I’ve just taken a couple of days off to get a few non-work related jobs done. One of these was kitting out the horsebox for overnight trips (Side Saddle Camp here we come), so I found myself at Asda with a basket that contained the following:
Inflatable mattress and foot pump
Jar of sliced beetroot (we’ve run out)
Impulse purchased bottle of limited edition Heinz ketchup with balsamic vinegar (which, incidentally, is delicious)
Try getting some insightful epos data of out that lot.
It occurred to me (as I asked the checkout lady if I’d won the prize for Random Purchases of the Day) that a significant minority of trolleys and baskets at your average supermarket must be similarly random. They’re full of boxes of beer that were only on offer that week, bread rolls and sausages for the Scouts barbecue or blank CDs, painkillers and a Thomas the Tank Engine toy (me at Tesco this morning).
It must skew the data somehow. I don’t believe that even Tesco’s DunnHumby clubcard database is sophisticated enough to flag every ‘random’ basket or trolley as data to be analysed as distinct to weekly and top up shops.
Perhaps I’m wrong. If I am, I’d love to know how they do it.
One thought on “does my kind of shopping disrupt the data?”
Well I imagine they can filter one person against their usual shop. Bread milk eggs… bread milk eggs… bread milk eggs… inflatable bed and toy gorilla *flag*
Plus they can probably filter against typical shop of each demographic to smooth out the data.
Good point though.