MARKETING MIX MODELING - Nielsen

1y ago
50 Views
1 Downloads
6.50 MB
32 Pages
Last View : 4d ago
Last Download : 6m ago
Upload by : Sutton Moon
Transcription

MARKETINGMIX MODELINGWHAT MARKETING PROFESSIONALSNEED TO KNOWMARCH 2014Copyright 2014 The Nielsen Company1

2MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

CONTENTSFOREWORD . . . 4EXECUTIVE SUMMARY . 6OVERVIEW OF MARKETINGMIX MODELSMarketing Mix Modeling Objectives and Requirements. 8Use of Marginal versus Average MarketingReturn on Investment. 16Gauging the Quality of a Marketing Mix Model. 16CONSIDERATIONS WHENINCORPORATING MARKETING MIXMODEL RESULTS INTO A MEDIA PL ANFINDINGS WHEN INCLUDING HISPANIC MEDIAINTO MARKETING MIX MODELING. 18Impact of Creative on Sales. 18Long-Term Impact of Marketing on Sales. 21Modeled Sales May Not Be Equivalent to Total Sales. 22Brand Size Matters.23One Media’s Marketing Return on Investment DoesNot Dominate Consistently. 24THREE DIRECTIVES FORBRAND MANAGERS .Keep a Lens Focused on Effectiveness.Choose a Future Strategy: Continuity versus Flighting.Knowing A Segment’s Wallet Share May LeadTo Growing Your Brand’s Share.25252528ILLUSTRATIVE EXAMPLE: THE IMPORTANCEOF USING MARGINAL VERSUS AVERAGEMARKETING RETURN ON INVESTMENT INMEDIA PL ANNING . 29BRAND MANAGER CHECKLIST. 31REFERENCES . 32Copyright 2014 The Nielsen Company3

FOREWORDATTENTION BRAND MANAGERS,MULTICULTURAL MANAGERS,MARKETING INTELLIGENCE TEAMSAND PL ANNING AGENCIES:Though you may already use primary and secondary media researchto guide your marketing strategy, you may be missing out on keyinformation if you’re not measuring marketing effectiveness too. Moreinformation is required to answer these compelling questions: How much media is enough?Which medium is most effective?What is the best media environment to use?Is it better to use flighting or continuity?When are ads worn out?Marketing Mix Modeling (MMM), the use of statistical analysis toestimate the past impact and predict the future impact of variousmarketing tactics on sales, can deeply inform marketing plans. Whilemarketing spend and bottom line results are often perceived asdisconnected, Marketing Mix Modeling closes the loop and shows thepath to improved return on marketing ort4MROICLOSING THE SIncrementalVolumeMARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

This paper will give you an understanding of the elements of a MarketingMix Modeling project and how to use its outputs to improve marketingefficiencies and Marketing Return on Investment (MROI). We’llexplore the importance of segmenting your audience to improve youreffectiveness, looking at the Hispanic segment as an example, since theyare one of the fastest growing groups in the U.S.In Section I, you’ll get an overview of the key components and phases ofa Marketing Mix Modeling project. Section II identifies what to considerwhen incorporating your results into planning, with a lens focused onHispanic media. Mastering Marketing Mix Modeling can enhance yourmarketing and advertising decision-making, giving you not only positiveresults, but confidence in your future marketing plans.As the world’s largest marketing mix modeling provider, Nielsen hasan unmatched ability to integrate a diverse set of data sources intostate-of-the-art marketing models to provide globally relevant andconsistent marketing mix recommendations. With local presence inmore than 100 countries, Nielsen delivers an end-to-end solutionusing integrated insights, global comparability and proven predictivesimulation and optimization tools. Nielsen is a founding member ofthe Digital Media Consortium with Google, Facebook and a group ofleading advertisers, focusing on identifying groundbreaking trendsand insights on measuring the return of digital marketing initiatives.Copyright 2014 The Nielsen Company5

EXECUTIVESUMMARYMarketing Mix Modeling (MMM) is an important part of anymarketing plan. It allows you to measure past performance and charta path for future success. To ensure a successful Marketing MixModel project, every project must begin with a checklist of businessquestions, which will keep you focused on your goals and make sureyour project is answering the right questions. You can see a samplechecklist at the end of this paper.The four phases of a Marketing Mix Modeling project are:1. D ata collection and integrity: Collaborate with your Marketing MixModeling vendor to decide which data needs to be included.2. M odeling: Test the models against your checklist. Ensure yourin-house analytics team is involved.3. M odel-based business measures: Interpret the model-based outputsand look at your campaign’s effectiveness, efficiency and MarketingReturn on Investment. Measure incrementality by campaign for alltactics so you can better understand drivers of incremental profit.4. Optimization and simulation: Determine the best marketing mixfor your next planning period.6MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

