Question

1. Review chapter 17 in Modern Business Statistics.
2. Read Case Problem 1, Forecasting Food and Beverage Sales, on page 868.
3. Perform an analysis of the sales data.
4. Prepare a report that summarizes your findings in five pages maximum, forecasts, and recommendations. Include the following:
   • A time series plot. Comment on the underlying pattern in the time series.
   • Analysis of the seasonality of the data. Indicate seasonal indexes for each month and comment on the high and the low seasonal sales months. Do the seasonal indexes make intuitive sense? Explain.
   • Deseasonalize the time series. Does there appear to be any trend in the deseasonalized time series?
   • Using the time series decomposition method, forecast sales for January through December of the fourth year.
   • Using the dummy variable regression approach, forecast sales for January through December of the fourth year.
   • Provide summary tables of your calculations and any graphs in the appendix of your report.
5. Submit your assignment for review and feedback.

Answer 1

Your sales forecast is the backbone of your business plan. People measure a business and its growth by sales, and your sales forecast sets the standard for expenses, profits and growth.

When it comes to forecasting sales, don't fall for the trap that says forecasting takes training, mathematics or advanced degrees. Forecasting is mainly educated guessing. So don't expect to get it perfect; just make it reasonable. There's no business owner who isn't qualified to forecast sales--you don't need a business degree or accountant's certification. What you need is common sense, research of the factors, and motivation to make an educated guess.

Your sales forecast in a business plan should show sales by month for the next 12 months--at least--and then by year for the following two to five years. Three years, total, is generally enough for most business plans.

If you have more than one line of sales, show each line of sales separately and add them up. If you have more than 10 or so lines of sales, summarize them and consolidate. Remember, this is business planning, not accounting, so it has to be reasonable, but it doesn't need too much detail. Here are some tips to get you started:

• Develop a unit sales projection. Where you can, start by forecasting unit sales per month. Not all businesses sell by units, but most do, and it's easier to forecast by breaking things down into their component parts. Product-oriented businesses obviously sell in units, but so do a lot of service businesses. For example, accountants and attorneys sell hours, taxis sell rides, and restaurants sell meals.
• Use past data if you have it. Whenever you have past sales data, your best forecasting aid is the most recent past. There are some statistical analysis techniques that take past data and project it forward into the future. You can get just about the same results by projecting your two most recent years of sales by month on a line chart and then visually tracking it forward along the same line. Statistical tools are a nice addition, but they're rarely as valuable in a business plan as human common sense, particularly if it's guided by analysis.
• Use factors for a new product. Having a new product is no excuse for not having a sales forecast. Of course you don't know what's going to happen, but that's no excuse for not drafting a sales projection. Nobody who plans a new product knows the future--you simply make educated guesses. So break it down by finding important decision factors or components of sales. If you have a completely new product with no history, find an existing product to use as a guide. For example, if you have the next great computer game, base your forecast on sales of a similar computer game. If you have a new auto accessory, look at sales of other auto accessories. Analysts projected sales of fax machines before they were released to the market by looking at typewriters and copiers.
• Break the purchase down into factors.For example, you can forecast sales in a restaurant by looking at a reasonable number of tables occupied at different hours of the day and then multiplying the percent of tables occupied by the average estimated revenue per table. Some people project sales in certain kinds of retail businesses by investigating the average sales per square foot in similar businesses.
• Be sure to project prices. The next step is prices. You've projected unit sales monthly for 12 months and then annually, so you must also project your prices. Think of this as a simple spreadsheet that adds the units of different sales items in one section, then sets the estimated prices in a second section. A third section then multiplies units times price to calculate sales. The math is simple--the hard part is making that estimated guess of unit sales.

• A fourth section of your projected prices will set the average costs per unit. You want to set costs because a lot of financial analysis focuses on gross margin, which is sales less cost of sales. For financial reasons, cost of sales, also known as costs of goods sold and direct costs, are different from the other expenses that come out of profits.

  The cost of sales isn't what you pay salespeople or for advertising. It's the amount you pay to buy what you sell. This is usually easy to understand. In any retail store, for example, the cost of goods sold is what the store pays for the products it sells. In service businesses, the costs of sales can be less obvious, but it can still be