Reference quantities are called reference for a reason. Accuracy matters. But it also kind of doesn’t.
There are some calculations where you do actually need your reference area and span to be pretty darn accurate.
Aspect ratio is one such calculation, where you find the ratio of your given wing area (Sref) to the area that would be created by making a square out of your wingspan (bref x bref). It’s basically quantifying how long and skinny your wing is. (The Lockheed U-2 has an aspect ratio of around 10, while its cousin the SR-71 Blackbird has an aspect ratio of around 1.7. If you compare their silhouettes, the difference is clear.)
Aspect ratio helps tell us a lot about the efficiency of a given aircraft configuration. It’s used to calculate a number of performance estimates, like endurance and range, which is why it’s important to get a reasonable calculation for it.
On the other hand, when you’re putting together a new configuration in some analysis software, or processing down raw force and moment measurements, reference quantities can be just “good enough”.
Why?
They’re exactly what it says on the tin—reference quantities. These are semi-arbitrary numbers that we use to make comparisons easier.
They’re like adjusting the denominator when you’re adding fractions: if you’re adding 1/3 cup of flour and 1/2 cup of flour together, how do you figure out how much total you have? You find the common denominator, adjust both fractions to match, and then add them. 2/6 + 3/6 = 5/6 cups of flour.
Similarly, we’re just using the reference quantities to make sure the way we nondimensionalize, and compare, all of our physical force and moment data is equivalent.
As long as you’re comparing lift coefficients that used the same reference quantities, you can use whatever you want as the reference. You can get a reasonable estimate of the wing area and use that, or you can say the reference is one square foot and save yourself some effort. Both are valid.
What matters far more than the actual reference is consistency in using it. The moment you start comparing coefficients that used different reference quantities, whatever it is you’re doing is no longer valid.
You can compare the lift coefficient of one airplane to the coefficient of another, different airplane, but it doesn’t tell you anything useful. The reference you used to turn the original lift forces into coefficients is different for each one. Sure, this aircraft has a max lift coefficient of 1.4, and the other one has a max of 1.2, but maybe the first one also has a smaller wing area which biases its coefficient.
This is why I always put such an emphasis on drilling any analysis down to the raw physics of what’s going on. If you take those coefficients and turn them back into forces, you’ll find out real fast which aircraft is actually generating more lift. And that’s what matters at the end of the day.
In short, when it comes to reference quantities, use something reasonably accurate. But use it consistently.