Remember your purpose. You are undertaking analysis activities to identify those characteristics of your target performers and of the workplace which are likely to influence the effectiveness of your package. You are also attempting to specify the goals and the objectives for your package. Finally, you will use the analysis data to select the most appropriate strategy. Don’t forget these purposes.
Skip the analysis phase. If you are familiar with the problem, the package requirement, and the workplace, don’t go through a lengthy analysis procedure. Skip the analysis activities if you have a tight deadline.
Use just-in-time analysis. Incorporate analysis procedures within the actual design and development activities. Do your analysis on the fly.
Exploit existing data. Most problems and opportunities can be clarified by reviewing available records and reports. You can use production data, customer complaints, and other existing information to analyze the problem.
Approach the situation from at least three levels. Collect information from your target performers. Also collect information from one level up (managers, supervisors, and mentors) and one level down (staff members, subordinates, and clients).
Remember multiple realities. The perceptions of people at different levels (and also of different sub-groups within the same level) may contradict each other. This does not mean that someone is wrong. In many cases, you need to consider all contradictory perceptions in order to define the problem or the opportunity.
Use a few good people. Don’t conduct a major survey using hundreds of people. In most of the analyses, you can get 80 percent of the necessary information from 10-20 people.
Use focus groups. Instead of mailing out anonymous questionnaires or conducting a telephone survey, assemble a small group and walk them through a series of interview questions. Gathering data from a group is often more cost-effective than gathering data from individuals.
Collect hard and soft data. Whenever possible, obtain hard numbers (such as units produced or profits). At the same time do not ignore soft data (such as opinions, perceptions, and fears).
Mix observation and interviews. In many occasions, what you see may not be what you’re being told. On the other hand, you should not believe everything you see, but should probe for the story behind different performances and outcomes.
Use space holders. If a critical piece of data is delayed, don’t hold up the project waiting for that information. Continue collecting other data or doing other tasks which are not affected by the missing data. In cases of emergency, make a guess about the missing data and proceed.
Set aside time and budget for analysis of data. All the data in the world is of no use unless you have time to make sense out of it. Data analysis takes time and requires people.
Use a computer for data analysis. You do not need sophisticated statistical packages. For most of your quantitative data, use a spreadsheet. For analyzing documents and qualitative data, use personal information management software, such as Agenda.
Document the procedure and outcomes. You do not have to write lengthy reports, but save all the data for reanalysis.
Select your strategies on the basis of various factors. Be sure to take into account the characteristics of the target performers, the constraints and resources in the workplace, and the nature of the target performance in choosing the most appropriate strategies for solving the performance problem (or benefiting from the performance opportunity).
2. Design and Development
Remember your purpose. The package you are developing is not an end in itself, but a means for solving a performance problem or benefiting from a performance opportunity. All the sophisticated design is wasted if it does not contribute to the achievement of your goal.
Maintain congruency. Make sure that the goals and objectives, procedures, and evaluation measures for your package are correlated to each other. For example, if you are developing an incentive system with a goal of selling more computers, it is much more congruent to appraise the productivity of sales staff by counting the number of computers sold rather than by testing their knowledge of computer programming.
Design for use. Your package is for specific target performers in a specific work setting. All analysis data about this target population and this organizational setting should be incorporated in the design and development of the package.
Use a computer. The design of any type of PT package can be made more efficient and effective by using appropriate computer software. In addition to speeding up the design process, the use of a computer also facilitates the frequent revisions of the package.
Design in a modular fashion. For example, a course can consist of a number of modules, each with their own objectives, content, activities, and final tests. Similarly, a packet of job aids for electronic troubleshooting can be designed in a modular form so that the user can select and use the most appropriate procedure.
Keep it lean. In the initial design of any performance technology package it is very important to avoid all superfluous, lengthy, redundant, and repetitious verbiage. During evaluation and enhancement, lean packages make it easy for users to discover additional procedures and content. A padded package, on the other hand, makes it difficult to discover superfluous and unnecessary materials and procedures.
