Dickeson–Conducting Job Pathology
Cost estimating lives in a world of virtual reality, forcing decisions based on predictions from tables that are not intended for such usage. Actuaries can only look at groups like white male smokers, construction workers, homemakers, etc., and provide averages and ranges of life expectancy. All we can do with our predictors is provide averages and ranges of job performance expectancy.
Dr. Quincy was a reactionary. He analyzed what happened, not what was to be. Let’s react to job costs and examine the job bodies after the production crimes have been committed. When we multiply transaction costs by actual results, we are doing this. No predictors, just actuals converted to the common denominator of dollars.
With this usage we’re prepared to do internal benchmarking. Our benchmarks are the results of all the other jobs we’ve done in the past. It’s our own set of actuarial experience tables. Compare the actual results of job A with actual experience of all other jobs in the last year, quarter, month or whatever.
Compare job A actual results with all other jobs with similar sets of characteristics. Group the actuarial experience of jobs by customer account. Compare account group X with account group Y. Constantly ask, “Why did we make money on this one and lose on that one?” Where and what are the job differentiators? Think. Discuss. We’re mining the database of experience for the gold that’s in it.
Not to worry about precise accuracy of transaction costs. Get them reasonably close and then freeze ’em, lock ’em into place. We want the jobs and customers to be relative to each other and to the past, so we hold those costs as constants while actual performance supplies the variables. Use a standard cost for paper and ink. Always multiply actual performance by the same transaction values to make the jobs and the accounts directly comparable. Now we can do internal benchmarking.