he side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health

That's a question often asked on one or two of the training fora I visit occasionally. Unfortunately, the question falls within the unknowable category much like the question, 'How long is a piece of string?' Equally unfortunate is that respondents give answers that are wild guesses; they've ranged from 40 hours per one hour of training and various nominal percentages during the past six months.

As humans, we like to place everything into quantifiable or qualitative boxes ... neat little boxes with comfortable, known dimensions. We can feel safe and secure with quantities or qualities we can understand and control. Of course, it is true that we also have to face the practical necessity to allocate costs to activities and therefore need to know how long these activities take. We need to know how much we should pay a specialist to develop a program. At best, there is always a large element of guesswork. The guesswork occurs because of the large number of variables that aggregate in any instructional design activity. For example, how much content needs to be covered to achieve the program learning outcomes? What is the expected take-up rate by the proposed audience? How will it be delivered (interventions for classroom delivery are less labour intensive than for example, e-learning designs)?

Not only are there the structural variables, there are also variables related to the instructional designer or design team. People with extensive content knowledge in a specific area can produce stepped instruction much more readily than someone with a less extensive knowledge. Often it's necessary to identify one or more subject matter experts to tell instructional designers how content knowledge, skills or attitudes are applied in workplaces.

My experience has been that while it is almost impossible to acccurately determine how long it will take to develop a training activity (one of substance), there are some ways in which an estimate can be made more reliable. Some of the folliwng methods might help.

Based on past experience, calculate costing on averages or work on maxima. That is, if it took 10 hours on average to design a range of instructional programs, it's a resonable proposition that it will take around 10 hours to design something similar. When estimating, it's always better to err on the side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health a

r qualities we can understand and control. Of course, it is true that we also have to face the practical necessity to allocate costs to activities and therefore need to know how long these activities take. We need to know how much we should pay a specialist to develop a program. At best, there is always a large element of guesswork. The guesswork occurs because of the large number of variables that aggregate in any instructional design activity. For example, how much content needs to be covered to achieve the program learning outcomes? What is the expected take-up rate by the proposed audience? How will it be delivered (interventions for classroom delivery are less labour intensive than for example, e-learning designs)?

Not only are there the structural variables, there are also variables related to the instructional designer or design team. People with extensive content knowledge in a specific area can produce stepped instruction much more readily than someone with a less extensive knowledge. Often it's necessary to identify one or more subject matter experts to tell instructional designers how content knowledge, skills or attitudes are applied in workplaces.

My experience has been that while it is almost impossible to acccurately determine how long it will take to develop a training activity (one of substance), there are some ways in which an estimate can be made more reliable. Some of the folliwng methods might help.

Based on past experience, calculate costing on averages or work on maxima. That is, if it took 10 hours on average to design a range of instructional programs, it's a resonable proposition that it will take around 10 hours to design something similar. When estimating, it's always better to err on the side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health

osed audience? How will it be delivered (interventions for classroom delivery are less labour intensive than for example, e-learning designs)?

Not only are there the structural variables, there are also variables related to the instructional designer or design team. People with extensive content knowledge in a specific area can produce stepped instruction much more readily than someone with a less extensive knowledge. Often it's necessary to identify one or more subject matter experts to tell instructional designers how content knowledge, skills or attitudes are applied in workplaces.

My experience has been that while it is almost impossible to acccurately determine how long it will take to develop a training activity (one of substance), there are some ways in which an estimate can be made more reliable. Some of the folliwng methods might help.

Based on past experience, calculate costing on averages or work on maxima. That is, if it took 10 hours on average to design a range of instructional programs, it's a resonable proposition that it will take around 10 hours to design something similar. When estimating, it's always better to err on the side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health

orkplaces.

My experience has been that while it is almost impossible to acccurately determine how long it will take to develop a training activity (one of substance), there are some ways in which an estimate can be made more reliable. Some of the folliwng methods might help.

Based on past experience, calculate costing on averages or work on maxima. That is, if it took 10 hours on average to design a range of instructional programs, it's a resonable proposition that it will take around 10 hours to design something similar. When estimating, it's always better to err on the side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health

he side of higher cost. Thus, instead of using average times for development, if you use maxima, your calculations may have some slack for delays. If I was considering averaging I would choose the maximum duration from the list of programs being used to calculate the average and use it.

It is also possible to allocate the hours and then design instruction to fit the time available. This can be risky, but it isn't as unprofessional as it may seem at first. Say for example, an hour had been allocated for everyone in an organisation to receive an update on occupational health and safety matters. Perhaps a suitable delivery strategy might be to do a 'show and tell' with a handout provided at the end for further consolidation. I've always held the view that 'chalk and talk' presentation methods are not 'training' because they lack a form of assessment; we don't know whether participants have learnt what was intended. However, at times, this type of presentation can be useful and it's hard to argue that people don't leave with knowledge they didn't have when they arrived.

A third method is to ask someone who has completed a similar project how long it took. This can be a rough guide, but often more helpful than guessing without any real rationale for doing so.

As a training manager, when I estimated my annual budget I included a bulk figure for development if I did it separately, or included a loading in my training delivery costs of say, 25%. The bulk development figure was a maximum I had calculated on previous design activities. Usually my team would overspend on one activity and underspend on the next and when the budget was gone, we either had the choice of cancelling other planned training activities and redirecting the funds, not developing any further activities for the year, or putting a business case to our Finance Department for additional funds.

The message in this article then is not to blindly follow the hours, percentages or other information people bandy about on training fora, often with such conviction that they are 'standard', 'industry accepted', or anything else. Do your own research and arrive at estimates that suit the sets of variables that exist in your internal and external environments.

A properly thought out training proposal (or outline) can be a great help in estimating development costs because it focuses on the facts eg, the target audience, learning outcomes, duration of training, delivery method, and much more. These are the data that inform your instructional design team. An added advantage is that if you have your training proposal approved by a funding delegate, your design and development estimate is more likely also to be approved.

Published May 2005. Copyright Robin Henry 2005