Accuracy in Surveys (4)

Space Database has the advantage that we manage these CAD files over a long period of time. We revisit the buildings and revise the plans each time there is a physical change. Managing information over time does let us validate and revisit our measurements. Updates on our drawings offer us the opportunity to check the work - to validate the accuracy of the file by comparing them to a new set of site measurements. These activities have allowed us to validate the fact that our processes produce results with far less variance than the industry is accustomed too. In contrast to other measurement methods, our results would have much less variance around the mean.

If Space Database surveyors measured a space several times with our method, the variance would be far less, the probability that any of our calculations would be more than 1% different than the 'actual' area is extremely small. As part of our continuous improvement process we are improving the accuracy of our surveys while remaining price competitive.
In planning a survey project there are two key criteria that must be established at the outset. First one must determine exactly what needs to be recorded to meet the intended use of the finished CAD file. Second, one must decide on the degree of accuracy or tolerance that is required. In executing the work, the surveyor must follow a process that will consistently and reliably deliver results that meet these established criteria.

Accuracy in Surveys (3)

At Space Database we believe that by applying an appropriate set of tools and techniques to simultaneously address all these four sources of error will lead to consistently more accurate area calculations than is currently the norm in the industry.

We carefully test and screen our staff to ensure that have the aptitude and skill to be accurate surveyors. We train people to record information in a ‘self-correcting’ manner that will uncover errors. We calibrate and test our equipment each day. We use multiple measurements of critical measurements to overcome the slight variations in built surfaces. And finally, all of the activities and steps in a survey are controlled through an ISO 9001 certified quality assurance management process. Even with this integrated set of processes, we do not perform ‘perfect’ surveys but we do create work of much higher quality than the industry norm. By focusing on controlling the major sources of error, we produce a product that is much more accurate.

Accuracy in Surveys (2)

You may find this definition of accuracy to be odd. How can two numbers BOTH be correct just because they are close to each other? This strange definition points to the fact that it is impossible to know the exact correct area. Each survey is a more or less accurate approximation of the actual area. It would be nice if after a survey, we could magically compare our result to the ‘true’ number – but this is not the case. We will always have to live with some degree of inaccuracy. According to BOMA, the acceptable limit is 2%. Using the building example from above, this would mean that there would still be $180,000 'up for discussion' each year. Even though this is the allowable limit of the accepted standard, our goal is to be significantly more accurate than this.


What are the causes of survey errors and why would the industry have a tolerance for a relatively large degree of error? Error and variation are caused by several different factors that can and do occur together. The first and most serious is human error - mistakes. People get confused, they measure from the wrong surface, they incorrectly record basic elements and they can transcribe and garble the information that their instruments provide. The second source of error is mechanical measurement error. Devices such as tapes can become stretched and worn. Laser devices can go out of calibration. A third source of error is the variations that occur in building construction. Walls are not straight, consistent or smooth. Structures are not built exactly according to plan. Very subtle angles and variations over long distances can lead to significant differences in interpretation of area. Finally we believe that the methodology, the strategy used to record information and to reconstruct it on a computer can also have a profound effect on the accuracy of the final calculations. Given all these sources of error, the industry has come to accept a variation of up to 2% as being a ‘high quality’ result.

Accuracy in Surveys

We often talk about the importance of accuracy in our surveys. It's generally understood that the quality of information is the essence of its value and usefulness. But surveys are never perfect. Measure something twice and you will always get a different answer. If you don't, then just use more decimal places - and you will see the variance!

So how can we rely on information that can never be 'perfect'?

The answer is that we have to set limits on how much variance we can accept and then strive to develop systems and controls to meet these objectives. Over time we should be able to improve accuracy and narrow the range of what we consider accurate. Although the information may not be perfect, we can set specific tolerances for what is the acceptable range of possible variation.

To understand how accurate CAD drawings and areas have to be, we have to determine the purpose that the information will serve. In commercial leasing, the rent paid for space is based on a square foot rate. Inaccurate areas will either cost the landlord in uncharged rent if the calculated area is too low or create a liability if the landlord has collected rent on too much area. How much could this amount too? If we take a modest office of 10,000 square feet at a rent of $30 per square foot for a term of five years, then an error of 5% would amount to $75,000. If a typical office building of 300,000 square feet were off by 5%, then the total lost revenue or liability would be $450,000 each year!

Another important use of surveyed office interiors is for interior design layouts. In calculating areas, columns are not subtracted from usable or rentable area and are therefore not accurately recorded. But if you are using a CAD file for construction and especially if you are using system furniture, the accuracy of every column and jog in the walls is crucial. A drawing that may have been very accurately drawn to calculate area may be unsuitable for an interior layout. It is therefore very important to determine the purpose of a drawing before doing the survey and creating the file. Because the exact location of drywall covered columns and walls may change during renovations, it is often best to create the 'as-built' survey of existing conditions just before new construction or when new furniture is required.
Once we have determined what needs to be measured and what the purpose of the information is, the next step is to determine how accurate the information has to be. What is the allowable margin of error? What is the maximum tolerance? Currently the BOMA standard accepts a variance of +/- 1% in area calculation. (If two surveys are within 2% of each other, then both are deemed to be correct!)