Introduction

Modern businesses are critically dependent on Information Technology for all aspects of their operations.
However as systems become more powerful and complex, it becomes harder to provide satisfactory
training by conventional means.

• Trainees are often geographically dispersed - often in different continents and time-zones.
  
  It is often expensive and inconvenient to bring staff to a central location for training sessions;
  but it can be equally expensive to send trainers to each site.
  
  
• Training is ideally given just prior to an employee's first use of the system. However with
  a fixed schedule of training sessions, weeks or months can elapse between the need for training
  and the availability of a course.
  
  
• Live training sessions cope poorly with the need for refresher training and consolidation of learning,
  and with variations in people's speed of learning and attention span.

Hence the attraction of self-study and distance learning solutions. These may take a number of forms,
but are typically a combination of e-learning sequences, text-based materials and in some cases videos.

In this article I will outline Mantissa's general approach to the development of training materials
for computer usage, which has evolved from our experience with many I.T. training projects.

The different types of training need

Computer systems differ enormously in their complexity, scope and human interface design.

So the first stage in a project is a training needs analysis, which determines the appropriate focus for the training effort. We study the system and discuss its use in the business with different stakeholders.

What are the common types of training need that tend to emerge from this analysis?

1. System process and data flows
   
   A large system usually involves a flow of information between different departments or business functions.
   
   It is often necessary to give all users a high-level view of the total system operation, so they can see how their activities fit into the larger picture. Specifically, staff are often unaware of the impact of any errors or omissions they make on other parts of the business.
   
   This high-level overview is particularly important where a new system is replacing existing a less sophisticated system or a manual process. The opportunity can also be taken to promote the new system by pointing out the business benefits and opportunities for more efficient working practice.
   
   
2. Concepts of data organisation
   
   Many difficulties in computer usage stem from misconceptions about such matters as what kinds
   of records the computer is holding, and how these are interconnected.
   
   This is a common issue for complex databases. The organisation of these is often not very intuitive to the new user. So a careful explanation is usually required of what these data entities are and how they relate to each other..
   
   Here is an example from a recent project for the commodities group Trafigura.
   
   One of their major systems maintains a database of ports where oil is loaded or discharged; voyages between these ports; and parcels, which represent amounts of oil to be loaded and/or discharged in the course of voyages.
   
   This data organisation is required to cater for the wide variety of trading activities in which the company engages, but it has some features that can confuse the newcomer, and which require clarification via a set of examples and exercises..
   
   
   
   
   
3. Navigational issues and user interface

   Modern computer software can offer a great richness of menu options, navigational paths, features and facilities. This can be a source of confusion to newcomers, who often has great difficulty in identifying the appropriate routes to achieve their purposes.

   Conventional documentation tends to be organised around system functions - how to open a file, how to amend a record. What is more helpful for the user, both in training materials and in documentation, is an organisation around business tasks, e.g. "Create a new customer", "Generate a purchase order", "Record a payment received"
   
   A well designed system will have a consistent look and feel across its various elements - information will be presented on screens in a consistent way, and the same key-presses or mouse clicks will be used to achieve program actions. This greatly simplifies the training task, because the user interface only needs to be taught once. Conversely a system that uses different user interface conventions for different modules or sub-sections will take longer to teach.

   The "user interface" training will also cover such matters as special keys, standard function keys for moving from screen to screen, handling of error message, availability of on-system help etc.
   

4. Procedural training

   Once the scene has been set by way of system overviews, conceptual training and user interface issues, students are ready to be shown how some typical operations are carried out.

   The classic e-learning solution is a series of guided "walk through" simulations of the actual software. At each stage, a training panel accompanies the simulated screen. Students can have data fields explained to them, and can practice undertaking typical business transactions.

   

   As students progress through the worked examples, the lesson will detect errors and give appropriate remedial feedback. It will also point out potential pitfalls and quirks of the software which may need special handling.

   How many procedures will need to be taught in this way? This question can only be answered by the initial training needs analysis, which will address questions such as:

   * What are the different user groups? How big are these groups, and which tasks are common to these groups?
   
   * How business-critical are the tasks? A task may be carried out infrequently, but may nonetheless be very important, and complicated enough to justify expenditure of effort on provision of good training .

   * Is it realistic to expect training on one set of tasks to generalise to tasks in other parts of the system? If the system is well-designed, and there are obvious similarities between the functions within its different components or modules, then a few well-chosen examples may be all that the student needs.

Just-in-time embedded training

There is a 'performance support' philosophy which claims that it is inefficient to provide training activities
as a precursor to use of the system, because users will forget what they are taught, and/or will waste
time by learning functions that they may not use in their day-to-day work.

Instead, users are encouraged to make use of short learning modules, embedded within the application itself. These can be called up at the time of need. In essence, the user is put in charge of selecting
whatever training he or she thinks is necessary for the immediate task in hand.

It is true that if training is available at the point of need, the user will be highly motivated, and retention should be good. So we think that it can make sense to embed procedural task training in this way.

However as the foregoing discussion makes clear, many problems of software usage arise from misconceptions about the "big picture" or about data organisation. The correct time for high-level overview and conceptual training is before the I.T. system is first used.

Roger Kreitman
Principal Consultant