Once seated, the wait staff would take the drink order and hand the customers a short survey rather than a menu. The survey would consist of multiple choice and yes/no answers which are selected by filling in the corresponding bubble. For example, 'What type of meat do you like most? ()Chicken ()Beef ()Fish ()Pork ()None', "Do you prefer: ()Sweet ()Salty' and so on. This survey is then fed into a computer which creates a 'taste profile' for the customer. Using this information, it searches a database of recipes to find one that matches the customers taste and sends the information to the chef.

The database could be limited to more simplistic recipes and ones that have common ingredients to help ease the burden of inventory and chef skill that would come with this idea. There is also the possibility of getting food you do not like, but as the profiling became more refined I think it could do a fairly good job at determining tastes.
-- rascalraidex, Aug 18 2013

"Once seated, the wait staff would take the drink order and ..."

Chef/waitress should be allowed to strongly suggest the beverage based on the survey. That way diner gets two surprises.
-- popbottle, Aug 19 2013


Unless your likes change, this Idea means you always get the same meal at that restaurant.
-- Vernon, Aug 19 2013


//this Idea means you always get the same meal at that restaurant//

That's largely the case for me as it is. Most restaurants don't seem to be exempt from Sturgeon's law.
-- ytk, Aug 19 2013


I think that peoples likes change enough to receive different meals. For example, one day you might want beef, the next chicken. BBQ pork one day, salt crusted tenderloin another, and so on. Depending on the length of the questionnaire, even a change in one of the selections could lead to a new meal. Now, I am not expecting a 2 page form or anything, so there is the possibility you would get something the same as last time, but with say, 20 questions, it could produce quite a bit of variety in the meal selection.

I like the idea of drinks being random too, but that would add to the length of the survey. In addition to that, there are not nearly as many selections for beverage as there are for food. However, a 'Do you prefer wine with the meal? []Yes []No' might work. A fitting wine will be selected to accompany the meal.
-- rascalraidex, Aug 19 2013


Ideally this survey would be filled out at great length, but used for marketing purposes later. The actual meal would be based only loosely on the survey. For places where the meal departed drastically from the survey - eg "I marked that I hated sour stuff! And you fed me gooseberries!", it could be justified in the name of edification. "The chef thought that was problematic, and that he could help you overcome those issues."
-- bungston, Aug 19 2013


"That which we love becomes that which eventually kills us."
-- RayfordSteele, Aug 19 2013



random, halfbakery