Estimating E-Learning Course Length

Knowing how long a course is going to be is important for many reasons.  In an ideal world, a course should be exactly as long as it needs to be to meet its goals—not a minute less, not a minute more.  But there are some real-world reasons why you might need to contain the length.

• You need to put together a cost estimate and schedule, and course length will drive that.
  Conversely, you may have budget, resource, or schedule constraints, and you need to keep the course length within the limit of what you need to accomplish.
• You may have external requirements on the course length.  For example, the state of California requires that all supervisors take a course in harassment prevention that is two hours long.  It can’t be any shorter than that.
• You may have restrictions on how long your learners can spend on the training.  For example, you may be training volunteers who may be scared away if they know they have to go through a six-hour orientation program or sales staff who would lose too much revenue if they had to take a day away for training.

If any of these are factors, then you’ll want a good length estimate up front that you evaluate as you go.

The up-front estimate is the hardest.  You may not have well-defined content, and you may not know how much detail you need until you dive in.  We recently completed a mobile course on troubleshooting heavy equipment that went from an initial estimate of 30 minutes to a final product of 2 hours.  No one really knew until we started gathering content.

If you have well-defined materials, it is a little easier.  The rule of thumb is that you can accomplish the same thing in e-learning as you can in the classroom in about half the time.  In the nine years we’ve been in business, we’ve found this to be true across clients and subject matter.  (Perhaps my next post will be on why that is.)  Even if you aren’t converting an existing instructor-led course, you can still use that metric to help.  I’ve found that if I ask a subject-matter expert how long they think an e-learning course needs to be, they often don’t know and won’t even hazard a guess.  But if I ask them for a “gut” estimate of how long it might take them to explain everything in the classroom, they can usually come up with something.  We then estimate the e-learning course will be half of that length.

So now you have a starting point.  As you design and develop the course, it is important to check back in on that number.  Assuming your business reasons for having a length goal are still there, you’ll need to monitor and adjust as needed.  For us, the best place to do that is after the first draft of the storyboards.  By that point, we have a very good idea of what needs to be covered and to what level of detail.  At the same time, we aren’t so far in that it would be too difficult to make some changes if we are off target.

So how do you estimate at this stage?  Well, a few years ago, we analyzed a number of courses to help us do just that.  We had people take a broad sample of courses that we had developed and keep track of how long it took them.  Then we analyzed the storyboards for number of slides, words in the narration, and words in the entire storyboard to come up with some averages.  Sure, there were some outliers, but a lot of the courses ended up with similar numbers.

Here’s what we found.  One hour of content contained, on average:

• 8680 storyboard words
• 5179 transcript words
• 53 screens

Having these numbers came in handy recently.  We were working on a volunteer training project, and we had a target length in mind.  As we were developing the content, the client started talking to their volunteer base about it.  They got significant feedback concerning the length of the course.  They wanted it to be easy for volunteers who might be using time-limited computers such as those at a library.  We analyzed the storyboards and got an idea of how long the course was and how much we might need to cut.  Sure, we had some rework, but better at the storyboard phase than after the courses had been built.

You are welcome to use our numbers for your own estimating, but what would be more useful would be for you to do your own evaluation.  For example, you might include more programmer instructions in the storyboards than we do.  You might have less audio and more text that the student reads.  Analyze how you tend to design courses and come up with your own numbers.  Then adjust the estimates if a certain project is different.  For example, we don’t use these numbers for computer simulations.

What do you do to estimate the length of projects?  What’s working for you?