This is why I hugely prefer shopping online to shopping at a store. Even not counting the time it takes to drive there, at a real store you are the store's bitch, to manipulate for maximum benefit. An online store, on the other hand, is your bitch--it has to attract you right away to the things you're looking for, rather than prolong contact with those for as long as possible.
I'm not sure that online shops are any less likely to try "to manipulate for maximum benefit". The palette may change (fresh bread smells & video surveillance get replaced by cookies and Referer fields) but the motivations and biases stay the same.
Of course, but my point is that the online shops have a much narrower range to work in. If a grocery store makes you walk around a quarter of a mile to get a ten-item list of necessities, people will accept it. If an online store makes you view half its inventory before you can purchase something, it's not going to make much business.
I greatly prefer shopping online for those things I'm just going to buy all the time anyway (groceries, etc), but strongly prefer shopping in person for things that I buy rarely, or when I'm shopping to have fun.
On the CS side, I would add that if you mash POS data with marketing spend the results are often shocking. Consumer goods companies have terrific raw data but they ignore a surprising amount of it to their peril.
I built a little sales analysis tool for Accenture that they have since sold to several consumer goods clients. It pulls in raw marketing data--pricing, promotion types, locations, revenues--then calculates a few dozen metrics related to ROI, cannibalism etc. Client managers then drill into their marketing data from a couple layers of abstraction and from there they can see fairly clearly how their choices are affecting P&L.
We built that because the leading consumer goods companies were essentially guessing at the repercussions a given sales promotion would have on the bottom line. Today the leaders are more sophisticated but the little guys continue to revere intuition and gut instincts over statistics.
Our pilot client was Procter & Gamble. I pulled in their POS sales and marketing data e.g. endcap on Tide detergent at [x% discount] in [region y] during [phenomenon z]. I mashed that with private and public data sources, laying out the results for trending and drilldown. P&G bought as did L'Oreal and some others. I'm curious how things are going so if anyone reading has worked with these or similar tools, I would love to know how they've evolved.
It's amazing how many expensive marketing decisions are done without regard for scientific analysis. A lot is decided on the basis of how the marketing director/committee feels. In the 90s that was acceptable technology but today it is called Not Learning From Your Mistakes. Compare Google's A/B testing with "throwing ideas at the wall to see what sticks" and you begin to see the problem. Trust me, there are big P&L opportunities in this stuff.
Given the razor-thin margins of grocery stores I'm not sure why they don't try a more ad-supported model. Hang huge colorful printed ads over each aisle, charge fees to the name-brands if they want to be placed anywhere but directly behind the store brand, etc. Seems like they could squeeze more profit than they're currently getting.
I think they do some of those things now, with ads on the floor, blinky-lighted coupon dispensers, end-caps (?) and so on. I suppose it could get uglier, though. ;)
I also think they've become savvier about promoting their generic brands.
