Market Mix Modeling _ Sas Programming 204t55

5/22/2017

Market Mix Modeling | SAS Programming

 (https://www.facebook.com/AnalyticsVidhya)

 (https://twitter.com/analyticsvidhya)

 (https://plus.google.com/+Analyticsvidhya/posts)

 (https://www.linkedin.com/groups/Analytics-Vidhya-Learn-everything-about-5057165)

☰

(https://www.analyticsvidhya.com)

(https://datahack.analyticsvidhya.com/contest/delhi-meetup-by-nagarro-analytics-vidhya/)

Home (https://www.analyticsvidhya.com/)  Business Analytics (https://www.analyticsvidhya.com/blog/category/business-anal…

Market mix modeling – Simpli摏�ed! BUSINESS ANALYTICS (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/CATEGORY/BUSINESS-ANALYTICS/)

SAS

(HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/CATEGORY/SAS/)

SHARE



(http://www.facebook.com/sharer.php?u=https://www.analyticsvidhya.com/blog/2014/08/market-mix-

modeling/&t=Market%20mix%20modeling%20–%20Simpli摏�ed!)



(https://twitter.com/home?status=Market%20mix%20modeling%20–

%20Simpli摏�ed!+https://www.analyticsvidhya.com/blog/2014/08/market-mix-modeling/) url=https://www.analyticsvidhya.com/blog/2014/08/market-mix-modeling/)





(https://plus.google.com/share?

(http://pinterest.com/pin/create/button/?

url=https://www.analyticsvidhya.com/blog/2014/08/market-mix-modeling/&media=&description=Market%20mix%20modeling%20– %20Simpli摏�ed!)

(http://events.upxacademy.com/infosession?utm_source=MLWeekAVBnr&utm_medium=Banner&utm_campaign=Info-Session)



US market spends on an average more than $140 Billion on just marketing every year.

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

1/11

5/22/2017

Market Mix Modeling | SAS Programming

Provided that marketing is such an important component of the total expense by every company, it is very essential to know the actual bene t from these marketing campaign. But today, marketing is not done using a single channel. Every company uses multiple channels like TV, radio, outdoor activities, banners, social media, and many more, to their products. With such a wide spread types of expense, it becomes very di cult to quantify the bene t from each of these  channel campaigns. Market mix model is a statistical model accepted industry wide to quantify these bene ts and optimize the budget allotment to di erent campaigns.

What is a Market mix model ? There is a very famous quote “BRAND IS BUILT OVER YEARS, but managed over quarters”. Quantifying bene ts in shorter window helps business to optimize their spends on marketing. The total sales can be divided into four primary components:  

(https://www.analyticsvidhya.com/blog/wp-content/s/2014/08/Bk-Deepak-rk.png) Market Mix Modeling is an analytical approach that uses historic information like point of sales to quantify the impact of some of the above mentioned components on sales. Suppose the total sale is 100$, this total can be broken into sub components i.e. 60$ base sale, 20$ pricing, 18$ may be distribution and 2$ might be due to promotional activity. These numbers can be achieved using various logical methods. Every method can give a di erent break up. Hence, it becomes very important to standardize the process of breaking up the total sales into these components. This formal technique is formally known as MMM or Market Mix Modeling.

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/



2/11

5/22/2017

Market Mix Modeling | SAS Programming

 

How is Market mix model quanti摏�ed? We have all heard regression and I think regression continues to be savior here as well. This raises a couple of important questions : Why do we need econometric techniques like OLS, multivariate time series models to estimate a marketing ROI? Why can‘t we be simply estimate the ROI, which can be calculated as the ratio of income to the investment or expenditure? The answer to the above questions is simple and straightforward: each brand has a di erent characteristic, tastes, and so di erent market and cost structure. Thus the marketing e orts of each brand are not homogeneous; it depends on di erent kind of promotions and advertising. It is very useful to separate and decompose the e ect of marketing e ort. The purpose of econometric modelling is to create response curves for each type of marketing spend and then use them to calibrate an optimization model to determine a more optimal marketing mix. Response curves, in turn, measure the incremental lift in actual $ Orders per additional $ spent on marketing activities like promotions, advertising, and so on. A response model forecasts the change in a dependent variable, Y, as a function of the change in one or more independent variables, X, e.g., the change in sales of a product as a result of a change in advertising or price of that product or the change in a person‘s preference for a service as a result of a change in the quality, timeliness, or price of the service.  

