Measuring Marketing Performance

Question 1

When can you estimate the effect of price promotions using transaction data?

Answer 1

No prior testing = No conclusions can be drawn
* If overtime there was no price variation, there is nothing to estimate.

Same person always finding the same price = No conclusions can be drawn
* If we offered different prices to different consumers in the past, we also can’t estimate a demand curve.

If we offered price promotions in our store over time. Can we estimate the demand curve? Yes, but it’s not that easy.

Question 2

When can demand estimation go wrong?

Answer 2

For example, ice cream sales depend on price, but they also depend on other factors, like the weather. Price may also be set as a function of the weather.

We may observe higher prices when demand is highest, an apparent violation of the law of demand.

This happens because the ceteris paribus assumption is violated, (you are not maintaining everything else constant), when sales are not only impacted by price, but also by other external factors.

Question 3

Ways of Solving the external factors Problem - Clean Variation

Answer 3

• Instrumental Variable Approach (using historical data): Filters good price variation (price variation that does not correlate with other factors of the market, the pure price).
  1. First we must predict prices based on input costs (e.g. price of tomatoes for the production of ketchup), to filter the bad variation.
  2. Then we can regress quantity on predicted price from step 1.
• Price Experiments: Respects the ceteris paribus rule - when we see price vary, this movement is not correlated with other demand factors. Hence, estimation of the price coefficient is not invalidated by other variables correlated with prices.

Question 4

Decision making process when doing advertising

Answer 4

• Need and problem recognition.
• Information research.
• Evaluation of alternatives.
• Purchase decision.
• Evaluate post-purchase behaviour.
• Be persuasive.

Question 5

Elaboration Likelihood Model: Individuals process persuasive messages and make decisions based on two distinct routes

Answer 5

Individuals process persuasive messages and make decisions based on two distinct routes:

Central Route to Persuasion: Uses facts and information to persuade potential consumers. Involves high elaboration and deep cognitive processing. Emphasizes detailed product information, features, benefits, and unique selling propositions.
* Example: Camera commercial, shows detailed technical specifications, advanced features and the superior image quality produced by the camera.

Peripheral Route to Persuasion: Uses positive associations with tips. Involves lower elaboration and relies on heuristics or surface-level cues. Utilizes attention-grabbing visuals, endorsements, emotional appeals, or other peripheral cues.
* Example: Advertisement features a well-known and admired celebrity photographer using the high-end camera in various picturesque locations.