lecture 3 Flashcards
According to marketing evolution
an analysis technique that allows marketers to measure the impact of their merketing and advertising campaigns to determine how various elements contribute to their goal, which is often to drive conversions (sales)
A more workable definition
Aims to measure the correlation between total sales/revenue in each day/week/month and advertising spending or impressions on that same day/week/month
Key differences to attribute modelling (media mix modelling)
Uses aggregate data
Longer time horizon in the analysis (months vs weeks)
Wider range of channels, both traditional and digital
Incorporate external influences such as seasonality, promotions
Modelling done “less often”
Seeking answers to the following strategic questions + data driven approach
1) how does a firms marketing activities correlate to a KPI of interest
2) What ist he optimal mix of marketing activites for a given KPI
Data driven approach:
1) Media mix modelling (MMM)
Estimate how different marketing activities impact a KPI
how? Via linear regression or some other statistical model
2) media mix optimization (MMO)
Adjust budget allocation across marketing activities to optimize a KPI of interest
Media mix modelling
seeks to understand the role each media type plays on driving the overall campaign performance
Marketing mix modelling:
take a more holistic view
It doesnt just condier media channels but also factors in other marketing activities
Pricing strategies, product distribution, and even macroeconomic indicators
About viewing the larger picture and determining how these varying elements interplay to impact the overall marketing performance
Comparing alternative mmm models simple linear regression
same coefficient generated for each panel for every week
Comparing alternative mmm models
Mixed model
coefficient now vary by panel (fixed + random estimates) but are similar across weeks
Comparing alternative mmm models
state space model
coefficient vary for each panel by week giving time varying dimensions
so far assumed Advertising spending in time period t only impacts
KIP in t
Spending yesterday doesnt impact revenue today
Marketing activities do not have a saturation effect
-> every dollar spennt per channel is equally effective
these are strong assumptions
Temporal effects of advertising
Current effect in bursts no carry into timeline
B: conveyor effects of long duration, slowly declining slope
C: conveyor effects of short duration, many spikes and slow fall off
D: persistent effect
Spike, decrease but stay
non linear response to advertising
Concave response, (concave curve)
Linear response (linear increase)
S-shaped response (s shape?)
How to allocate a marketing budget over multiple channels
assume there is one KPI we want to maximize (e.g. revenue)
We can predict revenue based on media mix model
We have a fixed budget to allocate
Simplification for today
We assume we want to max sales in this time period only by allocating spending across channels in this time period only
Media mix optimization as two step process
simulate different budget allcoations and their impact on a KPI choose the best allocation
Assumes the media mix model that predicts revenue already exists
There is a constrained optimization problem
How to do this is beyond the scope of this class but the intuition is not
Max predicted revenuet
subject to Total expendituret (facebook ad spending + tv ad spending <= budgett
Its hard to think of media mix models in causal terms because
omitted /lurking variables
MMarketing spending is not randomly allocated
-> careful when interpreting outputs
As a prediction model this is OK
How I think about MMM, “can I predict sales with marketing activities”
But then, the coefficients dont have causal interpretations, so does MMO make sense?
its a tricky space
Some recent developments trying to wrestle with this…
