Scheduling and Pricing Strategy for Television Media’s Commercial Time

Client Background:

Client is a multi-national media conglomerate, with India office headquartered in Mumbai, India.

Problem Context:
  • Television media has 12 minutes of commercial time per hour to sell to advertisers and the value of commercial time is directly proportional to gross rating point (GRP) of the respective program. Based on this, the media company decides the price of each slot P with a +/- k% window
  • Advertisers’ requirements are minimum target viewership and limit on budget
  • Given advertisers constraints, media company has to schedule advertisers for the next quarter with an objective of minimizing media’s total GRP usage and also suggesting optimal price for each slot
Business Problem:

Media company has to schedule commercials from advertisers on a quarterly basis. Each advertiser may have different target audiences and different budget. Advertiser wants to reach minimum target viewership within the specified budget. Business wants to schedule advertisers with minimum possible GRP, subject to advertisers’ constraints and also suggest optimal pricing for each slot.

Solution:
Phase I – Forecasting future GRP for each program:

GRP of a program is directly proportional to viewership. We forecast the viewership for each program for the next quarter using the past data.

Inputs:
  • Program ID
  • Airtime
  • Viewership
Output: 

Viewership (GRP) for each program for the next quarter

Challenges: 

We will not have past data for the new programs. So we also built a forecasting model on time slot level to estimate viewership for the new (or upcoming) programs.

Phase II – Optimally schedule advertisers and suggest price:

Once we get the forecast GRP for each program, we developed an optimization model on top of that to optimally schedule advertisers’ commercials and also suggest price for each time slot.

Input:
  • Program ID
  • Forecast viewership for each program (output of phase I)
  • Slot ID (slot is a duration of 30 mins)
  • Lower and Upper bound for price of each slot (Model should suggest price between this range)
  • Minimum target viewership for each advertiser
  • Budget for each advertiser
 Output:
  • Price for each slot
  • Slot to advertiser mapping
Objective:

To minimize the overall media’s GRP usage so that media can serve as many advertisers as possible to increase revenue.

ROI:

Quarterly revenue increased by approx. 4-5% using the proposed approach.

JOIN OUR COMMUNITY
I agree to have my personal information transfered to MailChimp ( more information )
Join over 3.000 like minded AI enthusiasts who are receiving our weekly newsletters talking about the latest development in AI, Machine Learning and other Automation Technologies
We hate spam. Your email address will not be sold or shared with anyone else.