SlideShare a Scribd company logo
1 of 11
Download to read offline
Confidential
Multivariate Media Models :
Providing a metric tool that
identifies significant Media Placement
Multivariate Algorithms Inc
www.multivariate.com
Confidential
Overview :
Multivariate Algorithms Inc has a statistical tool that can
identify key characteristics of high call volume spots.
Looking at the creative, placement and cost of the ad, we have
identified “efficient” combinations that can assist the client .
In addition, if adopted could have significant
insights into what is doing well and leverage this as a point of difference.
Confidential
Multivariate Media Models :
Used Call and Spot files for Oct 1999
Coded final calls into 10+ and less than 10 per spot
“Really defines the MOST reactive callers within the best spots”
Predictor variables used were :
Weekend versus Weekday - Day part
DRTV – Network – Program Type – Net Cost
Creative - Platform
Confidential
Multivariate Media Models :
The Technique used to model build is proprietary.
It can be classically described as
a Non-sequential CHAID that uses Interaction Terms
Rather than assume a constant effect for each predictor variable this
approach calculates ALL POSSIBLE interactions and then
chains them into thousands of selection rules
Confidential
Multivariate Media Model Results :
Examples of real combos include –
------------------------------------------------------------------------------------------------------------------------------------
| QTMS |
| V3.2 ALL 7-WAY SEGMENTS OF PROJECT NOCOST, SORTED BY RESPONSE INDEX, DETERMINED BY VARIABLES |
| CABNET DAYOWEEK DAYPART LENGTH NCOST NETWORK PLAT |
| CONDITIONS ARE CONFIDENCE LEVEL=95%, LIFT=50%, MINIMUM NUMBER OF SOLICITED=100, MINIMUM NUMBER OF RESPONDERS=0 |
| TOTAL NUMBER OF SOLICITED=42,210, TOTAL NUMBER OF RESPONDERS=7,210, OVERALL RESPONSE RATE=17.08% |
| |
------------------------------------------------------------------------------------------------------------------------------------
| NO.OF NO.OF RESPONSE CONF. INTERVAL| |
| # SOLICITED RESPONDERS INDEX RATE(%) LEFT - RIGHT | DESCRIPTION OF SEGMENT |
------------------------------------------------------------------------------------------------------------------------------------
| 1 107 46 252 42.99 33.61 - 52.37 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F |
| | DAYPART=9AM - 3:59PM LENGTH=60 |
| | NCOST=2 NETWORK=America One |
| | PLAT=Unlimited |
------------------------------------------------------------------------------------------------------------------------------------
| 2 100 4 23 4.00 0.16 - 7.84 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F |
| | DAYPART=7PM - 10:59PM LENGTH=60 |
| | NCOST=2 NETWORK=America One |
| | PLAT=Sprint Sense Anytime XTRA |
------------------------------------------------------------------------------------------------------------------------------------
| 3 107 4 22 3.74 0.14 - 7.33 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F |
| | DAYPART=7PM - 10:59PM LENGTH=60 |
| | NCOST=2 NETWORK=Fit TV |
| | PLAT=Sprint Sense Anytime XTRA |
------------------------------------------------------------------------------------------------------------------------------------
Confidential
Multivariate Media Model Results I :
( Using 1,187 significant combinations we classified
Keepers and Losers unto the 42,210 spots )
Keepers = 17,852 spots = 153,979 calls = $52,784,516 = 342 CPC
Losers = 24,358 spots = 78,808 calls = $32,019,299 = 406 CPC
Savings
Average CPC is 364 ( 84MM / 232,000) = 364 – 342 = 22 * 153,979 = 3,387,538
Confidential
Multivariate Media Model Characteristics:
0
10
20
30
40
50
60
9AM-3:59PM
4PM-6:59PM
7PM-10:59PM
11PM-1:59AM
2AM-5:59M
6AM-8:59AM
Keepers
Losers
Confidential
Multivariate Media Model Characteristics:
0
10
20
30
40
50
60
70
80
15Seconds
30Seconds
60Seconds
Keepers
Losers
Confidential
Multivariate Media Model Characteristics:
0
10
20
30
40
50
60
70
80
90
W
eekDays
W
eekEnd
Keepers
Losers
Confidential
Multivariate Media Model Characteristics:
0
10
20
30
40
50
60
70
80
90
100
Cable
NETW
ORK
Keepers
Losers
Confidential
Conclusions :
There appears to be a preliminary model that can
* Score and separate out spots that reduce the CPC
by shifting Day part placement, weekday placement.
*Shifting 15 second spots to 30 second spots.
Achieving a $22.00 CPC savings for 150,000+ calls

