3.3.1 Quantitative sales forecasting Flashcards
Correlation:
-a statistical technique used to establish strength of relationship between 2 variables
- stronger = closer line of best fit (regression line)
- scatter diagram
dependant variable
y-axis
-being influenced
independant varibale
x-axis
-one causing other to change
Line of best fit/regression line
- forecast sales & identify factors influencing demand
- strong correlation = relationship used to make marketing predictions/decisions
positive correlation
- direct relationship
- close to 1
- as one increases so does other
e.g. sales & advertising or income & sales
negative correlation
- inverse relationship
- close to -1
- as one variable increases other decreases
e.g. price and demand or as interest rate rises there is a fall in demand for new house
4 Pros of using correlation to forecast sales:
1-predict sales & demand factor
2-link = influenced for benefit of business
3-simple & useful tactical thinking
4-regularly = chance correlation exists
3 Cons of using correlation to forecast sales:
1-uncertainty
2-past wont repeat
3-cause & effect or causal link (other factors?) coincidental ?
Extrapolation
- extending line of best fit (dotted line)
- predict future levels such as sales
- analysing trends in past data
3 Affects to extrapolation
1-new competitor
2-increase price
3-external factors
3 Pros of extrapolation to forecast sales:
1-quick and easy to implement & simple & cheap
2-accurate = based on past sales trends (static)
3-quantitative target to predict future sales
5 Cons of extrapolation to forecast sales:
1-doesn’t account external factors/qualitative (trends)
2-assumes past repeats -not likely (comp bus env)
3-some markets (dynamic) (fast move consumer goods)
4-not statistically valid (unreliable significant fluctuations)
5-ignoring significant outliers
Moving averages
- time series analysis
- statistical calculation of an underlying trend in data
3 period: (add 3 together divide by 3)
5 reasons why a moving average is useful:
1-dealing with erratic/personal data
2-average of multiple time periods
3-minimise effect extreme value - take average
4-emphasise direction of a trend
5-reduce ‘noise’ = confuse interpretation
Analysing markets
-several periods at time & averages out the data
- iron out all peaks/troughs in demand
- more accurate figure = whether sales have risen/fallen in market over time
variation
sales in a specific time period - moving average sales
4 ways a sales forecast useful
4 things it creates?
- cash flow forecast = predict receipts and repayments
- revenue budgets = predict budgeted revenue and cost data
- ops plan = used for production/operations plan = predict scale of production = stock requirements and likely C.U
- HR plan = predict staffing levels required
3 key factors affecting sales forecasts:
- consumer trends
- demand changes tastes/fashion
- MS existing competitors - economic variable
- sensitive change in variable (ER, IR, tax)
- strength of economy GDP growth = important - competitor actions
- cant predict
- significant reason sale forecast = over-optimistic
6 situations where sales forecasts are likely to be inaccurate
1-new business (start-up = difficult to forecast)
2-mkt subjective - disruption from tech change
3-demand = sensitive-change price/income
4-product is fashion item
5-significant changes in mkt share (new entrants)
6-management= poor sale forecast ability in past
3 things that makes quantitative techniques more effective in forecasting sales
- business mature (lots past data = see trends)
- industry mature (rapid change = less likely)
- stable external environment (econ, tech ,comp, and legislation less likely to change)