Your Marketing Return on Investment (MROI) will be a key metricto look at during your Marketing Mix Modeling project, whether thatbe Marginal Marketing Return on Investment for future planning orAverage Marketing Return on Investment for past interpretation. Thebest projects also gauge the quality of their marketing mix model,using Mean Absolute Percent Error (MAPE) and R2.Once you feel comfortable with the way a Marketing Mix Modelingproject should work, it’s time to take it to the next level: applying whatyou’ve learned to specific audiences. We’ve chosen to focus on oneof the fastest growing groups in the U.S., the Hispanic population.Looking at data from previous Nielsen studies, we note these fivekey findings when including Hispanic media into your Marketing MixModeling project:1. A d creative is very important to your sales top line and yourMROI, especially if you can tailor it to a segmented audience.This paper presents five best Spanish language creative practicesto drive MROI, which should also impact top-of-the-funnelmarketing measures.2. The long-term impact of marketing on sales is hard to nail down,but we have found that ads that don’t generate sales lift in the nearterm usually don’t in the long-term either. You can also expect longterm Marketing Return on Investment to be about 1.5 to 2.5 timesthe near-term Marketing Return on Investment.3. M odeled sales may not be equivalent to total sales. Understand howmarketing to targeted segments will be modeled.4. Brand size matters. As most brand managers know firsthand,the economics of advertisement favors large brands over smallbrands. The same brand TV expenditure and TV lift produces largerincremental margin dollars, and thus larger Marketing Return onInvestment, for the large brand than the small brand.5. One media’s Marketing Return on Investment does not dominateconsistently. Since flighting, media weight, targeted audience,timing, copy and geographic execution vary by media for a brand,each media’s Marketing Return on Investment can also varysignificantly.Knowing these key findings, you can follow the best practices fora Marketing Mix Modeling project, such as focusing on campaigneffectiveness and deciding between continuity and flighting. It alsohelps to get to know your category’s wallet share, the effect yourbrand size may have on your results and the differences betweenvarious forms of media. Once you understand these best practices,you’ll be well on your way to executing a successful Marketing MixModeling project.Copyright 2014 The Nielsen Company7

OVERVIEW OFMARKETINGMIX MODELSMarketing Mix Modeling (MMM) is the use of statistical analysis toestimate the past impact and predict the future impact of variousmarketing tactics on sales. Your Marketing Mix Modeling projectneeds to have goals, just like your marketing campaigns. As a brandmanager, you are responsible for setting those goals and seeingthem through. Before you begin working with a modeling vendor,make a checklist of questions for your vendor to address, like theone at the end of this paper. Think of your checklist as a roadmap tosuccess—you’ll never get anywhere if you don’t know where you’regoing, so don’t skip this step.MARKETING MIX MODELING OBJECTIVEAND REQUIREMENTSThe main goal of any Marketing Mix Modeling project is to measurepast marketing performance so you can use it to improve futureMarketing Return on Investment (MROI). The insights you gainfrom your project can help you reallocate your marketing budgetacross your tactics, products, segments, time and markets for abetter future return. All of the marketing tactics you use should beincluded in your project, assuming there is high-quality data withsufficient time, product, demographic, and/or market variability.Each project has four distinct phases, starting with data collectionand ending with optimization of future strategies. Let’s take a lookat each phase in depth:Phase 1Data Collection& IntegrityPhase 2ModelingPhase 3Model-BasedBusiness MeasuresPhase 4Optimization &SimulationPHASE 1: DATA COLLECTION AND INTEGRITYTo kick off your project, do your due diligence and collect the data thatwill be used in the statistical model. Determine which products will beanalyzed, the timeframe you’ll look at, the time-dimension granularity,and which markets to model. Finally, determine the sales performancemeasure to be analyzed – dollar sales, volume, units or somethingelse?i You’ll also need to gather brand margin rates and marketingtactic spend, which are needed to calculate Marketing Return onInvestment down the road.18Regardless of the actual sales performance measure modeled, we will refer to it as “sales.”MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

HIGHLIGHTTHE MAIN GOAL OFANY MARKETING MIXMODELING PROJECTIS TO MEASUREPAST MARKETINGPERFORMANCE SOYOU CAN USE IT TOIMPROVE FUTUREMARKETING RETURN ONINVESTMENT (MROI).Copyright 2014 The Nielsen Company9

