Estimating Instructional Development Time

(From the archives: This is a decades-old article by Michael Greer. Can you apply this to your current project planning? It's up to you... Enjoy!) For many years I have been teaching my class Planning Successful ID Projects for IDs and trainers from all different industries.  In that class participants discuss their own rules of thumb for developing instructional materials and then use my worksheets for estimating time and costs for typical ID projects. Like any instructor, I’ve spent a lot of time listening and learning from what my students tell me. And they have told me a lot! Here are some things I’m fairly certain about when it comes to estimating time for instructional development projects:
  • Rules of thumb such as 10:1 or 30:1 (development time vs. training time) vary tremendously, depending on who is espousing them. While I have never heard anyone using a ratio as low as 10: 1, many different people from many different reputable organizations have told me that they place their confidence in 15:1, 30:1, 50:1, and 80:1 for instructor led training. For CBT (computer based training), students tell me that they use 150 – 200:1, depending on complexity. One reputable CBT developer, speaking at an ISPI (International Society for Performance Improvement) conference, said that CBT development time could go as high as 1500 hrs for 1 finished hour! …At the same time, this developer uses 400:1 as an average estimator.
  • Whatever type of training you are developing, what matters most when estimating the development time is the instructional development project management model you will be using. Your model should account for all the activities you will perform that are not directly related to writing and revising instructional materials. I will call this “non-writing” time.
  • Non-writing time typically consumes most (history shows about 80%) of the project time! This non-writing time includes front-end analysis, brainstorming, preliminary design, sponsor/SME review, sponsor/SME feedback, administering/debriefing student tests, and all manner of handholding and administrivia. The ID project management model should not only account for these activities, it should provide means of attaining closure (through sign-off, etc.) on each of them and for keeping things moving. Needless to say, your model should be custom-tailored to your sponsor/project environment.
  • The rules of thumb seldom account for the fact that the deliverables we are developing for a one-hour training session may differ enormously from one course to another. Will instructor-led training provide lecture only, lecture with video, lecture with interactive simulations, case studies with two student roles, case studies with six student roles? You get the point! We need to think about exactly what is happening in this “one hour” of instruction, how many training paths (real-world or CBT) will we be needing to build for students and what specific materials will each path require that we develop? Once we know this, then we can make a reasonable estimate of the writing/revising time and add it to the non-writing time.
  • If you carefully examine your organization’s unique ID project management model and then collect your organization’s historical data related to time spent executing various steps of this model, you will be able to create some fairly accurate custom project estimation worksheets with your own supporting historical data to help you accurately estimate projects. Given such worksheets, you can usually make a detailed, easy-to-negotiate-and-defend project time/cost estimate in as little as 1 1/2 hours. And estimates you come up with will then make a lot more sense for your company than using some vague ratio. (If you don’t have any project history, you still can develop a fairly good estimate by using a detailed, step-by-step listing of all the activities, hand-off points, and so on — in other words by visualizing, in detail, the deliverables to be created, the steps involved in creating them, and the time required to complete each step.)
The bottom line: Don’t trust the ratios, unless they evolved from your specific projects using your own specific people — and even then, I believe such ratios are probably too simplistic and inaccurate to apply to a particular project. By the way, if you want to see an example of a detailed time estimation worksheet for ID projects, see my award-winning article “A Manager’s Guide to Determining Project Scope,” Performance and Instruction, May/June 1988. (This and 36 other job aids are also in my book, ID Project Management.) For an interesting related article discussing many organization-specific rules of thumb for estimating project time, see “How Long Does It Take” by Ron Zemke and Judy Armstrong in the May 1997 issue of Training Magazine (page 69 – 79).