Design in the medium. Messages to be presented in a particular medium should be designed in that medium. For example, create the draft version of an audiotape job aid by talking into a tape recorder instead of writing a text outline. Similarly, create a graphic poster for a team-building activity by sketching on a large sheet of paper rather than typing a description of what the poster is supposed to convey.
Design a total package. While it is important to keep your package lean and simple, it should not lack critical components. The designer should anticipate and provide all materials and instructions for necessary procedures. In addition to materials and procedures for target performance, a complete package should also include materials and instructions for those who are delivering it.
Don’t ignore packaging. Unfortunately, decision makers and users often judge a book by its cover. It is not enough to have good content; you should also have a good package. Frequently, designers of performance technology packages ignore the cosmetic appeal of packaging. This usually leads to less than total acceptance of the package.
Use plain language. Almost all performance technology packages involve the use of language. The designer should keep in mind the level and preferences of the performers who will be using the package. All unnecessary technical jargon and convoluted sentences should be avoided. If technical terminology is critical, it should be carefully defined and illustrated before being used.
Use templates and formalize. Using the same sequence and approach for presenting different components of the package not only speeds up design and development but also makes it easier for users. For example, if a task consists of 20 subtasks, and each of these is presented through a job aid, it is much more efficient to use the same type of job aid for all of the 20 subtasks. In addition, each of these job aids should use the same shapes and symbols to present action and decision boxes.
Provide opportunities for personalization. One of the advantages of modularized design is that it permits different performers to assemble different collections of modules for their use. For example, a powerful incentive system will actually provide a smorgasbord of alternative benefits so that each employee can assemble the combination of benefits best suited for his or her needs.
Keep your eyes on the costs. Throughout the initial design and development period, it is very easy to forget the total cost of the final package. It is very important to frequently estimate the cost of the total package in its final form and to explore various ways of cutting costs in the final production.
3. Implementation and Institutionalization
Plan for implementation from day one. Do not wait until the package is completed before planning for the implementation. It is important that implementation variables affect the design and development of the package.
In planning for implementation, take different factors into account. Your implementation plan should be based on the characteristics of the target performers, the people delivering your package, the workplace, organizational variables, and the characteristics of the package itself. You cannot afford to ignore any of these variables without raising the possibility of major problems later on.
Involve decision makers and end users. In all stages of the project, keep your clients and end users informed of what is happening. This enables you to obtain their inputs and – much more importantly – increase their level of awareness of the project and of the package.
Keep in touch with reality. While the project is progressing, conditions which necessitated the package in the first place may be rapidly changing. However sophisticated and effective your package may be, it is totally useless if it solves a problem that no longer exists. Avoid making your package obsolete by constantly monitoring the current conditions and making appropriate changes to make it more compatible with reality.
Market your package. Just because you have a better mousetrap, it does not necessarily follow that everybody will use your package. You need to inform decision makers and end users of the availability of the package. In doing this, you should take a lesson for good marketing practice and focus on the benefits of the package and not on its features.
Make your package visible. One of the major factors in the adoption of new technological packages is the visibility of the package and its results. This is why it is difficult to get intangible concepts accepted while it is easy to get tangible products accepted.
Make your package divisible. If you have a complex package, provide it in a modular form so that different users can use different components depending on their needs. Avoid making rigid packages which cannot be used in more than one particular way. Keep your package flexible.
Make your package compatible. Any innovative package should be compatible with the needs of your clients and end users. It should also be compatible with the beliefs and values of the target performers. Finally, the procedures used with the package should be as compatible as possible with the habitual way of doing things.
Make your package cost-effective for the end user. Usually your package will be designed to be cost-effective from an organizational point of view. But it is equally important that every user of the package is able to perceive that a significant number of benefits will result from a little effort on their part.
Identify and respond to user concerns. Before you can train a new group of target performers to use a package, they may be concerned about various implications of the package in their life. For example, any computerization in a bank or a retail store is likely to increase concerns among the employees about their future welfare. It is important that you identify these concerns during early stages of implementation (or better yet, while planning for implementation) and genuinely respond to them. Unless the target performers are reassured on various fears and anxieties they have, they are not in a position to effectively implement your package.
