Flashcards
Market Analysis
Cards in this deck (57)
- 1Principles of Data Reduction (9)1. Headings
2. Comparison elements in columns (not rows)
3. Labelling
4. Gridlines and Gaps
5. Summary figures: averages and totals
6. Ordering: sort by decreasing brand size
7. Brands in rows
8. round to 1-3 significant figures
9. Display relevant units (% or $) and scales ($0,000)
10. Give brief written summary
11. Include a title
12. Alignment (usually to right) - 2Market Share (%)definition: percentage of total brands sales in a market
calculation: (brand purchase total / category purchase total) x 100 - 3Penetration (%)definition: proportion of shoppers who are buyers
calculation: (brand buyers / shoppers) x 100 - 4Purchase Frequency (times)definition: Purchase rate relative to the buyers of a brand
calculation: brand purchase total/brand buyers - 5Purchase Rate (times)definition: the purchase rate relative to the shoppers of the category
calculation: brand purchase total/shoppers - 6Share of Category Requirements (SCR%)def: The proportion allocated to a brand of the purchases made in the category by the buyers of same
calculation: (brand purchase total/total accumulative category purchases by buyers of that brand) x 100 - 7100% Sole Loyaltydefinition: The proportion of buyers who buy only the brand
calculation: (YES/Buyers) x 100
(If buyer of brand is loyal count yes) - 8What are the brand size measures?Market Share and Penetration
- 9What are the brand loyalty measures?Purchase Frequency
Purchase Rate
Share of Cat Req.
100% Loyalty - 10What are 2 patterns we see within BPM's?Double Jeopardy and Duplication of Purchase Laws
- 11Double Jeopardy• Small brands have fewer buyers who purchase less often
• Big brands have more buyers who purchase more often - 12Duplication of Purchase Law• Small brands ALWAYS share their buyers with big brands
• Big brands SOMETIMES share their buyers with small brands
• Brands share their buyers with other brands in line with their market shares - 13What is the purpose of the Dirichlet Model?The Dirichlet is an empirical generalisation that explains 'how people buy' and 'how brands compete and perform' in the market
- 14What tasks is the Dirichlet used for?Forecasting tool to develop marketing strategies and assess the effectiveness of current marketing practices (e.g. promotion, distribution)
Analysis of market trends and BPM's - 15How do people buy? (4)1. INFREQUENTLY - non and light buyers
2. RANDOMLY - buy regardless of previous purchases (zero-order)
3. REPERTOIRE - multi-brand buying
4. HETEROGENEITY - people are different in their preferences - 16Dirichlet models fit when?1. Deviations are small (+/- 10% relative to average observed values of BPM)
2. The Mean Absolute Deviation (MAD) is small
3. There are no obvious patterns in the data - 17How brands compete and perform? (4)1. Double Jeopardy
2. DoP Law
3. Repertoire buying - No Partitioning
4. New or light buyers are important (Penetration over PF) - 18Purpose (Applications) of Polarisation (3)1. Comparing a brand against the rest of the category
2. Comparing a small brand against a big brand
3. Evaluating the effectiveness of loyalty programs - 19Advantages of PolarisationStandardised measure of brand loyalty without the effects of brand size - enables benchmarking
- 20Types of loyalty (2)Attitude: Consumer preferences e.g. I like coke
WEAKER FORM - not directly captured in BPM's - less relevant
Behaviour: Measured by looking at actual purchases made by consumers over a given period of time - enhances predictability - directly affects brand performance - 21Polarisation index formula 0:1(repeat rate - market share) / (1 - market share)
0= pure heterogeneity, no loyalty
1= maximum homogeneity, extreme loyalty - 22Conditional Trend Analysis (purpose)-Predict future purchasing behaviour of customers of a brand
-Identify the source of sales growth or decline of a brand
-Explain the impact of marketing activities on non, light, medium, and heavy buyers - 23random nature in period-by-period purchasingin period-by-period purchasing, it is normal that non and light buyers increase their purchases, while heavy buyers decrease their purchases.
- 24Implications for marketers1. Theoretical benchmark for future sales based on past performance
2. Helps in marketing planning activities
3. Tracking of brand performance - 25CTA FORMULAS2|x = (x+1) * B(x+1)
Predicted sales in Period 2, conditional on the shoppers who made (x) purchases in period 1.
