Apr 2 2022

Performance Testing and Pay For Performance Systems

Based on a Performance Improvement Project at Prudhoe Bay in 1987

The Foundation of a Performance-Based Pay System

In most skill-based pay systems, your level or job grade is based on three factors
• Time on the job
• Performance
• Qualifications/skills.

In order to assess skills, tests need to be developed that accurately and fairly assess employees. We completed such a project for Alaska’s Prudhoe Bay oil field, where 1/3 of the oil used by the U.S. was produced back in 1987.

The company wanted a skill-based pay system for sixteen different types of maintenance jobs – which grew to over 20 by the time we finished – as other functions and departments saw what we were doing and how we were doing it – and they wanted in as well.

Along with the environment and the diversity of the maintenance jobs, which varied from communications technicians to diesel mechanics, the biggest challenge we faced was how to test knowledge and skills in a manner that was accurate and easy to administer.

Building the Performance Foundation

I was very busy with client work at the time that my 2 business partners got this project – so my role was limited to modifying my methodology for Curriculum Architecture Design projects and bolting on at the end of those methods, the development of the Performance Tests and then Pilot Testing them.

I did that, and then tested the methods in a parallel effort with my 2 business partners, running the analysis efforts, and then developing the first batch of tests, for 3 target audiences. All as the skeptical clients and informal team leaders from all of the target audiences looked over our shoulders and “grilled” the Master Performers from the first three Pilot Test efforts – out of the box (a track racing metaphor).

Then I developed the training for our staff and our subcontractors, delivered that training, and then went back to Chicago and my other client obligations.

Designing a skill-based compensation system is similar to the front end of designing a comprehensive training program/curriculum, except no training materials are actually developed.

As with most projects, we began first with a job performance analysis – identifying all the outputs and tasks and their key measures and then proceeded to derive the enabling knowledge and skills needed to perform each job.

We have found that the most effective way of getting this data is to use small group meetings with the Master Performers – being the people actually doing the jobs being analyzed to a recognized level of Performance Competence.

Once all the critical knowledge and skills have been documented, the next task is to sort them, based upon criteria such as criticality to the ability to meet the various Stakeholders’ Performance Requirements.

You’ve got to imagine that operating in the old fields on the north slope of Alaska’s tundra, in the arctic circle, were subject to strict regulatory reviews by a multitude of federal and state agencies.

They too looked over our shoulders as we implemented our planned approach.

Developing Performance Tests

The third step in the process is to develop the tests that assess whether or not a person has mastered certain knowledge and skills and can apply them to their Performance Requirements.

We had to be a lot more careful here when writing tests that tie to a Pay for Performance Program. Much more than if they are simply a component of a typical training program – most, but not all, of the time.

People’s pay was going to be based upon these tests, so they had better be “damn” valid, as some of the onlookers observed and both whispered and shouted.

Having a thorough job analysis not only allows you to develop accurate test items, it protects you from a legal standpoint, should anyone challenge the validity of the tests. Having Master Performers create the tests and then others test them went a long way to creating buy-in for the final battery of tests.

The performance tests were simple and yet conclusive. The technician had to demonstrate each task needed to produce the Output, at the worksite, using actual information (such as work orders) and equipment, including any guidance that might be allowed on-the-job (such as vendor manuals). So it was an open book test, so to speak.

An evaluator/assessor – a Master Performer themselves – used a checklist (built into the Performance Test) to monitor the performance as it occurred and then checked-off whether each task in the process was performed per requirements.

If a safety requirement was violated during the test – the test was stopped right then and there – and rescheduled for another day – upping the cost for violating safety policies – as it was sometimes very inconvenient to everyone to have to reschedule a Performance Test.

Safety is a big concern in an environment like this one, where the outdoor temperatures often limited work to 3 or 4 minutes before one had to head into a “warming-hut” to warm up before venturing back out into the cold, cold, cold.

Did I mention that it was cold outside?

My 16-minute video from 1987 – wandering our spartan living and working environment – 16 feet off the tundra.

2 things to note:

at the 3:19 minute mark – note the outdoor temperature on the monitor at one of the doors to the outdoors
at the 7:39 minute mark – note my video selfie – a first for me.

This video has been viewed more than 4500 times.

Performance Testing

My firm did another similar project for technical populations working on the Alaska Pipeline when our client moved to a new company.

We also developed these for branch managers, salespeople, and technicians for an HVAC firm headquartered in the western Chicago suburbs – that I had been working for since 1983 and was one of my long-term clients until I moved from Chicago to North Carolina in 2004. They had titled their initiative “Time to Performance” which is something that many of our clients were doing. Our efforts helped them win an internal award in 1994 from the President of their firm (which has gone through numerous owners and names over the years).

Books

In 2007 my former business partner (1982-1997), the late Ray Svenson, and I wrote a book about these kinds of consulting efforts. That 1987 effort was his project and my methodology.

My more recent mini-book (March 2022) also addresses this kind of effort…