More Related Content

Similar to Media

CX for Utility Companies - Uses Cases & Future | SoftClouds
CX for Utility Companies - Uses Cases & Future | SoftCloudsCX for Utility Companies - Uses Cases & Future | SoftClouds
CX for Utility Companies - Uses Cases & Future | SoftCloudsSoftClouds LLC
 
Algoritmo di text-similarity per l'annotazione semantica di Web Service
Algoritmo di text-similarity per l'annotazione semantica di Web ServiceAlgoritmo di text-similarity per l'annotazione semantica di Web Service
Algoritmo di text-similarity per l'annotazione semantica di Web ServiceMichele Filannino
 
Optimizing SaaS Growth with Effective Product Metering
Optimizing SaaS Growth with Effective Product MeteringOptimizing SaaS Growth with Effective Product Metering
Optimizing SaaS Growth with Effective Product MeteringPrasanna Hegde
 
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...VMworld
 
mobilda_media_kit_2016_web.PDF
mobilda_media_kit_2016_web.PDFmobilda_media_kit_2016_web.PDF
mobilda_media_kit_2016_web.PDFAlexander Shyfer
 
Servicing Optimization - Improving Performance in Highly Regulated Times
Servicing Optimization - Improving Performance in Highly Regulated TimesServicing Optimization - Improving Performance in Highly Regulated Times
Servicing Optimization - Improving Performance in Highly Regulated TimesClaren Financial Services, Inc.
 
Sap performance testing best practice guidev1 0-130121141448-phpapp02
Sap performance testing best practice guidev1 0-130121141448-phpapp02Sap performance testing best practice guidev1 0-130121141448-phpapp02
Sap performance testing best practice guidev1 0-130121141448-phpapp02Kamalaksha Das
 
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02Pompee Das
 
SAP Performance Testing Best Practice Guide v1.0
SAP Performance Testing Best Practice Guide v1.0SAP Performance Testing Best Practice Guide v1.0
SAP Performance Testing Best Practice Guide v1.0Argos
 
Alteryx Telco Use Cases
Alteryx Telco Use CasesAlteryx Telco Use Cases
Alteryx Telco Use CasesTridant
 
IRJET - An Auction Mechanism for Product Verification using Cloud
IRJET - An Auction Mechanism for Product Verification using CloudIRJET - An Auction Mechanism for Product Verification using Cloud
IRJET - An Auction Mechanism for Product Verification using CloudIRJET Journal
 
Part1 of SQL Tuning Workshop - Understanding the Optimizer
Part1 of SQL Tuning Workshop - Understanding the OptimizerPart1 of SQL Tuning Workshop - Understanding the Optimizer
Part1 of SQL Tuning Workshop - Understanding the OptimizerMaria Colgan
 
CA Performance Management 2.6 Deep Dive
CA Performance Management 2.6 Deep DiveCA Performance Management 2.6 Deep Dive
CA Performance Management 2.6 Deep DiveCA Technologies
 
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...IRJET Journal
 
smarter-field-work-management-systems
smarter-field-work-management-systemssmarter-field-work-management-systems
smarter-field-work-management-systemsRadha Swaminathan
 
Rapid Fire Analytics Attribution Strategy
Rapid Fire Analytics Attribution StrategyRapid Fire Analytics Attribution Strategy
Rapid Fire Analytics Attribution StrategySameer Khan
 

Similar to Media (20)

Sourcebook 2017
Sourcebook 2017Sourcebook 2017
Sourcebook 2017
 
CX for Utility Companies - Uses Cases & Future | SoftClouds
CX for Utility Companies - Uses Cases & Future | SoftCloudsCX for Utility Companies - Uses Cases & Future | SoftClouds
CX for Utility Companies - Uses Cases & Future | SoftClouds
 
Algoritmo di text-similarity per l'annotazione semantica di Web Service
Algoritmo di text-similarity per l'annotazione semantica di Web ServiceAlgoritmo di text-similarity per l'annotazione semantica di Web Service
Algoritmo di text-similarity per l'annotazione semantica di Web Service
 