Decoding Market Mix Modeling  with a Retailer’s sample dataset Suppose there is a large retailer and we have 2 departments, 2 year sales for 2 DMA’s i.e. marketing regions. Wk_end_dt

Region

Department

SBU

Sales

Tvsn_game

06/07/2014

500

D001

FOOD

$10000

467

06/13/2014

500

D001

FOOD

$10500

467

06/20/2014

500

D001

FOOD

$10476

462

06/07/2014

500

D002

CONSUMABLES

$10001

467

06/13/2014

500

D002

CONSUMABLES

$10502

467

06/20/2014

500

D002

CONSUMABLES

$10503

462

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/



3/11

5/22/2017

Market Mix Modeling | SAS Programming

06/07/2014

501

D001

FOOD

$10000

467

06/13/2014

501

D001

FOOD

$10500

467

06/20/2014

501

D001

FOOD

$10476

462

06/07/2014

501

D002

CONSUMABLES

$10001

467

06/13/2014

501

D002

CONSUMABLES

$10502

467

06/20/2014

501

D002

CONSUMABLES

$10503

462

Here is the description of the dataset: Wk_end_dt: Cumulative sales till the weekend date Region: Sales at store level is rolled up for each Region Department: Every retailer combines each item into a category and each category is mapped to a Strategic business unit i.e. SBU Sales: Total sales rolled up at Region wk_end_dt and department level. TVSN_game:TV grp’s for a particular campaign rolled out by retailer. This dataset represents an example of correlated data, several sales measurements may be taken over time on each of the several subjects.Since the same response variable i.e. sales is being measured at di erent times ,we typically refer to such data as repeated data  and the collection of responses on the same subject is often called a cluster responses. When such repeated measures are taken over time, the study is called longitudinal study. There are several approaches to analyze repeated measures data and one of them is general linear mixed model. The SAS MIXED procedure can carry out the computations required to t such a model.  

Market Mix modeling with SAS mixed procedure: Typical SAS syntax to illustrate the use of mixed procedure:  proc mixed data =dataset name;

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

4/11

5/22/2017

Market Mix Modeling | SAS Programming

class region department wk_end_dt;

model sales= base_sales tvsn_grp/s;

run;

This program will give us a coe cient for TVSN_grp which when multiplied by GRP value will give us the contribution by this variable. Contribution =coe cient * value of GRP. GRP= reach X frequency, where reach (expressed in %) is a measure of the number people in the target market exposed to advertising, and frequency is the average number of people have the opportunity to see/hear advertising. For example, 100 GRPs mean 100% of market exposed once or 50% of market exposed twice or 25% of market exposed 4 times and so on. We can then roll up the contribution for the total time period i.e. the period in which we want to know the e ciencies of the campaign and calculate the values given below in the table.   MEDIA DRIVER

Contribution

Investment

ROI(Average)

TV

X$

Y$

X/Y (say 190%)

RADIO

A$

Y$

A/Y(say 100%)

Internet Search

B$

Y$

B/Y(say 90%)

PRINT

C$

Y$

C/Y(say 50%)

In this example, TV is lucrative marketing channel. But all other marketing channel seem to have a higher investment than the additional value they bring in. Radio is a channel which can be considered, provided it brings in some other non-quanti able value. This table is a typical output shared with the clients when dealing with MMM projects. This helps clients to optimize their media mix across various channels.

