delivery are obvious--speed of transmission, clarity of the document received, ease of transmission as compared with fax. It is the preferred method of document delivery for an increasing number of libraries worldwide as well as for commercial document suppliers, such as CISTI and the National Library of Medicine. For high-volume operations there is also a cost advantage over most other delivery methods.

Within TexShare, when delivery cost is the primary consideration, it is necessary to factor in TExpress, TexShare's Courier Service. Like Ariel, TExpress is most cost effective for high-volume operations. Each participant pays a fixed annual charge for 2-3 day delivery of any number of packages at any weight to other TExpress libraries. The cost per package decreases as the volume increases. For example, at $2100 per year, a library that sends 1000 packages per year pays $2.10 each; a library that sends 5000 per year, pays $.42 each.

The startup cost for TExpress is considerably less than that for Ariel, although TExpress startup costs must be paid annually while Ariel equipment is a longer term purchase. Ariel has regular maintenance and upgrade costs, and requires ethernet connectivity as well as supplies (e.g., paper and toner). TExpress transports books as well as photocopies, broadening its usefulness. Low-volume operations may not need to invest in both Ariel and TExpress.

The Ariel survey showed that the majority of TexShare Ariel sites transmit and receive fewer than 100 documents per month. Low-volume institutions considering using Ariel may not find Ariel cost effective, primarily because of the high startup costs. Financial assistance from some source, such as TexShare or a TIF grant, would make startup costs easier. In the past, TexShare has provided training and technical support (initial and ongoing), although some on-site technical support is probably necessary. Another option for low-volume libraries is a "receive only" Ariel workstation.

When considering Ariel and TExpress, per unit cost is not the only relevant factor. Rapid delivery of needed materials is high priority as is cost effectiveness for the total consortium. Athough operations that borrow more than they lend may find per unit cost of TExpress and/or Ariel expensive, it is important to remember that high volume lenders supplying their needs will find TExpress cost effective. What seems expensive for a low-volume lender on a per unit basis may be very cost effective for TexShare as a whole.

There is a wide range of participating libraries in TexShare ILL, and the best option for some may not be the best option for others. Traditionally 10 percent of the participants in TexShare ILL supply 50 percent of the loans/copies, and the biggest lenders are usually the biggest borrowers. For these libraries, it is obvious that using both Ariel and TExpress is cost effective. Low-volume libraries may find neither Ariel nor TExpress cost effective.

Summary

"TexShare libraries contemplating the purchase of Ariel software should base their decision on their broader ILL needs and patterns, not just on their anticipated use of Ariel for TexShare," advises TexShare management (http://www.texshare.edu/About/faq.html#ariel) . When making the choice, a library should weigh the following factors:

*Importance of delivery speed
*Importance of clarity of documents received and sent
*Total ILL volume
*Availability of funds
*Percentage of ILL activity in and outside of TexShare (Ariel is used nationally and internationally by libraries and document delivery vendors; TExpress is not)
*Multipurpose use of Ariel equipment
*Integration of delivery methods within existing ILL workflow
*Possibility of delivering documents electronically directly to users

Submitted by the ILL Working Group

Lisa Anderson
Scott Downing
Brad Eden
Karen Hendricks
Keiko Horton
Rebecca Linton
Sara Lowman
Jeanette Mosey
Nancy Paine, Chair
Heather Perkins
Tim Prather
Carol Roberts