Optimizing SaaS Growth with Effective Product Metering
Optimizing SaaS Growth with Effective Product MeteringOptimizing SaaS Growth with Effective Product Metering
Optimizing SaaS Growth with Effective Product Metering
 
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...
VMworld 2013: Create a Key Metrics-based Actionable Roadmap to Deliver IT as ...
 
mobilda_media_kit_2016_web.PDF
mobilda_media_kit_2016_web.PDFmobilda_media_kit_2016_web.PDF
mobilda_media_kit_2016_web.PDF
 
Servicing Optimization - Improving Performance in Highly Regulated Times
Servicing Optimization - Improving Performance in Highly Regulated TimesServicing Optimization - Improving Performance in Highly Regulated Times
Servicing Optimization - Improving Performance in Highly Regulated Times
 
Sap performance testing best practice guidev1 0-130121141448-phpapp02
Sap performance testing best practice guidev1 0-130121141448-phpapp02Sap performance testing best practice guidev1 0-130121141448-phpapp02
Sap performance testing best practice guidev1 0-130121141448-phpapp02
 
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02
Sapperformancetestingbestpracticeguidev1 0-130121141448-phpapp02
 
SAP Performance Testing Best Practice Guide v1.0
SAP Performance Testing Best Practice Guide v1.0SAP Performance Testing Best Practice Guide v1.0
SAP Performance Testing Best Practice Guide v1.0
 
Alteryx Telco Use Cases
Alteryx Telco Use CasesAlteryx Telco Use Cases
Alteryx Telco Use Cases
 
Supply chain network design
Supply chain network designSupply chain network design
Supply chain network design
 
Supply chain network modelling
Supply chain network modellingSupply chain network modelling
Supply chain network modelling
 
IRJET - An Auction Mechanism for Product Verification using Cloud
IRJET - An Auction Mechanism for Product Verification using CloudIRJET - An Auction Mechanism for Product Verification using Cloud
IRJET - An Auction Mechanism for Product Verification using Cloud
 
Part1 of SQL Tuning Workshop - Understanding the Optimizer
Part1 of SQL Tuning Workshop - Understanding the OptimizerPart1 of SQL Tuning Workshop - Understanding the Optimizer
Part1 of SQL Tuning Workshop - Understanding the Optimizer
 
CA Performance Management 2.6 Deep Dive
CA Performance Management 2.6 Deep DiveCA Performance Management 2.6 Deep Dive
CA Performance Management 2.6 Deep Dive
 
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...
IRJET- Supply Chain Network Design for Plant Location and Selection of Capaci...
 
NAME's Appendix - H
NAME's Appendix - HNAME's Appendix - H
NAME's Appendix - H
 
smarter-field-work-management-systems
smarter-field-work-management-systemssmarter-field-work-management-systems
smarter-field-work-management-systems
 
Rapid Fire Analytics Attribution Strategy
Rapid Fire Analytics Attribution StrategyRapid Fire Analytics Attribution Strategy
Rapid Fire Analytics Attribution Strategy
 

More from Daniel Kocis Ph.D. - Chair (11)

Overview
OverviewOverview
Overview
 
Channel Strategy1
Channel Strategy1Channel Strategy1
Channel Strategy1
 
Results for Keith
Results for KeithResults for Keith
Results for Keith
 
Case Conditions
Case ConditionsCase Conditions
Case Conditions
 
How_to_maximize_the_number_of_spots
How_to_maximize_the_number_of_spotsHow_to_maximize_the_number_of_spots
How_to_maximize_the_number_of_spots
 
AD1
AD1AD1
AD1
 
Consumer Models
Consumer ModelsConsumer Models
Consumer Models
 
BuildHistoryfinal
BuildHistoryfinalBuildHistoryfinal
BuildHistoryfinal
 
Weekly Forecasts
Weekly ForecastsWeekly Forecasts
Weekly Forecasts
 
QTMS-EM-Combinatorical Model
QTMS-EM-Combinatorical ModelQTMS-EM-Combinatorical Model
QTMS-EM-Combinatorical Model
 
