Suppose you are the manager of Phone Number Database a cinema in Rotterdam, and you wonder why your candy sales are lagging behind those of colleagues in other cities. Then you can dive into the data. This shows that the frequency is lower, the turnover is lower, the quantities sold are lower, that certain times peak and others do not, and so on. Big data on devicesBut with larger and more complex data sets, there's little chance that you'll get all those Phone Number Database correlations by chance. And in the example above, you may also overlook that the combination of candy and beer is less common than in your previous job as a cinema manager.
Or that sales of children's films Phone Number Database are performing as expected, but not of adult films (which are leads for further research). Testing gut feeling with data That is why it is often better to get that why question answered from the gut, and to test it against the data you have. If you ask 10 people 'have Phone Number Database you considered buying candy', you might be told that adults Phone Number Database find the popcorn packaging childish. Everyone complains about the childishly designed containers for the popcorn and the straw bottles aren't that attractive either.
Adults feel like a little kid: they want Phone Number Database a bowl of popcorn with a classic American flag and prefer a beer at a real bar. Is that representative? Only then do you enter the data. You look for adults and sales. You quickly see that the turnover of your candy store is particularly disappointing Phone Number Database on days when few children's films are shown. In the data you look for a relationship between the type of films and the turnover in the candy store. This turnover is particularly disappointing for films for '+12'. Perhaps the complaints of a few visitors were more frequent.