Demonstrate your package. No amount of talking about how a package is used can replace an actual hands-on (or videotaped) demonstration. Without a demonstration, your target performers are unlikely to see the big picture and be ready to learn the new package.
Provide OJT. Delay training people on how to use the package until you have gradually increased their level of awareness of the problem (or the opportunity) and the package. You should also have taken care of their primary concerns before training them on use. Provide such training in the workplace and in a format as close to OJT as possible.
Stick around to see what happens. Do not leave immediately after installing your package at the workplace. Stay around and provide troubleshooting help for some time. Provide hotline access for consultation by the users for some more time to come.
Let go. Do not interfere with the implementation of the package any longer than necessary. Encourage local people to continue implementing the package and making appropriate revisions as needed. Remember, your job is not completed until people are able to use the package in your absence.
4. Evaluation and Enhancement
Remember your purpose. All evaluation during the project should be undertaken to either improve your package or to prove that it produces measurable results. The same evaluation activity and the resulting data can be used for these two purposes. Make sure you are not collecting more data than is needed for these purposes.
Integrate evaluation with other activities. Think of evaluation as a component of all analysis, allocation, design, development, implementation, and institutionalization activities. Conduct evaluation with a specific payoff in terms of these other activities.
Cultivate an informant. Make friends with a few representative members of the target performer group. You will be able to collect valuable and valid information from these informants throughout the project. Such information is often more useful than systematic data collected through surveys and posters.
Balance experts and users. Only an expert can provide evaluative data on technical aspects of your package. At the same time, only the user can give you evaluative data on performance outcomes. For example, you need a legal expert to tell you if a new incentive system violates comparable worth laws. You need employees from the target population to see if the incentive actually increases sales calls.
Use more than one expert. You need experts who know the technical area, the implementation aspects, and the end users. For example, in designing a team-building activity, you need to talk to managers who are familiar with the characteristics of the workers, psychologists who are familiar with trust-building activities, and your client who is familiar with the organizational culture. You may want to use more than one expert in each category in order to ensure a wider range of feedback.
Control your experts. Limit each expert to his or her domain of expertise to prevent anyone from commenting on other areas. Also limit the information you give to each expert. Use a checklist or a questionnaire specifying the feedback you want. Ask for revisions rather than indications of problem areas.
Begin with blueprints and mockups. Use blueprints and mockups to get feedback from the target group as early as possible. For example, if you are testing a videotape, begin by asking for opinions on the storyboard and the script. Produce an inexpensive camcorder recording for initial testing of the concept. If you are testing a new economically-designed computer disk holder, begin with an inexpensive mockup to adjust and check on the physical dimensions. In this way you can gather initial feedback inexpensively.
Use automatic data traps. Don’t design new measuring instruments to collect data that can be obtained from existing sources. For example, don’t create a questionnaire to discover how long it takes people to use a computer program. Just program the computer to record that information automatically.
Ask direct questions. Don’t try to use complicated indirect methods of obtaining data unless it is necessary. The best way to find out if the users find a job aid easy to use is to ask them directly.
Collect data on all relevant variables. Performance and accomplishment data are very typical. However, you also need data on the acceptance of the package and other user preferences. Also, collect data not only from the target population but from supervisors and subordinates.
Get to the target performance. The most critical data you need is changes in the performance of the target population and increases in their accomplishments. Make sure this critical information is available and used. It does not really matter how much people like your package if it does not result in measurable accomplishments.
Set aside time and budget for data analysis and revisions. One major purpose of collecting data is to enhance your package. You need time, personnel, and budget in order to get maximum information from your data analysis. You also need to carefully set aside resources for making appropriate changes in your package.
Proceed from informal to formal. During early stages of evaluation, use fewer subjects, do the evaluation yourself, collect the data informally, and make on-the-spot modifications. During final stages of evaluation, use larger numbers of subjects, hire external evaluators, use various controls to collect reliable data, and analyze the data carefully to identify suitable revisions.