Partner with your Marketing Mix Modeling vendor to decide whichtactics to include in the model. You can use our list, below, for ideas.SAMPLE MARKETING MIX MODEL SALES TACTICDISTRIBUTION &PRICINGPRODUCTPAID MEDIAPROMOTIONSEXTERNALFACTORSDistributionProduct Life CycleTVMerchandisingSeasonality &Weather PatternsPricingProduct ChangesMagazinesCouponingCompetitive FactorsCRM & OffersNew ProductsNewspaperPublic RelationsMacroeconomic InputsChannel IncentivesSegment TrendsRadioLoyalty Program ActivityRetail Format ChangesProduct RecallsOutdoor/Out-of-HomeEvent MarketingQuality MetricsOnline MediaSponsorshipsAwardsWord-of-MouthThird-Party ReviewsSamplingInventory LevelsSales Force ActivityCustomer SatisfactionProduct PerformanceProduct PlacementIn Phase 1, your team will answer questions like: How can you makesure that your chosen data is consistent over its entire life cycle? Andare you using the best available data for your project? Phase 1 alsoincludes verifying data integrity, which requires coordination amongall project stakeholders. To ensure that your Marketing Mix Modelingproject meets expectations, key stakeholders must participate in aData Review session before you move to Phase 2 (Modeling). Thisreview will ensure proper handling of the data by the vendor, who willeventually be processing and synthesizing large amounts of disparatedata for you.PHASE 2: MODELINGThe statistical method used is usually determined by the vendorafter collaborating with the advertiser to ensure the modeladdresses your questions. Brand managers should partner withtheir in-house analytics team during this phase of the project. It’simportant for the in-house analytics team to dive deeply into thestatistical details and specifications. Any concerns the in-houseteam has with these details should be raised with you immediately.10MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

PHASE 3: MODEL-BASED BUSINESS MEASURESThe outputs from your Marketing Mix Modeling project – that is, thedata that comes out of your statistical model – needs to align withyour checklist and address the questions you listed. Your projectwill produce a host of outputs that measure how each tactic affectssales, and before sharing results with a wide audience you shouldreview your vendor’s proposed outputs to make sure they supportyour goals.A fundamental output of a Marketing Mix Modeling project is thedecomposition of sales, often represented by a pie chart, showingsales volume broken down by each modeled tactic. This outputdifferentiates core and incremental marketing tactics – the coreincludes all marketing tactics not controlled by the marketing/trade team (for example, distribution, weather, seasonality,competitive trade, competitive advertising and more). You can alsothink of it as the sales that would be generated in the absence ofany marketing efforts. Incremental tactics are just the opposite,those controlled by the marketing/trade team. Your project shouldmeasure incrementality by campaign for all media-specific tacticsexecuted. This way, you’ll know whether existing campaigns shouldbe continued, and if so, to what extent.The pie chart below shows the percentage of total sales attributedto each marketing tactic in Year 2 and Year 3 2.2.8%.6%.2%5.6%19.4%CORE SALESYEAR 2 SALESDECOMPOSITIONTRADETVOTHER MARKETING.8%.3%20.1%CORE SALESYEAR 3 SALESDECOMPOSITIONTVOTHER MARKETINGCOUPONSCOUPONS76.9%TRADE73.2%The charts in this section are illustrative and should not be considered to be the output of an end-to-end marketingmix modeling project.2Copyright 2014 The Nielsen Company11

Another fundamental business output of a Marketing Mix Modelingproject is the year-over-year impact of each tactic on sales. Thischart usually includes comments about the changes in marketingsupport that caused changes in incremental sales.YEAR 3 VS YEAR 2, PERCENTAGECHANGE IN TACTIC .22%DISTRIBUTION.59%.23%OTHER MEDIACOMPETITIVE MEDIA.13%.05%COUPONS-.16%SEASONALITY AND WEATHER-.25%CONSUMER MARKETINGHOLIDAYS-.30%COMPETITIVE TRADEBASE PRICE-1.32%-3.19%.09%TOTALCOMMENTSTV: TV spend increased by 50% and GRPs by 55%Trade: Trade support increased by 10%Distribution: Average # items per store sellingincreased by 6%Other Media: Radio spend increased by 30%Competitive Media: Though competitors increasedTV spend by 20%, the year-over-year affect isrelatively constantConsumer Marketing: In store promotionsdecreased by 4%Price: Base price increase by 1.9% on average12MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

