…one common occurrence that warns you that a performance discrepancy may be lurking around is the announcement that takes some form of “We’ve got a training problem.”  Someone has detected a difference between what is desired and what is actually happening.

But statements such as “We’ve got to train/teach…” are pits into which one can pour great amounts of energy and money unproductively.  Such statements talk about solutions, not  problems.  Training (teaching, instruction) is a solution, a remedy – a procedure used to achieve desired results.  It implies transferring information to change someone’s state of knowledge or ability to perform.

But lack of information is often not the problem.

Robert Mager/Peter Pipe, Analyzing Performance Problems, p. 8

If lack of information is not the problem how do you find out what the problem is?  One of the first questions one should ask when looking at a performance problem is, “what does ‘right’ look like?”  In performance terms “right” is defined as optimal performance.  Actual performance is the current way the work is being done.  When starting a project, one should not assume the problem lies is the “actual” way the work is being done.  One should also not assume that training will resolve any perceived problem. The problem may have a cause nobody has considered that cannot be resolved through training alone.

A fundamental step in understanding a performance problem is documenting the actual and optimal performance.  The difference between actual and optimal performance is the performance gap.  In many organizations it is likely that optimal performance has been documented at some point.  This can be in the form of a manual or an existing training course.  One should consider that the problem may be caused by a lack of accountability to existing standards or failure to send people through existing training.  Reviewing existing documentation can provide a good starting point for any needs analysis.  It provides valuable background and a reference point from which to begin data collection.  However, one should also consider other factors that may be causing a performance problem.

The focus of this post is how to document and analyze a performance gap.  I will cover how to close a gap in a later post.

There are many ways to approach a gap analysis.  To effectively understand an issue a combination of analysis techniques is usually required.  Below are seven types of analysis ISPI recognizes:

  • Job or Task Analysis—Identifies the required activities, information, processes used, and outputs produced and then compares that to actual practice.
  • Process Analysis—Identifies the cycle time compared to process time; time at task compared to time on rework, waiting, or checking; resources consumed and the cost of those resources; and what drives activity (customer or product requirements).
  • Work Environment Analysis—Identifies and evaluates the effectiveness and efficiency of feedback, the reward and incentive system, information and communication systems, work and process designs, and work tools and equipment.
  • User or Audience Analysis—Identifies current expectations, perceptions, physical capability and capacity, and knowledge and skills.
  • Communication Systems Analysis—Identifies and evaluates the availability, capability, capacity, upgrade ability, and cost to use and maintain.
  • Market Analysis—Identifies the size, competition, growth, current and potential constraints or limitations, organizational expectations, initiatives, capabilities, and capacity.
  • Data System Analysis—Identifies and evaluates the capability, capacity, availability, upgrade ability, and cost to use and maintain.

Speaking from experience it is unlikely that a performance specialist would possess all the skills required to successfully complete each of these forms of analysis.  This is one reason why ISPI emphasizes partnerships.  If the stakeholders on a project feel strongly that any of the methods listed above are necessary to gather data they should also be prepared to employ a specialist to complete the data collection.  Although the work can be done by a layperson it will add time and may impact the overall quality of the data.  Ultimately the decision is a matter of cost and the benefits derived from the expenditure.

Regardless of who is conducting the analysis, the participants play a critical role in the quality of the data.  Participants should be representative of the entire team in knowledge, experience, and responsibilities.  A control group can be helpful in validating and clarifying data.

Access is another critical success factor.  Being able to observe and interact with job performers in their work environment may be necessary.  However, striking the right balance between observing and interacting can be a challenge.  In my experience it is very important to be open with the participants about your needs and expectations.  It can be unsettling for an employee to be observed.  This is where it is critical to have a sponsor who can communicate with employees and address their concerns.  Interacting with participants may influence your data so should be considered with caution.  Direct contact with participants can also negatively impact their productivity which may also negatively impact the data.

Once optimal performance is documented, the analysis should attempt to validate while documenting the actual performance of workers.  This is where the control group is useful.  It is possible that changes have taken place since previous standards were written that require an organization to revisit their expectations (optimal performance).  Finding out how to compensate for the gap is where the needs analysis can get complicated.  It would be a mistake to assume the gap is caused by a lack of knowledge.  Unfortunately, many organizations treat all performance issues this way.  If the problem is caused by performers lacking knowledge, it is logical to assume the solution is traditional training.  However, there may be another cause.  But that is a discussion for a future post.