You may want to look at the method described in this article:
I believe your are trying to perform a market basket analysis, which simply requires a specific self-join to achieve. This analysis is really looking at the combination of products that customer may purchase together, which sounds a lot like what you are trying to achieve.
Hope this helps!
1 of 1 people found this helpful
Good answer there from Wilson regarding a quick and dirty affinity analysis solution in Tableau. It can however get pretty messy if you have very many products and you also have that mirror-image view as well which isn't the most optimal use of screen pixels.
You might want to run your data through some proper data mining routines first to get your associations and accompanying statistical significance and other measures. You can also then get common triplets, quads and more, rather than just pairs. RapidMiner is a good opensource tool for this sort of analysis.