I work in a personal training environment. My manager periodically runs contests "stretch weeks" throughout the year in an effort to boost sales. We see a spike in sales but also a dip just prior to and after the contest. This is due to the staff holding their best leads and customer renewals for the contest week. My manager and I are in disagreement as to whether there is a benefit to continuing this practice.I have acquired two years of weekly sales data. My question is what would be the best way to determine if we actually benefit (monetarily) from stretch weeks using only the sales data before considering additional expenses?I have applied a 3 and 5 week rolling average to smooth the data, but am at an impasse as to how to take that information and say we make (or lose or break-even) $X by holding the contests. If that is even the best approach.Your assistance is greatly appreciated.
      Answer1:Kinda complicated... well its easy, just alot of work, unless the data is already in excel.Basically, you just need to compare the weeks without "stretch" and weeks with "stretch". You gotta section out the weeks. What I mean by that is, say you have 11 weeks of stretch weeks for 1 year. Add w1+w2+w3+w4+w5 divide by 5 (total number of weeks used). This is your first average. Then w2+w3+w4+w5+w6 divided by 5. This is your second average. Then w3+w4+w5+w6+w7 divided by 5. This is your third average. Keep goin until you get all 11. Then plot the averages. Then do the same for the normal weeks, and plot those averages against the stretch averages. Youll see which is better.This is called Moving Averages. You can look it up if you need a better explanation. You can also use Exponential smoothing, which is a little more complicated and I dont think you need it for this.