Pharma
PharmaPharma
Pharma
 

Media

  • 1. Confidential Multivariate Media Models : Providing a metric tool that identifies significant Media Placement Multivariate Algorithms Inc www.multivariate.com
  • 2. Confidential Overview : Multivariate Algorithms Inc has a statistical tool that can identify key characteristics of high call volume spots. Looking at the creative, placement and cost of the ad, we have identified “efficient” combinations that can assist the client . In addition, if adopted could have significant insights into what is doing well and leverage this as a point of difference.
  • 3. Confidential Multivariate Media Models : Used Call and Spot files for Oct 1999 Coded final calls into 10+ and less than 10 per spot “Really defines the MOST reactive callers within the best spots” Predictor variables used were : Weekend versus Weekday - Day part DRTV – Network – Program Type – Net Cost Creative - Platform
  • 4. Confidential Multivariate Media Models : The Technique used to model build is proprietary. It can be classically described as a Non-sequential CHAID that uses Interaction Terms Rather than assume a constant effect for each predictor variable this approach calculates ALL POSSIBLE interactions and then chains them into thousands of selection rules
  • 5. Confidential Multivariate Media Model Results : Examples of real combos include – ------------------------------------------------------------------------------------------------------------------------------------ | QTMS | | V3.2 ALL 7-WAY SEGMENTS OF PROJECT NOCOST, SORTED BY RESPONSE INDEX, DETERMINED BY VARIABLES | | CABNET DAYOWEEK DAYPART LENGTH NCOST NETWORK PLAT | | CONDITIONS ARE CONFIDENCE LEVEL=95%, LIFT=50%, MINIMUM NUMBER OF SOLICITED=100, MINIMUM NUMBER OF RESPONDERS=0 | | TOTAL NUMBER OF SOLICITED=42,210, TOTAL NUMBER OF RESPONDERS=7,210, OVERALL RESPONSE RATE=17.08% | | | ------------------------------------------------------------------------------------------------------------------------------------ | NO.OF NO.OF RESPONSE CONF. INTERVAL| | | # SOLICITED RESPONDERS INDEX RATE(%) LEFT - RIGHT | DESCRIPTION OF SEGMENT | ------------------------------------------------------------------------------------------------------------------------------------ | 1 107 46 252 42.99 33.61 - 52.37 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F | | | DAYPART=9AM - 3:59PM LENGTH=60 | | | NCOST=2 NETWORK=America One | | | PLAT=Unlimited | ------------------------------------------------------------------------------------------------------------------------------------ | 2 100 4 23 4.00 0.16 - 7.84 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F | | | DAYPART=7PM - 10:59PM LENGTH=60 | | | NCOST=2 NETWORK=America One | | | PLAT=Sprint Sense Anytime XTRA | ------------------------------------------------------------------------------------------------------------------------------------ | 3 107 4 22 3.74 0.14 - 7.33 | CABNET=CABLE TV DAYOWEEK=Weekdays M-F | | | DAYPART=7PM - 10:59PM LENGTH=60 | | | NCOST=2 NETWORK=Fit TV | | | PLAT=Sprint Sense Anytime XTRA | ------------------------------------------------------------------------------------------------------------------------------------
  • 6. Confidential Multivariate Media Model Results I : ( Using 1,187 significant combinations we classified Keepers and Losers unto the 42,210 spots ) Keepers = 17,852 spots = 153,979 calls = $52,784,516 = 342 CPC Losers = 24,358 spots = 78,808 calls = $32,019,299 = 406 CPC Savings Average CPC is 364 ( 84MM / 232,000) = 364 – 342 = 22 * 153,979 = 3,387,538
  • 7. Confidential Multivariate Media Model Characteristics: 0 10 20 30 40 50 60 9AM-3:59PM 4PM-6:59PM 7PM-10:59PM 11PM-1:59AM 2AM-5:59M 6AM-8:59AM Keepers Losers
  • 8. Confidential Multivariate Media Model Characteristics: 0 10 20 30 40 50 60 70 80 15Seconds 30Seconds 60Seconds Keepers Losers
  • 9. Confidential Multivariate Media Model Characteristics: 0 10 20 30 40 50 60 70 80 90 W eekDays W eekEnd Keepers Losers
  • 10. Confidential Multivariate Media Model Characteristics: 0 10 20 30 40 50 60 70 80 90 100 Cable NETW ORK Keepers Losers
  • 11. Confidential Conclusions : There appears to be a preliminary model that can * Score and separate out spots that reduce the CPC by shifting Day part placement, weekday placement. *Shifting 15 second spots to 30 second spots. Achieving a $22.00 CPC savings for 150,000+ calls