A typical 4 wire load cell has 4 resistors in a bridge.
They are wired so that opposing resistors are to experience the same strain and adjacent resistors are to experience the opposite strain. This results in the excited load cell being forced out of a balanced condition under load and a voltage to appear on the sense pair under load.
When measuring a load cell you can find opposite terminals by measuring the largest values between two terminals. This will be equal in value to the element resistance R=(R+R)||(R+R). If you measure across two adjacent terminals you should read a value of R'=R||R+R+R=0.75R.
If the top and bottom pairs are not equal in value you will have a much harder time determining the exact values as you will have to take more readings and possibly short out other terminals to gain enough known values to calculate the unknowns. It is certainly possible but less fun.
Once you have found the opposite pairs of terminals you need to provide a excitation current and then see which polarity the sense pair supplies and determine the gain of the bridge with applied maximum load. If you plan to use some cheap devices with little documentation then I would purchase a bunch, and test them to load and current and temperature destruction (dielectric insulation too), check the gain and resistance tolerance and yield and THEN if they meet your need I would order a lifetime supply hoping to receive the same type and redo the tests in case they are from a different batch.
If you need to reorder do so from a reputable source with a data sheet and pay for them from the profits of your first successful run.