I hear a lot about “ticket average”, however I think that ticket average is not a very valid way of judging the actual average size of your customers orders.

Shouldn’t we be more concerned about the median ticket price? I’m sure we all have those high dollar tickets that really skew the ticket average for a day - not to mention the very low dollar tickets that skew it just as much the opposite way.

My POS shows “ticket average” - but I’m not sure it really mean much when I’ve had 10+ orders during the day that are 5 time that average.

Anyone else use a “ticket median” - or at least compare the median to the average?

Analysis overkill: :shock:

I mean, I enjoy crunching numbers as much as the next flour covered accountant wanna-be but… geesh. :smiley:

Just get your POS to pop out the standard deviation from the mean and calculate variance statistic and a chi square. I mean, you can even get into MANOVA (multiple analyisis of variance) to determine if a rise in sales on any given night is due to chance or is a statistically significant event.

All fuin and kidding aside (too much graduate stats), You do get a very skewed picture of “average” when you have known large variations from that mean. That goes for really small orders, too. Like when someone pops back in for a 2-liter and it is rung up on a new order.

You can try to eliminate those aberrant items from the report, but that couldrun you into some complicated report generation and programming if you don’t write sequel queries, and don’t have back end access to the database itself.

its not about the stats…its the trends…

up over time is good…

down over time is bad…

(accountant/auditor speak - trend analysis)

:shock: … leafs through the Accountant/Auditor To Pizza Man Translation Book.

Which by the way I’m selling on Ebay right now… more valuable than that “Black Book” thing. :lol:

My thoughts also.

Bank account growing = good trend
Bank account shrinking = bad trend

Oh c’mon - I think it’s a valid comment. And he’s right - mean probably would be more meaningful. Now, I’m not going to calculate it - but I might shoot an email to my POS vendor with a software suggestion. Since most of my software suggestions (and bug reports) are valid, they tend to listen to me…

He was asking about using the median instead of the mean (average) ticket value.

MEDIAN would be a more descriptive statistic when data set has large variance (some really big or really small, aberrant values).

Here is what I do. I run my POS reports every week then every month then every 3 months then 6,9 and 12 months. The 12 month I feel is the best average and I can compare that to each of the other reports so say for example my avg ticket is 22 for week 1 and 28 for week 2 (I had a school order) week 3 was 19 (it was easter weekend) week 4 was 23 my average for the month would be 23 that is going to be pretty average but obviously the longer you go the more accurate you will be. I dont find a huge difference from 6 months to 12 but my market is steady year round.