3.3.1 quantitative sales forecasting Flashcards
define Sales forecasting
is the process of predicting future sales levels by volume or value and future trends
Quantitative sales forecasting is based on
data which can be historic or the result of quantitative research
Sales forecasting will be used to:
- Inform resource management about inventory levels, production output and logistics
- Inform cash flows and budgets
- Aid workforce planning
Time-series analysis shows
past sales figures in date order
Marketers use this historical data, after fluctuations have been smoothed out, to identify trends
Trends are then used to predict future sales
describe moving averages and what they show
Moving averages
Shows whether a trend is significant by smoothing out fluctuations in data
Allows for better identification of an overall trend
Identifies influencing factors on future sales e.g. seasonal, cyclical or random fluctuations
Sufficient data is needed to give validity to the trend identified
how do you calculate moving averages
add up the first 3/4 sales depending on if its quaternary or tertiary then divide it by how many there are eg 3/4
then continue with the rest
what do scatter graphs do
Scatter graphs plot the relationship between 2 variables to identify correlation
what is a correlation
Correlation is the identifying of a relationship between 2 variables
correlations can be
positive
negative
zero
describe positive correlations
the 2 variables move in the same direction
e.g. as temperature goes up ice cream sales go up
describe negative correlations
the 2 variables move in opposite directions
e.g. as road tax prices go up the sales of new 4 x 4s goes down
describe zero correlations
there is no relationship between the factors
e.g. average rain fall and sales of text books
The strength of correlation can be expressed on a spectrum from
-1 to +1
Limitations of quantitative sales forecasting techniques
The further into the future the greater the uncertainty
Sales will be influenced by external influences which are difficult to accurately predict
The past is not always a fair indication of the future
May be manipulated or biased
Inadequate market research
Unexpected events
what is extrapolation
Using past data to extend an identified trend into the future