Articles
Software Utilities
How to Choose Help Desk Software By Dr. Frances Tischler, P.H.D
Originally printed in July, 1995 in AFSM International
Dr.Frances Tischler is Executive Vice President and Chief Operating Officer of Professional Help Desk. Before applying her skills to PHD, Dr.Tischler worked as Marketing Director at Applied Cognetics and served as a member of the faculty at the City University of New York, where she taught Chemistry and Mathematics. Dr.Tischler has also been widely published in the fields of Chemistry, Physics, Computer Science and Marketing.
More than ever before companies are searching for cost-effective ways to increase employee productivity. Introducing the PC to the organizational mix has been a mixed blessing. On the one hand, PCs enable the knowledge worker to manipulate and analyze data in ways that, just a few short years ago, were impossible. On the other hand, the PC has introduced an extraordinary degree of technological complexity into places where none existed before. Today, for the organization to get the biggest bang for the PC buck, it is not only important that the PC function flawlessly, but the end-user understand how to best utilize the software that helps him manipulate his information.
A Nolan Norton & Co. study performed late in 1994 found that the cost of not having a help desk is between $6,000 and $15,000 a year per PC. This is the cost of lost productivity as PC users turn to other PC users for help. Implementation of a Help Desk by these organizations dramatically reduces these costs as well as boosts productivity.
It becomes obvious, then, that the proficiency of the Help Desk is critical to the health of the company's bottom line. When the Help Desk is accurately and speedily handling all of its calls, the entire workforce is more productive. This boosts profitability and also promotes employee satisfaction.
In today's typical client/server environment, there is no longer the tight control over the hardware and software that existed in the mainframe world. Each department often uses its own preferred software. As a result, the Help Desk is called upon to field an ever-increasing range of questions and problems. These calls demand that the Help Desk expertise constantly grow.
It is simply not possible to run a moderate to large Help Desk operation without the use of sophisticated software.
This white paper will explain in detail the features of a robust Help Desk software package. The reader may choose to fill out the associated evaluation matrix that is printed in Appendix A for the purposes of comparing one or more Help Desk packages.
The Help Desk Mission
Before setting out to evaluate the over 60 vendors competing in the Help Desk marketplace, it is worthwhile to first consider the purpose, or mission, of utilizing Help Desk Software. The following questions will be useful in determining if a Help Desk is applicable to your corporate situation.
* How many end-users are being supported?
Obviously, a small office does not require the support of sophisticated Help Desk Software. Small organizations, however, might want to consider the use of Help Desk Software if their function is to support multiple external customers.
* How many hardware platforms are being supported?
Distributed computing (i.e. using mainframe, workstations and/or PCs) raises the support ante considerably. It is important to note, however, that whether the organization is supporting one, two or twelve hardware platforms, its Help Desk function should run on only one platform to avoid unnecessary complexity (i.e. it is not worthwhile to have a Help Desk function to assist the Help Desk function).
* How many different software programs are being supported?
The typical corporation runs both vendor-supplied as well as in-house written software. This may include desktop publishing, word processing, spreadsheet, graphics, and any number of other lesser or more involved systems. The point here is that software diversity requires the ability to support a large number of users with quite different requirements. It is not humanly possible, even for he smartest computer jock, for any one person to be an expert on so many different types of software.
* What is the timeliness of support required?
A mobile sales force out in the field needs answers immediately, but casual users might be able to wait 24 hours.
* What is the level of end-user expertise?
If you're supporting NASA, you're lucky since these end-users are most likely self-sufficient and can take care of their own problems. But the rest of us have to deal with varying levels of end-user expertise - from the power-user to the computerphobic.
* What is the level of Help Desk support personnel expertise?
Unfortunately, the Help Desk usually experiences a high turnover. This means that the majority of Help Desk support staff will be at a mid of even junior level. This type of diverse staff requires Help Desk software to contain a wide variety of capabilities.
The majority of readers answering these questions will draw the following conclusions:
- The Help Desk should be able to handle any number of end-users (preferably simultaneously)
- Help Desk software should have teh ability to store knowledge about the intricacies of any type of hardware and any type of software.
- Help Desk software should be able to distinguish between urgent requests for help (i.e. priority for he mobile sales force) and other types of requests.
- All questions, whether they be from naive end-users or power-users, should be able to be answered from the Help Desk's knowledgebase.
- The Help Desk software used should be able to make "experts" out of junior level Help Desk analysts
Ultimately, the Help Desk should act as an extra staff of support personnel. To do so requires that the software have the functionality - and the intelligence - to tailor itself to the organization and grow with it as the organization grows.
The Help Desk Matrix
What it boils down to in the end is choosing the functionality that best fits your organization. Not every Help Desk package is appicable to evey organization. In this section, we present a matrix of functionality that can act as a guide to intelligent selection of "smart" Help Desk software. It is presented in the form of enumerated questions that can be answered through a careful review of product literature and a set of evaluation disks.
Software/Hardware Environment:
1. Is the software easy to install, learn and use?
If you have problems so early in the game, it seems reasonable to assume that you will have problems throughout the lifecycle. As a rule, quality software is intuitive and therefore easy-to-learn and use. Speedy on-line (customizable) graphical systems with built-in help are the recommended environment.
2. Does the software conform to your current operating environment of choice?
Choosing a Help Desk should not require the organization to revamp its environment. This means that the Help Desk should run in conjunction with the software environment the organization is now using including its network and database of choice. It comes as no surprise, therefore, that the majority of organizations desire a client/server Help Desk solution. In this way, the organization can take advantage of its individual environment by dispersing the Help Desk engine to the server and the client interface to the PC, if desired.
In addition, the Help Desk software of choice should permit the organization to add-on modules (i.e. expert system tools) when appropriate.
