Scribd Logo
Upload

Welcome to Scribd!

  • 102 views

    Is Deep Learning Is A Game Changer For Marketing Analytics

    Uploaded by

    T BINDHU BHARGAVI
    This document discusses deep learning and its applications in marketing. It provides an introduction to deep learning, gives examples of use cases like segmentation and media strategy, discusses technical aspects for analysts, and covers implications for companies. It finds that deep learning models can predict consumer choices more accurately than traditional methods, though improvements were modest. The document also discusses costs/benefits, advantages, and applications of deep learning as well as thoughts on the future.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Is Deep Learning Is A Game Changer For Marketing Analytics

    Uploaded by

    T BINDHU BHARGAVI
    102 views23 pages
    This document discusses deep learning and its applications in marketing. It provides an introduction to deep learning, gives examples of use cases like segmentation and media strategy, discusses technical aspects for analysts, and covers implications for companies. It finds that deep learning models can predict consumer choices more accurately than traditional methods, though improvements were modest. The document also discusses costs/benefits, advantages, and applications of deep learning as well as thoughts on the future.

    Original Description:

    #Is Deep Learning is a Game Changer for Marketing Analytics

    Original Title

    Is Deep Learning is a Game Changer for Marketing Analytics

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses deep learning and its applications in marketing. It provides an introduction to deep learning, gives examples of use cases like segmentation and media strategy, discusses technical aspects for analysts, and covers implications for companies. It finds that deep learning models can predict consumer choices more accurately than traditional methods, though improvements were modest. The document also discusses costs/benefits, advantages, and applications of deep learning as well as thoughts on the future.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    102 views23 pages

    Is Deep Learning Is A Game Changer For Marketing Analytics

    Uploaded by

    T BINDHU BHARGAVI
    This document discusses deep learning and its applications in marketing. It provides an introduction to deep learning, gives examples of use cases like segmentation and media strategy, discusses technical aspects for analysts, and covers implications for companies. It finds that deep learning models can predict consumer choices more accurately than traditional methods, though improvements were modest. The document also discusses costs/benefits, advantages, and applications of deep learning as well as thoughts on the future.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    of 23

    • Abstract

    • Definition of Deep Learning


    • Introduction
    • Example
    • Three use cases
    • Technical Note For Analysts

    CONTENTS • Implications for Companies


    • Accuracy rate
    • Costs and Benefits
    • Advantages
    • Applications
    • Future and Conclusion
    • References
     •In the future, deep learning is likely to substantially change both
    marketing strategies and customer behaviors.
    ABSTRACT  •Building from not only extant research but also extensive
    interactions with practice, the authors propose a multidimensional
    framework for understanding the impact.
    DEFINITION OF DEEP
    LEARNING 
    Deep learning is an AI function that mimics the
    workings of the human brain in processing data
    for use in detecting objects, recognizing speech,
    translating languages, and making decisions.
    Deep learning is delivering impressive results in AI
    applications.

    INTRODUCTION The technology that underpins deep learning is becoming


    increasingly capable of analyzing big databases for patterns
    and insights. 

    It isn’t difficult to imagine a day when companies will be able


    to integrate a wide array of databases to discern what
    consumers want with greater sophistication and analytic power
    and then leverage that information for market advantage.
    EXAMPLE
     Segmentation
     Brand optimization
    THREE USE CASES
     Media strategy
      Marketers take on segmentation projects to more
    efficiently target customers and refine and optimize
    marketing strategies.
     AI-driven market research can assist by making it
    simple to build dynamic micro-segmentation models,
    SEGMENTATION
    create archetype segmentation for messaging design
    purposes, develop attitudinal segmentation to
    understand the customer mindset and optimize global
    marketing investment with country clustering.
     AI-driven market research tools can deliver a brand
    measurement framework centered on the customer,
    accessing diverse information sources to provide a
    comprehensive picture of brand health.
     This gives marketers insight about brand equity and
    BRAND the actions they can take to capitalize on it most
    OPTIMIZATION effectively.
     Practical applications can include a brand health
    dashboard and identification of key brand loyalty
    drivers.
     The right AI-driven market research tool can assist
    with media planning and execution by defining
    campaign success and identifying impact.
     It can bring diverse data together, including survey
    information, social media and in-house sales data to
    MEDIA provide recommendations that improve customer
    STRATEGY interactions.
     AI-driven research can also measure ROI from media
    spend and promotions and recommend channels to
    optimize reach.
    The first section is used to “train” the model using
    algorithms that minimize the losses and that measure how
    well the model fits the data. This is manipulating the data
    to describe past choices. To avoid overfitting, analysts can
    configure models so they are less sensitive to noise.
    TECHNICAL The second section of the data is used to determine the
    NOTE FOR model’s overall architecture (the number of hidden
    ANALYSTS layers and the type of nonlinear transformations
    between them).
    The final section is used to test the predictive
    accuracy.
    IMPLICATIONS FOR COMPANIES