Where B(x+1) is the predicted number of shoppers who made (x+1) purchases in period 1from the NBD model. - 26Best-Worst formula tasks-Need to understand and predict consumer preference
-Need to rank importance of one attribute to another
-preferred attributes by consumers
-identify consumer segments for product attributes - 27Methods to measure consumer preference RATING METHODRating method
1: not imp 10: ext imp
advantages: simple, easy to present, not costly
disadvantage: questionable data quality - 28Methods to measure consumer preference RANKING METHODRanking method
Most important to least important
(numbers do not reoccur)
advantages: most appropriate method for conceptual mapping of values
disadvantages: difficult for respondents, complex, time-consuming, costly - 29Methods to measure consumer preference CONSTANT SUMCONSTANT SUM
divide 100 points between attributes
advantages: Able to capture small differences in consumer preferences disadvantages: Complex when the number of attributes is too high - 30Methods to measure consumer preference Method of Paired ComparisonMETHOD OF PAIRED COMPARISON
Compare between 2 attributes at one time
advantages: good information on the reliability of a respondent’s choices
disadvantages: requires many comparisons - 31BEST-WORST METHODtick which is most important (best) and which is least important (worst)
- 32Underlying theory and advantagesrandom utility theory
Easy to conduct and interpret
Good for comparative research (i.e. segmentation techniques can be applied to the data)
Provides reliable information for respondent choices
High forecasting ability - 33Best Worst FormulaCount (best) - Count (worst) / frequency*# of respondents
- 34Conjoint Analysis definitionConjoint analysis is a method that determines how consumers value different attributes/attribute levels that make up products and services
- 35Conjoint Analysis Marketing Tasks-design new products to enhance consumer value
-predicting market share of a new launch
-identifying market segments
-measure consumer preference for a product attribute
-identifying combination of attribute levels of product attributes that has the highest choice probability - 36Conjoint Analysis PurposeUsing data from fieldwork (discrete choice experiments) to determine the feasibility of product attributes, and consumers' choice probability (or likelihood) of product attribute combinations.
- 37Conjoint Analysis Applications-Calculating dollar values for attributes
-Calculating predictive market shares
-predicting purchase intentions - 38Conjoint Analysis dollar value formulachange in price/absolute b value
- 39Bass Model Purposeused in forecasting adoption of new products
-attempts to predict how many customers will adopt the new product and when they will adopt - 40Bass Model implications1. Forecasting new product adoption
2. Decision-making about viability of new products (particularly durable products and technology)
3. Thinking of new product performance - 41Bass Model Assumptions (8)1. The market potential N remains constant.
2. The customer decision is binary e.g. either the adopt or not adopt.
3. Uniform mixing e.g. everyone can come into contact with everyone else.
4. No repeat or replacement purchase of the innovation.
5. Marketing strategies supporting the innovation do not influence adoption.
6. The value of q is fixed throughout the lifecycle of the innovation.
7. Imitation always has a positive impact- no negative word of mouth.
8. Sales of an innovation are considered to be independent of the adoption/or non adoption of other innovations. - 42Bass Model Coefficientsq: the coefficient of imitation (internal influence word of mouth)
p: the coefficient of innovation (external influence knowledge/advertising)
q>p = Imitation dominates innovation (slow growth)
p>q = Innovation dominates imitation (instant hit) - 43Bass Model ParametersN : Total number of potential buyers of new product (constant)
n(t) : Number of new adoptions at exactly time (t).
N(t) : Cumulative adoptions of the product up to time (t). - 44ADBUDG Response Functions PurposeIdentify the effects (relationship or correlations) between two or more elements of comparison, such as
-effects of marketing activities: advertising costs, promotion spend
-product performance: sales performance, market share, profit - 45Linear Response ModelY = a+bX
Dependent variable Y = sales level
Independent variable X = actions of marketing
a = y-intercept
b = coefficient that defines increase/decrease and strength of increase/decrease - 46Pros and Cons of Linear Response ModelPros:
-easy to understand
-interpolation
-given market data, can use standard regression methods to estimate parameters
Cons:
-unrealistic - no upper bound on Y
-extrapolation
-gives managers unrealistic guidance on decisions because it creates a false impression that sales continue to increase as advertising increases - 47ADBUDG response functionshows relationship between advertising and sales
-estimates optimum level of advertising
Y = sales
X = advertising
a = maximum boundary for sales values
b = minimum boundary for sales values
c = determines shape of the curve
d = affects the slope/angle of the graph function - 48ADBUDG response function s-shape/concavec > 1 = s-shape
c < 1 = concave shape - 49Price elasticity definitionMeasure of the responsiveness of a change in demand for a good or service to a change in its price.
- 50Price elasticity tasks-How much to increase price to cover production cost increase
-How much to reduce price to maintain market share when competitors lower prices? - 51Price and demands PE valuesPE = 1
Demand rises by same percentage that price falls
PE > 1
Demand rises by more than price falls (gold price)
PE < 1
Demand rises by less than price falls - 52Demand Curves variablesIndependent variable (P)
Marketing actions that the marketer can control (e.g. price, advertising)
Dependent variable (Q)
The measure used for monitoring and evaluating marketing actions (e.g. sales quantity or demand) - 53Price Elasticity formulaPE = (Q1-Q0)/Q0)/(P1-P0)/P0)
- 54Demand Curves Linear Demand Curvepros: simple
cons: does not reflect what we expect, with extrapolation to higher prices it generates unrealistic demand (negative demand for high price) - 55Demand Curve Constant Elasticity Demand Curvepros: better reflects what we expect, with extrapolation to higher prices it generates more realistic results
cons: with extrapolation to lower prices it generates unrealistic results (for low prices it estimates very high demand)
- 56Demand Curve Legit Demand Curvepros: better reflects what we expects, generates more realistic results when extrapolating to either higher or lower prices
Cons: complicated
- 57—
Information
- Description
- Theory aspect of Market Analysis
- Created by
- Danielle Weir
- Views
- 187
- Tags
- Rating