Typical Challenges faced during MMM: Data issues:

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/



5/11

5/22/2017

Market Mix Modeling | SAS Programming

1. GRP data is basically given by media agencies .The data collected by them can be at di erent levels, i.e. di erent from the requirements of our project.The data has to be sliced and aggregated at various levels which sometimes a ects the accuracy of data. 2. Veracity of data: Sometimes there arises a discrepancy between what is actually planned and what is actually done in the market .There is a certain advertising budget planned for the year and correspondingly there are planned GRP’s but what actually happen’ s is a di erent scenario. These discrepancies sometimes lead to spurious results in the regression Modeling issues:

In MMM we typically calculate response curves of advertising on sales and assume a S shape response curve but due to change in media i.e. typically digital data we can’t expect the same response and hence we get spurious.  

Food for thought Narendra Modi, current prime minister of India,  has rewritten the rules of the game and rede ned Indian politics. Brand Modi has not only captured popular imagination but also trumped Brand BJP but for the next elections, it will be worth knowing the role of various mediums in his stellar electoral performance and MMM is the most apt technique to do that. Dependent Variable: Votes recorded in each region of BJP. Independent Variable: Impressions recorded across mediums or trp’s of news channels, radio when NarendraModi is giving a speech.  

This

article

was

submitted

by

Amogh

Gupta

(http://in.linkedin.com/pub/amogh-

gupta/6/b14/906)  as part of his application for Analytics Vidhya Apprentics programme (https://www.analyticsvidhya.com/blog/2014/06/unveiling-analytics-vidhya-apprenticeprogramme/). Amogh is a master’s graduate in business economics from Delhi University, having a total work experience of ~3 years in analytics.  He has worked across multiple domains like Hospitality, Retail, Manufacturing performing various kinds of analysis and has worked on several techniques like which includes response modeling, market mix modeling, and exploratory analysis. Amogh has worked extensively on SAS products and uses Tableau for visualizations. He is also an avid Swimmer and likes reading books in his free time. 

If you like what you just read & want to continue your analytics https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

6/11

5/22/2017

Market Mix Modeling | SAS Programming

If you like what you just read & want to continue your analytics learning, subscribe to our emails (http://feedburner.google.com/fb/a/mail? uri=analyticsvidhya), follow us on twitter (http://twitter.com/analyticsvidhya) or like our facebook page (http://facebook.com/analyticsvidhya). Share this:

     

 (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=linkedin&nb=1) 154

 (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=facebook&nb=1)  (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=google­plus­1&nb=1)  (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=twitter&nb=1)  (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=pocket&nb=1)  (https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/?share=reddit&nb=1)

RELATED

(https://www.analyticsvidhya.com/bl

(https://www.analyticsvidhya.com/bl

(https://www.analyticsvidhya.com/bl

og/2016/08/bringing-analytics-intoindian-摏�lm-industry-with-backtracing-algorithm/)

og/2015/10/tips-tricks-awesomdata-science-jobs/) 13 Tips to make you awesome in Data Science / Analytics Jobs (https://www.analyticsvidhya.com/bl

og/2016/07/3-analytical-conceptsevery-professional-fresher-inanalytics/)

og/2015/10/tips-tricks-awesom-

(https://www.analyticsvidhya.com/bl

Bringing Analytics into Indian Film Industry with Back Tracing Algorithm (https://www.analyticsvidhya.com/bl og/2016/08/bringing-analytics-into-

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

3 Must Know Analytical Concepts For  Every Professional / Fresher in Analytics

7/11

5/22/2017

Market Mix Modeling | SAS Programming

og/2016/08/bringing-analytics-intoindian-摏�lm-industry-with-backtracing-algorithm/) August 22, 2016 In "Business Analytics"

og/2015/10/tips-tricks-awesomdata-science-jobs/) October 27, 2015 In "Business Analytics"

(https://www.analyticsvidhya.com/bl og/2016/07/3-analytical-conceptsevery-professional-fresher-inanalytics/) July 12, 2016 In "Business Analytics"

