Showing recommendations based on previous purchases might indeed be a difficult feature to implement. But if it's just a list of best sellers from the same category, than it would be fairly simple to add and a good ASP programmer could tackle it in a limited amount of time (www.earlyimpact.com/cpd).

We don't plan to add such a feature in the short term. If we ever do, we would definitely make it an option that the customer can turn off. That way you don't upset customers that believe the feature is not useful and/or annoying. They would simply turn it off.

This area of the forum is for posting suggestions for additions to ProductCart - that's what I'm doing.

It's not right for you, fair enough.  But without ideas and dialogue it's hard to evolve a product.  It might be right for others.  The use of ProductCart is varied and ranges from small few product stores to bigger ones with thousands.  I wasn't even concerned if it was relevant to my store, I was thinking of it as a generic tool to feed into consideration of future ProductCart evolution.

I was using Amazon as an example not as a definitive source.  I know its first generation e-commerce style won't fit with the more communal approach going forward that is emerging.   Your experience of their recommendations is clearly poor, but I have found it works far better for me.  

However there are all kinds of ways to do recommendations - not just a crass version, ignorant of customer choices.

I genuinely wasn't trying to be patronising, you never know the knowledge base or experience of who is posting and giving an illustration is often helpful.

I agree that cross selling in PC is powerful - I use it myself in a sensible and careful way and have found my sales have gone up significantly, particularly when combined with a simple reward scheme.

A big factor for me around cross selling, recommendations and similar is in building and keeping trust over time with customers. 

cheers

Mark

E-commerce stores do this all the time without problem.   The biggest in the world - Amazon, makes a substantial part of its revenue this way on precisely the product types you mention.   I have used it many times and it is very good indeed.  For example in music it picks by genre and artist, it's not complicated but it works.  The reason it works is that customers will follow leads they are given so long as its a matter of their own choice and exploration.

Not everyone would have to use it - like best sellers or specials selections.  Stores have choice.  However, recommended selections is an industry standard that people understand.

Nobody expects it to be perfect.  There are all kinds of simple ways of profiling the data that goes in.  It's our job to help customers find the products they like.

Some other software stores already have this feature.   EarlyImpact are an innovator, especially at the pricepoint of the software so I like to offer practical suggestions either I can use or I have already seen in action. 

cheers

Mark