3. Does the Help Desk provide ODBC support?
As in point two above, a robust Help Desk environment works "with" rather than "against" the current operational environment. From this perspective, ODBC support is crucial. ODBC, which is an acronym for Open Database Connectivity, is a Microsoft-driven specification (quickly becoming a standard) for an application program interface that enables applications to access multiple database management systems using Structured Query Language (SQL). ODBC permits maximum interoperability and is therefore something that no Help Desk software should be without. In other words, the Help Desk's knowledgebase should be equally at home on SQL server as it is on Sybase, DB2, DB2/2, Oracle, Access, Paradox, Ingres or even Btrieve.
4. Does the Help Desk support simultaneous user access?
Unless you're planning oto have only one support analyst, it is important that the software you choose enable multiple analysts - even end-users - to access and update the knowledgebase.
5. Are alternative access paths to the Help Desk supported?
Not all users will use the telephone to notify the Help Desk of problems. A good product will enable end-users to e-mail problems. A good product will enable end-users to e-mail problems. A good product will also enable the Help Desk analyst to e-mail, fax or beeper the problem resolution out.
Call Tracking and Management:
6. Does the Help Desk provide automatic call tracking?
When he telephone rings, the analyst should be provided with the highest level of automation. The knowledgebase should enable the analyst to track a large amount of information about the client and his problem. Trackable information should include: name, client ID, department, phone, fax, address, analyst ID, date call opened, date call closed, priority level of call, subject of problem, resolution of problem, session notes, hardware environment, software environment. It would also be useful if the Help Desk provided the ability to create custom fields. Because calls come rapidly, the Help Desk software must be able to add new users, applications, etc. on the fly.
Alternatively, the organization must be able to archive data and remove user names from the active user list without deleting associated problems.
7. Does the Help Desk enable the analyst to forward and/or escalate the call to another analyst or superior?
The Help Desk analyst should have the ability to easily reassign the case to another analyst or superior at the same or higher level of priority. The software, however, should be able to retain the fact that the case has been reassigned, in effect logging the progress of the case every step of the way.
8. Are there flexible reporting facilities?
Standard reporting is usually insufficient, therefore the ability to perform ad hoc reporting is essential. The ad hoc reports created by the organization should be storable and recallable. In addition, there should be an ability to do a quick print of a problem ticket.
9. Are there management facilities?
Managers and authorized analysts should be provided with a graphical representation of status of all calls by analyst, group, aging, date or other specified criteria.
10. Does the Help Desk software provide the facility to track both hardware and software inventory?
Organizations are finding that they must keep accurate records of their stock of hardware and software programs from both a financial as well as legal point of view.
Artificial Intelligence:
11. Is the Help Desk software intelligent?
Every thing we've mentioned so far comes under the guise of call tracking and management software. While full of functionality, along this does not constitute a "smart" Help Desk. For Help Desk software to be fully functional it should be intelligent. This means that call tracking and management functionality mst be coupled with artifical intelligence techniques including:
a. Natural Language: Using English syntax to converse with the computer when describing a problem (e.g. the screen went blank) is a lot more expedient then typing in a series of difficult to understand codes.
b. Expert Systems: Since knowledge can be efficiently encoded as a series of rules, rule-based or expert systems provide a natural medium to store information such as: If the screen is blank and the cord is out of the wall then plug in the cord. Expert systems provide various techniques to efficiently move backwards and forwards through the stored rules (and/or objects) to rapidly solve a problem.
c. Fuzzy Logic: This technique makes use of incomplete or even somewhat conflicting information. It "guestimates" a solution to a difficult problem if there is no exact match.
d. Case-based reasoning: This AI technique stores solved cases in the knowledge base and learns from this knowledge.
e. Neural Net: The more problems you have to solve, the more data the neural net has to work with. It solves current problems based on stored solutions in the knowledgebase.
A robust Help Desk system uses statistical methods to select the best approach to solve the given problem. That is, one problem might require use of Fuzzy Logic and another might be more efficiently solved using Neural Nets.
12. Does the Help Desk provide a database of common known problem/resolution pairs?
A certain percentage of problems will be easily resovled. The ability to locate these resolutions quickly saves time and money. A rapid resolutions database, therefore, is an effective Help Desk resource.
13. Can the Help Desk handle multimedia?
Since a picture is worth a thousand words, the ability of the Help Desk to pictorially enable analysts and end-users to solve problems is desirable. Images and video are worthwhile additions to the Help Desk knowledgebase.
14. Can end-users access the knowledgebase?
A combination of expert systems and multimedia provide a perfect end-user interface. Enabling end-users to query the Help Desk knowledgebase frees Help Desk analysts for more difficult tasks.
Vendor Support:
15. How long has the vendor been in existence?
The bigger the company the more resources they have to support you and enhancements to their product.
16. Does the company provide technical support?
For how long? And in what way? Most savvy software firms support users in a variety of ways including telephone, fax, and the Internet.
17. Does the vendor provide in-house training?
18. What kind of documentation does the vendor provide?
A manual is nice, but on-line help and a quick reference card is better.
Conclusion
The eighteen questions above cover a lot of territory. Unfortunately, a simple yes or no to each quesion will not by itself help you select the appropriate software. Rather it is better to determine the importance of each question and assign a weighting factor on the scale of 1 to 3 (1 is least important criteria and 3 is the most important criteria).
Next, comparatively rate each product using each of the eighteen criteria. Use a scale of 0 to 5 to assess how well the product you are reviewing meets the specified criteria (0 means that the product does not meet the criteria; 3 means that the product sufficiently meets the criteria and 5 means that the product excels). Now multiply the weightings against the ratings to get the total point value for each of the 18 criteria. Sum up each product's score. The highest score wins.
Article Source: http://www.helpdesk-support.com