Calculate your training ROI

Assumptions

• Participants work for 40 hours a week and follow training outside of working hours
• Job performance is rational and linear, and assumes all extra work is equally important
• Data collection is accurate and sample size is adequate and randomised
• Survey results create a valid and workable hypothesis for productivity calculation
• Weighting - the employee cost and recruitment and onboarding are averages, so hiding any variance e.g. larger productivity boost for a higher cost employee will produce larger savings

Considerations

• The prudence principle - when presented with outlying feedback, or numbers in a range, it is normally better to round down or take the lower number.
• Pre- and post-course data - the information required should be decided on and the data collection performed before the start of the intervention in order to create a benchmark, to then re-survey post-course. Relying on hindsight or memory increases the error substantially.
• Error margins - an error margin should be calculated (with at least 90% confidence, if not 95%) as the actual ROI will never be exactly the number you have calculated, but will lie in a range around that number. As an extra step you may also want to consider performing a Monte Carlo analysis - with a number of variables, each with their own error margin, contributing to the final figure, this type of analysis will show you how the result changes according to the range of actual likely numbers for each of the variables.
• For large programs or pilot programs it is always a good idea to have a control group (who are unaware of this fact) in order to compare the results between those that receive training and those that don't. This is the best method to eliminate from the calculation other external effects that were not due to the training.
• In order to improve the accuracy of survey data you should consider extending the data collection to include also the participants' supervisors/peers, so reducing the element of subjectivity in the data.
• Soft benefits and other intangibles do have a place in a business case for training, in the same way that no one number or metric can satisfactorily create an entire business case - for example, improved customer service feedback and response times due to more English speaking staff available, a deeper pool of resources to create multinational teams. These are business benefits, often included as or contributing to strategic objectives, and can be highlighted in the business case even though they cannot satisfactorily be included in an ROI calculation.
• The effective period of the training used in the calculation should be kept to a reasonably short duration, but, as with any training investment, the financial benefits should also be amortised over that period.
• The costs of training normally include only the cost of the actual course itself and any accommodation, travel etc, but not the fixed costs of any internal resources that may have contributed to the marketing, organisation and co-ordination of the training. You may want to consider allocating a portion of the fixed HR costs to the total training cost depending on your company's internal budgeting and cost centre procedures.
• An ROI analysis can not logically be used for any training that forms part of an employee's basic skills portfolio or contributes to compliance training.