Billboards, signage and eye-catching advertisement paraphernalia of different direct selling companies are sprouting everywhere, either local or international. Many companies established names and compete to prolong their standing in the business world. Defined in businessdictionary.com, direct selling is a face to face presentation, demonstration, and sale of products or services, usually at the home or office of a prospect by the independent direct sales representatives. Direct Selling contributes greatly on the economic development of the country; it manifests the Filipino spirit of enterprise and self-reliance. This industry gains greater popularity today than its early years.
One of the popular direct selling companies in the Philippines is Sundance Direct Sales (Footworks Marketing Corporation). It was established on August 1999. Before they came with the business’ name, they first thought of fancy Italian names but they had decided on an original and easily remembered name- SUNDANCE. It was originated from the Hollywood movie, Butch Casedy and the Sundance Kid. Mr. Peter Yu is the managing director of Sundance Direct Sales.
It is a fashion clothing and shoe retailing company with wide expertise in manufacturing industry. This company is engaged in selling of clothes, shoes, bags, cosmetics, accessories and infant and children lines. Currently, it is partnered with Maybelline New York, Afficionado and I2I eyewear. In its 12 successful years of existence, it has approximately 500,000 dealers and continuously increasing. It produced 16 branches and 400 local outlets nationwide.
The mission of Sundance Direct Sales in helping fellowmen is to bring standard of excellence to all parts of the world. In lieu of their mission, they have reached and served international countries such as HongKong, Dubai, UAE, Qatar, Oman, Abu Dhabi, Singapore.
Sundance Direct Sales is an industry that continuously progressing, competing and creating new styles of fashion, awarding them as the Best Direct Selling brand of Apparel.
Many business organizations used tools or techniques, like quantitative forecasting, that helped them determine the possible result of the business operation in the future. Quantitative forecasting technique bases its forecast from past data. This tool helps the manager or the decision maker to accomplish their organizational goals. Specifically, if a company has the record of its past 30-year sales, then it can project the sales for the next year and this may help him to determine the inventory levels, scheduling of production and the like.
However either quantitatively or qualitatively, forecasting is not 100% certain; it has uncertainties so we need to measure the accuracy of the forecast. Forecast accuracy can be measure by MAD (mean absolute deviation) MSE (mean square error) and MAPE (mean absolute percentage error); the best model depends on the measure. The goal of this study is to project the total sales for one of the Sundance branches –Calamba -2011 so that the branch manager can have preparations in their inventory levels.
In order to get the projected total sales of Sundance Direct Sales Calamba branch in 2011, time-series regression and smoothing linear trends were used. Time-series regression is the process of estimating the relationship between two variables- in our case time and sales per month. Smoothing linear trends is just the same with simple smoothing however the intercept and the slope of the trend line are continually adjusted in each period. Two methods were used for comparative purposes. The data were also tested for occurrence of seasonality. MAD was used to evaluate the forecast accuracy since most of the errors were too large. It weighted the errors equally. MSE is not advisable for this kind of problem because it will result to a very large number.
The data collected was a 3-year monthly sales of the Sundance Direct Sales Calamba branch for the years: 2008, 2009 and 2010. The sales representative allowed the author to have the data with proper and legal consent. A hard copy of the data is handed down to the author (Appendix A).
To analyze the data for occurrence of seasonality, the data were graphed first. In figure 1 is an illustration of the comparison of the sales in the three years of operation of Sundance Direct Sales –CALAMBA.
Figure 1.Monthly Sales of Sundance Direct Sales – Calamba in 2008, 2009 and 2010.
Seasonal time series repeats over a specific period such as day, monthly, quarterly or yearly. According to Levin, to determine seasonality two questions must be satisfied. First, are the peaks and troughs consistent? Looking back to Figure 1, there are peaks and troughs in year 2009 and year 2010 that are consistent but if you look at year 2008 it did not follow those peak and troughs. Second, there is an explanation for the seasonal pattern? Since there is no consistency in the trends of the data points, then we will not answer this question anymore. Both questions were not satisfied so we can say that the data didn’t exhibit seasonality. We can now proceed in forecasting the sales using the two methods mentioned earlier.
Forecasting models are evaluated by dividing the samples into two parts: warm-up samples and forecast samples. Warm-up samples are used to fit the forecasting model while forecast samples are for testing the model. In a long time series, data are divided into half.
Using time-series regression, warm-up samples – periods 1-18 –was used to get the equation the best-fitting trend line. Using the data in Appendix B, it was determined that the equation for the best-fitted line is Ft = 2340883.46 + 3800.51(t) .