Estimating is driven by two main categories of data: objective and subjective. Objective data—such as material pricing, wages, and equipment pricing—can be researched, quoted, and verified. On the other hand, subjective data—such as productivity rates, material coverages, and fuel consumption—are more often based on judgment and experience.
Subjective data can sometimes be verified or validated using historical data as a basis, making the data more objective. Estimates prepared in a fashion that shows the means, methods, and the data used to arrive at an amount can be reproduced, and in this reproduction any errors in the logic or math used would be exposed. Exposing these errors would allow for correction, and any subsequent adjustment in the amount.
However, this is not the case in some fields where estimating plays a pivotal role. For instance, when a thorough analysis is done on an estimate for the purpose of settling a construction-related dispute, checking for errors in the estimate is critical prior to analyzing for reasonable cost, as an error in the estimate could contribute to the amount being incorrect. Adjustments made to the amount by correcting any known errors allows for the analysis to be performed on the corrected amount.
It is all too common to see estimates prepared for the use in settling construction disputes, whether litigated or not, where the results of the estimate cannot be reproduced. The ways in which some of these estimates disguise their inability for reproduction is slight and not always obvious. There does, however, seem to be several issues in preparing estimates that cause not only an inability to reproduce, but also in some cases an inability to properly analyze the estimate for any purposes. These include: units of measure, labor productivity, and material coverage. Let’s explore some examples of how these issues affect an estimate.
Units of Measure
Units of measure are the unit by which the quantity of a certain item is described, such as square feet of flooring, cubic yards of concrete, squares of roofing, or lineal feet of baseboard. There are four distinct categories related to construction estimating: area, linear, volume, and count. Collectively, these categories form the basis for all the units of measure.
An issue arises when we see units of measure that are uncommon, incompatible, or not units of measure at all, such as location (abbreviated, “loc.”), unit(s), or lump sum (abbreviated, “ls”). A location is a point in space, whether on a map, at your house, or on a piece of paper. It is not a term that measures. Unit is mostly used to describe apartment, hotel, or condo units where a grouping of individual items needed to complete a project may exist, but, again, it is not a unit that measures. Lump sum, while common in all construction estimating, is incorrectly used when it is applied to items that have a native unit of measure, such as a count of an item, or when used in quantities greater than one.
When uncommon units of measure are used, analysis becomes more difficult. A line item in an estimate for removing and replacing window sealant with a quantity of 10 locations does not tell the reader how much sealant needs to be replaced, as sealant is measured by volume comprised of the lineal feet of sealant by the depth of the joint. Most commonly, sealant is stated with a unit of measure of lineal feet, and the depth of joint is contained in the scope or specification for the project. For any thorough analysis to be performed on a line item such as this, one would need to first determine where the locations exist, and then the quantity of sealant in each location. Gathering this information and confirming it can lead to potential issues when reproducing the estimate to check for errors. With units of measure such as “units,” typically a group of tasks is stated with one unit cost, such as remove and replace drywall, door frame, door, prepare for paint, prime, and paint. This grouping of line items with dissimilar native units of measure makes analysis of the individual items an arduous task. It also exposes both parties to include subjective interpretation on the scope and quantities therein. If the scope was thoroughly created, then the quantities of all the underlying tasks in the group would be known. Not stating them allows for interpretation instead of analysis of scope and cost.
Labor Productivity
Of all the subjective data used to create an estimate, labor is typically one of, if not the, largest driver of cost. Labor also happens to contain the largest risk category for most construction projects. Labor in general requires three parameters to properly estimate:
- The labor rate – wages on an hourly, weekly, or monthly basis, burdened or unburdened.
- Production rate – the amount of time it takes to complete a given quantity of the item.
- Quantity of the items to be installed.
For example, say we have a painter that can paint at the rate of 160 square feet per hour, his hourly wage fully burdened is $37.50, and we need to paint 1,000 square feet of surface. We would take our quantity (1,000) and divide it by the production rate (160) and multiply the result by the wage rate to get our labor cost of $234.38 (1000 / 160 = 6.25 * 37.5 = 234.375). The math involved is simple and straightforward when we have all the parameters needed to perform it.
What is commonly presented in estimates is a lack of detail regarding labor, from not having any labor rate listed, to having a lump sum listed for the labor in a given item, to having no labor detail broken out at all. This makes analysis of arguably one of the most important portions of an estimate difficult and requires a fair deal of assumption to be made regarding the underlying parameters of the labor cost.
If given a line item for removing and replacing window sealant with a quantity of 100 lineal feet and a labor cost of $500, we can calculate that the unit cost of labor is $5 per lineal foot. This still does not tell us if the labor productivity or wage is reasonable for the project. We can use a standard labor wage for the given geographical location—say, $42.50—and, dividing our labor total by this wage, we can calculate that there are 11.76 hours. Using this, we could calculate the productivity at 8.5 lineal feet per hour. Yet, after all this analysis, since we do not have the estimator’s actual parameters, we would be making assumptions, and basing our analysis on those assumptions and not the actual parameters used.
Material Coverage
As with labor, material costs included in estimates leave us without the data required for thorough analysis. Materials in general require four parameters to properly estimate:
- Material cost – can include taxes or fees.
- Material coverage – how much does one package of material cover for the given item, for example a gallon of paint may cover 250-300 square feet.
- Material waste – depending on the installation method waste factors can vary, but most tasks waste some percentage of material.
- Quantity of the item.
The quantity of the line item and the quantity of the material can be in different units of measure—an example is that paint comes in gallons, which is a volume measurement; paint is applied at a certain thickness and is measured by square feet. Using the quantity for our item and the coverage, we can determine the quantity of material needed.
Where we run into problems is in the same areas as with labor: We are given simply a material unit cost or lump sum of material costs. We could perform the same analysis we did on the labor and come to a set of costs, coverages, and waste factors, but we would be basing all these on assumptions rather than data used to prepare the estimate in the first place.
Finding a Solution
If the data that is required to create an accurate estimate does not exist, you are not holding an estimate at all but rather a set of theoretical costs that are fabricated and not based in reality. Units of measure are required to quantify the scope of work and break down the tasks needed to complete the scope in a fashion that allows for costs to be applied. Certain parameters are required to apply costs to the scope and the quantities contained therein. Without these, there is no estimate.
In an ideal scenario, we would have estimates containing all the information required to perform adequate analysis, but this has proven not to be the case in estimates presented for the resolution of construction claims. It starts with scope: Defining the work to be done in a way that is clearly communicated and contains proper units of measure so that the final quantities can be analyzed, is paramount. Then there is a need to request the parameters required to prepare costs relating to the labor and material (in some cases equipment and other costs are included in this) that are required to have properly estimated the scope in the first place.
If these are not available, we must assume that the estimate was created using costs calculated in an unconventional way and not in a way common to the mathematics of estimating. This should drive the conversation back to agreeing on the parameters that should be used: the labor wages, labor productivity, material costs, and material coverages. If the scope of work can be agreed to, and the parameters needed to properly estimate the scope can be agreed to, then an estimate using this data could be prepared resulting in an agreed-to cost.
The general hierarchy of inputs that influence cost are as follows: the scope (what needs to be done), the quantities (how much needs to be done), the means and methods (how is it going to be done), and the parameters (wages, production, materials, coverage, etc.). Focusing on these inputs and findings where agreements can be made reduces not only the divide between estimates, but also the introduction of unnecessary subjective data.