Maybe we shouldn’t use computers for estimating. I can’t believe I wrote that. I am Mr. Computer. My first estimating computer had a whopping total of 16 kilobytes of RAM with two 5¼-inch, single-sided floppy drives. Yeah, I can hear you young whippersnappers asking, “What’s a floppy drive?” That hardware is certainly archaic by today’s standards, but it was enough for me to fall in love with computers. There is no doubt that we gain tremendous benefits by using computer estimating programs. However, there also are some potential problems. 


During my tenure with one of my employers, we switched from manual to computerized estimating. Here is how a typical estimate went in our office, before and after the change to computerized estimating. 


B.C. (before computers)


Takeoff was done as many of you still do it today. We had paper drawings, colored pencils, highlighters, a thumb wheel (counter/clicker) and paper forms. For linear measurements, we had a Minerva mechanical wheel. The ⅛ scale only went up to 2,000 feet. I can’t tell you how many times I looked down to see a measurement of a few hundred feet, wondering how many times I had gone over the 2,000-foot maximum. 


The spots and dots (fixtures, receptacles, etc.) were counted and listed on an 11-by-17-inch takeoff form. The branch was listed on a branch takeoff form, and the feeders were listed on a feeder takeoff form. What a pain that was. I must have erased some counts a thousand times because I found just one more outlet or switch. I almost bought stock in the Pink Pearl eraser company. 


Then the real pain started. We had to break each count down to its component parts and list them on a pricing sheet form. We then had to price and labor each line on the form and then multiply each line for material and labor totals. Finally, the material and labor columns had to be totaled. And, of course, as soon as I finished, the boss would revise the labor, or some better pricing would come in, and I would have to do the math all over again. 


The calculator also would inevitably run out of paper or ink with only 30 minutes left to get the bid out. 


A.D. (after digital)


Then I got my first estimating computer (yes, the one with 16 kilobytes of RAM). No more forms and calculators. No erasures and recalculating. The counts and measurements went directly into the computer. The system even had a linear wheel that didn’t max out at 2,000 feet. The system exploded the assemblies, applied pricing and labor to each component, and then did all the math. All I had to wait for was the nine-pin dot-matrix printer (there were no onscreen reports). The system saved about six to eight hours per estimate.


So why do I ask if we should be using computers? Sometime after switching from manual to computer estimating, I noticed I was losing touch with my estimates. I no longer had material prices memorized. I did not have the confidence of having labored the material as the job conditions required. I was not as sure as I used to be that everything was in the estimate. I had fallen into one of several computer traps. I was using all of the time saved by the computer estimating program to get more estimates completed. I should have been using some of the saved time to study the estimates. Instead, I allowed the computer to take the art out of the estimate. It was the old “the computer is always right” trap, and, of course, it is not. 


In spite of the amazing advances in estimating technology, human intervention and review is still required. There are many items to be looked over in every estimate. Here are a few: Is the labor factoring accurate? Are there any large quantities that may need to be quoted? Are the components of the assemblies what you expected? Have you made any typos? Does the material list match the way you built the project in your mind? Are the material prices what you expected? 


Some estimating software packages are capable of making a few of these decisions. However, it is still your responsibility as an estimator to review the computer’s results and make any needed changes and corrections. You still need to “massage” the estimate. 


When first learning estimating, I found it was good to learn the manual method, as it gave me a complete understanding of the process. Moving on to computer estimating should be done with caution. Make sure that what comes out of the computer matches your expectations. If it doesn’t, find out why.