Know-How Exchange : E-Marketing

Presumably each promotion had an objective -- incremental unit or dollar sales, increased average sale, incremental/variable profit, increased web traffic, ROI, etc. In each case you need to measure against the objective.

To do this best, you need to consider baseline sales (i.e., what would have happened if you hadn't promoted), promotion lift (i.e., how much the results during the promotion differed from baseline), and any post-promotion dip (i.e., how much were results after the promotion below baseline). And, of course, you need to know what your own costs were and what the variable profit was.

That should be enough to start making comparisons across promotion type. Of course, you also want to adjust your data for any seasonal variations and/or unusual competitive activity.

If your business is broad-based geographically, you may want to do separate analyses for each geographic area, as results can vary significantly. Ditto if you have different products, if you think there might be cannibalization within your own business, etc. (The more granular you can track the data, the more insights you are likely to get from your analysis.)

Reviewing how others do this is not likely to shed any additional light on the subject, and you may find it difficult to get hard information on how other companies conduct their analyses. Most companies don't discuss those sorts of things in public forum settings.

If the analysis seems too daunting you can always outsource the analytics. There are experts who do this sort of thing every day, and they can often do the job more efficiently and effectively than a first-timer. (Let me know if you need a recommendation.)