    Be Be alert to the future opportunities. 

    Create Create rich databases. 

    Build capability through training


    Build and new hires. 

    Develop Develop plans for experimentation. 


     Based on our analysis, the two models that drew on
    deep learning were able to predict credit card choices
    with more precision than the traditional approach.
    What We Found (See “How Deep Learning Can Outperform
    Traditional Marketing Analytics.”) But the
    improvements were not as large as we thought they
    would be.
     To compare deep learning with traditional methods for
    marketing analytics, we studied a large database of
    click-streams, demographics, and ad exposures
    relating to the credit card market from NerdWallet, a
    EXPERIMENT large online vendor of credit cards, based in San
    Francisco.
     We wanted to see if a multilevel deep learning model
    could predict credit card choices more accurately than
    traditional models.
    The first model was a straightforward linear regression
    of choice as a function of the individual user
    demographics and card attributes.
     Each variable had a simple direct effect on choice
    through one equation, with no interactions among
    variables.
     Estimating sales response to marketing inputs can be a
    relatively straightforward analytics exercise.
     Many organizations use a variety of classic statistical
    methods and tools to simulate the effects of things like
    How Deep Learning price changes and shifts in advertising, promotions,
    Can Out perform and distribution.
    Traditional
    Marketing Analytics  It’s common for brand managers to track their
    progress with simple models and online dashboards.
     In some cases, they use A/B testing: If you have a
    statistical sense of how customers are likely to react to
    a marketing offer, you can look for ways to adjust the
    variables to improve the outcomes.
     The simple regression model had an accuracy rate of
    70.5%, meaning that in 70.5% of the cases, we were
    able to predict correctly which card a particular
    consumer would apply for and which ones he or she
    ACCURACY RATE would not apply for.
     The simple deep learning model was slightly more
    accurate, at 71.7%, and the more sophisticated model
    had an accuracy rating of 73%.
    COSTS AND
    BENEFITS
      Costs to acquire deep learning technology, staff
    required to implement it, and additional data costs that
    may need to be incurred.
     Finally, even with today’s fast computers, deep
    learning models need huge amounts of computer
    power and may experience long run times.
     This becomes a real limitation if real-time
    implementation is required. 
    Advantage is tied to the
    technical criteria used for
    This improves the model of
    estimating success. Traditional
    consumer response and the
    statistical methods use the
    probability that predictions will
    ability to describe historical
    actually occur.
    relationships as the criteria for
    success. 

    ADVANTAGES
    DISADVANTAGES

    • A major drawback of deep learning is that it's difficult to identify which variables drive
    the biggest response.
     Deep learning might also be used to design
    APPLICATIONS products to meet consumers’ personal needs,
    which could then be produced and delivered
    through automated 3D printing systems.
    Improve estimates of market response to help
    Improve
    maximize profit and marketing ROI, 

    THOUGHTS ABOUT
    Reveal new opportunities for product THE FUTURE AND
    Reveal CONCLUSION
    development, and

    Allow for more targeted product design,


    Allow
    distribution, promotion, and media optimization.
      P. Dhillon and S. Aral, “Modelling Dynamic
    User Interests: A Neural Network Approach,”
    working paper, MIT Sloan School of
    REFERENCES Management, Cambridge, Massachusetts,
    2019.
     https://www.kaggle.com

    You might also like