TAGS: BUSINESS ANALYTICS (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/TAG/BUSINESS-ANALYTICS/), CASE STUDY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/TAG/CASE-STUDY/), MARKET MIX MODELING (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/TAG/MARKET-MIXMODELING/), MARKETING ANALYTICS (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/TAG/MARKETING-ANALYTICS/), SAS (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/TAG/SAS/)

Next Article

Step by step guide to extract insights from free text (unstructured data)

(https://www.analyticsvidhya.com/blog/2014/08/step-step-guide-extract-inforation-free-text-unstructured-data/)

Previous Article

Bring it on! Analytics Vidhya Author identi摏�cation challenge

(https://www.analyticsvidhya.com/blog/2014/08/author-identi摏�cation-challenge/)



https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

8/11

5/22/2017

Market Mix Modeling | SAS Programming

(https://www.analyticsvidhya.com/blog/author/guest-blog/) Author

Guest Blog (https://www.analyticsvidhya.com/blog/author/guest-blog/)

This is article is quiet old now and you might not get a prompt response from the author. We would request you to post this comment on Analytics Vidhya Discussion portal (https://discuss.analyticsvidhya.com/) to get your queries resolved.

5 COMMENTS

sumit says:

REPLY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/?REPLYTOCOM=17289#RESPOND)

AUGUST 18, 2014 AT 6:34 AM (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/#COMMENT-17289)

q1) model sales= base_sales tvsn_grp/s;



how you are getting “base_sales” please explain

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

9/11

5/22/2017

Market Mix Modeling | SAS Programming

q2) you have use only TV_GRP .. so if I have to use other GRP as well how we use “simultaneously ” in one code explain sas code for that q3) If we dont have “GRP data” can we use MMM & how q4) what will be problem for “Arimax”

Amarjeet (https://www.linkedin.com/pub/amarjeet-sahoo/7/609/145) says: REPLY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/?REPLYTOCOM=17394#RESPOND) AUGUST 18, 2014 AT 6:56 PM (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/#COMMENT-17394)

Hi Amogh, It is really a nice way to present MMM brie y. I would like to ask you if you could explain me the way to calculate the carry over e ect/decay e ect of the Adstock of the TV variable. Thanks.

Amogh Gupta says:

REPLY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/?REPLYTOCOM=17911#RESPOND)

AUGUST 21, 2014 AT 2:38 PM (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/#COMMENT-17911)

Hi Amarjeet thanks for the question. Here are the calculations: Wk GRP Adstock 80% Adstock 80% ,Power .2 Adstock 80% ,Power .2,Lag1 1 100 100 2.511886432 – 2 0 80 2.402248868 2.511886432 3 90 154 2.738445765 2.402248868 4 0 123.2 2.618919453 2.738445765 5 0 98.56 2.504610166 2.618919453 6 80 158.848 2.755474208 2.504610166

aditya says:

REPLY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/?REPLYTOCOM=60494#RESPOND)

JANUARY 3, 2015 AT 7:10 PM (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/#COMMENT-60494)

Please lets stretch on things beyond codes …..how do we know which is the optimum point in a linear regression…..how do we convert this from an in ninte to a nite series where it peaks out  SAS and other things makes everyone believe that we know a lot – which can be dangerous

https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

10/11

5/22/2017

Market Mix Modeling | SAS Programming

Rakesh Jha says:

REPLY (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/?REPLYTOCOM=84708#RESPOND)

APRIL 19, 2015 AT 7:22 AM (HTTPS://WWW.ANALYTICSVIDHYA.COM/BLOG/2014/08/MARKET-MIX-MODELING/#COMMENT-84708)

Hi Amogh How are the interaction e ects and random e ects captured in the model? To have a very simple linear relationship would be not very representative of real examples. Often there are interactions and random e ects? Can you please explain this? Also, I am interested in knowing if any t-test or z-test is done? If so in what stage?

LEAVE A REPLY Your email address will not be published.

Comment

Name (required)

Email (required)

Website

SUBMIT COMMENT



https://www.analyticsvidhya.com/blog/2014/08/market­mix­modeling/

11/11