I have searched the forum briefly before posting, but didn’t find what I was after…
Im curious of what method some of you may be applying to roughly forecast your monthly, weekly, or more importantly daily sales.
Ive never forecasted sales before, and after the last two days of googling it, my head is starting to spin. Never the less, I believe I understand; there is no magic formula, Everything you can possibly imagine on the planet could be used from complex formulas to weather and construction and more, and one persons method won’t necessarily work for the next. I also understand in the end the forecast is still just an educated guess, and the importance of consistently updating, tweaking, comparing, and adjusting your forecasting methods as you move forward. -I have read in a few places this may have more weight than the method used to forecast. Well, I hope I understand these points correctly…
My question is: Can somebody give me some examples of simple forecasting formulas to start off with?
Possibly some examples that have worked for them in the past?
I feel like simply averaging my last 3 Friday nights in a row for example, isn’t the best I can do? Possibly taking said average and averaging it with maybe the same three weeks from last year? Im just not sure where to start…
Again, Im not looking for advice on the millions of statistics that could or should be placed in a complex forecast, Im Looking for some help getting my foot in the door with some possibly proven examples slightly more complex and hopefully more reliabe than just a simple average…
Note: I have access to 5+ years of sales data from my POS.
I simply plan for the busiest I have been on a given night of the week and add 10% for good luck. Sometimes I am close and others I am not even in the ballpark.
Hi Daddy, Thanks for taking the time to reply to my question.
Sorry I’m not following you exactly. Could you elaborate slightly? When your looking back at the busiesnt night of the week, How far back are you talking? last week? Last year?
I look at my 13 years of history for the busiest Monday and that is what I plan for on Monday, then I look at each day of the week the same way. There are exceptions such as when I have catered an event I adjust for the difference.
We look at the past three weeks and schedule based off of the average (last 3 Mondays averaged for upcoming Monday, etc). During transition times or one-off events, like when summer shifts into school starting or for our local college football games, we’ll look at the previous year and base our staffing off of those numbers.
It’s definitely more art than science sometimes. For example, right now our 3rd round of Mailshark mailings are hitting and that’s battling against what’s been a downward cycle in our sales. So to my managers I sound like a broken record with my “schedule for growth” reminders.
I had a spreadsheet that aligned the current year, previous year and the year before that by day of the week rather than date. So for today, Monday Sept 17 the columns would be aligned so that Monday Sept 18, 2017 and Monday Sept 19, 2016 were visible to the right. I would start with what we did last year and round to the nearest $100 and forecast by day for months into the future that way. I would update the forecast numbers with actual numbers as time goes by so that all numbers in the past were actual sales. I then had a column for rolling 7 day sales for each year. Each value in that column represented the seven days previous. It is this column that I used to compare sales to the prior year(s) for trend.
About once a week I would update my forecasts for the next few weeks based on current trend and events that I know about. These updates would automatically update the 7 day trend values. For example, if we were trending up 15% from the last year I would bump my forecasts for the coming weeks up by that percentage. I also had a field for notes such as “big soccer tournament” “snow storm” “superbowl” etc to help me remember what was going on when a number comes in sharply up or down from forecast. If I was aware of an event that was coming or one from the previous year that was not repeating I would take that into account. This forecast is what the GM was responsible to buy and schedule for.
I also linked that sheet with another that generates the daily dough prep requirement so that the manager on duty did not have to figure out how much to make… they can just look at the sheet. If the sheet was wrong that was on me. If they did not make the quantity on the sheet that was on them.