How to Validate an HPLC Method

Ask the outside auditor what is required.

I would imagine a proper job of it would involve a calibration curve done in triplicate , and determining the detection limit. The calibration curve should cover well above and below the expected concentrations. Then test any samples of known concentration to see how you compare. The detection limit is important so that it you report "Not Detected", it is known to less than that value. Also, if you have a auto sampler and programmed analysis capabilities, make sure they are properly set up and documented so if you get hit by a cement truck, your replacement can come in and pick right up on it. Doing in triplicate allows you to do statistics on repeatability of your technique. But I agree with the other post , and ask specifically what they require to make sure you are both on the same page.

Asking the auditor is a wonderful idea. I know from my own experience with our hplc, that doing a 5 point calibration works wonderfully. Some people might do a 3 point calibration as mentioned earlier, but honestly, the more points the better. 1 point is basically meaning less, 3 is good, 5 is better. The more you have the better you are statistically, and you will have a much better time proving to your auditor the reproducibilty of your method.