There are three important metrics to look at after performinga decomposition of sales. The first is Effectiveness. If you divide the incremental sales (thosethat came as a result of marketing efforts) by the support (theexecution of each marketing effort, such as Target Rating Pointsfor media) for each tactic, you get near-term Effectiveness.3The second metric is near-term Efficiency, which you can findby dividing incremental sales by expenditures (typically workingspend for media) for each tactic.Lastly, dividing a marketing tactic’s incremental margindollars (the gross profit) by its spend yields Marketing Returnon . Sales/TRP)200 unitsTV AdvertisingOUTPUT3Inc. Profit 150,000Incremental Sales100,000 UnitsEfficiency(Inc. Sales/Cost).50 unitSupport500 TRPsMROI(Inc. Profit/Cost) .75Near-TermMetricsCost 200,000Near-term is usually 5 to 8 months after the first media exposure.Copyright 2014 The Nielsen Company13

Brand 1 MROI by Media 3.50 3.20 3.00 2.74 2.62 2.50 2.00 1.50 1.15 1.00 1 Break Even .90 0.50 0.00OverallTVDigitalPrintRadioEach media type in the graph shown above is made up of a number ofdistinct campaigns. The Marketing Return on Investment for each ofthese campaigns by media, or across media, are also available sinceMarketing Mix Modeling estimates the impact of each campaign bymedia. The chart below shows the Marketing Return on Investmentfor each of the 11 TV campaigns executed for Brand 1. These insightswill be key in helping you address business challenges and improveMarketing Return on Investment.Brand 1 MROI by TV Campaign 5.00 4.50 4.00 4.44 3.97 3.63 3.50 3.15 3.00 2.95 2.50 2.68 2.53 2.00 1.98 1.97 1.93 1.80 1.50 1 Break Even 1.00 0.50 0.00#114#2#3#4#5#6#7#8#9#10#11MARKETING MIX MODELING: WHAT MARKETING PROFESSIONALS NEED TO KNOW

PHASE 4: OPTIMIZATION AND SIMULATIONThe final phase of a Marketing Mix Modeling project essentiallyturns your outputs into inputs for future marketing planning. Afterthe completion of modeling, you can perform an optimization/simulation exercise, which provides insights to use when planningfuture marketing campaigns. These exercises simulate the effectthat varying each marketing tactic might have on future sales (alsoknown as simulation, or “What-if Analysis”), and determine thebest combination of tactics for reaching your goals (also calledoptimization). In other words, a mathematical model and varioususer-supplied constraints can provide insights to make the way youapproach marketing even better in the future.THE FINALPHASE OF AMARKETINGMIX MODELINGPROJECTESSENTIALLYTURNS YOUROUTPUTSINTO INPUTSFOR FUTUREMARKETINGPL ANNING.Copyright 2014 The Nielsen Company15

USE OF MARGINAL VS. AVERAGE MARKETINGRETURN ON INVESTMENTSo far we’ve been using the term “Marketing Return on Investment”to mean Average Marketing Return on Investment. But Marketing MixModeling can also provide insights into Marginal Marketing Returnon Investment, and it’s important to know the difference betweenthe two. Marginal Marketing Return on Investment measures thefinancial impacts of the next 1 in spend. While Average MarketingReturn on Investment can help you gauge past performance, MarginalMarketing Return on Investment is what you should use for futuremedia planning, such as in Optimization and “What-if ” exercises. Seeillustrative example on page 29.TABLEGAUGING THE QUALIT Y OF ASECTIONHEADERMARKETING MIX MODELBefore you make changes to your marketing plan based on the outputfrom your Marketing Mix Modeling project, it’s a good idea to evaluatethe quality of the model.Though brand managers are not necessarily modelers, you shouldhave a basic understanding of how to gauge the quality of a MarketingMix Model. For instance, models that do not accurately predict salesshould be a cause of concern. It’s a good idea to include a section onmodel quality in your checklist.You can use many different diagnostics to measure the quality of amodel. Some diagnostics can also be used to determine if you havedata quality issues. Two popular measures that determine how wellthe model will predict sales are Mean Absolute Percent Error (MAPE)and R2.How to use MAPE. This diagnostic validates the quality of a MarketingMix Model by comparing the MAPE between a training sample and aholdout sample. The training sample is used to build the model, andthe holdout sample is used to validate the model.4 A smaller MAPEfor both samples indicates a better model fit. A model that fits thedata perfectly has a MAPE 0%. No model should be expected to fitthe data perfectly. Moreover, models that fit historic data perfectlyusually have poor future performance. As a guideline, the differencebetween the training sample’s MAPE and the holdout sample’s MAPEshould be les

Marketing Mix Modeling (MMM) is an important part of any marketing plan. It allows you to measure past performance and chart a path for future success. To ensure a successful Marketing Mix Model project, every project must begin with a checklist of business

Related Documents:

Figure 1. Model of the Customer Market offering dimensions of the Marketing Mix (Lipson, et al.) Marketing mix development for target market process involves four important steps: 1. Division of the marketing mix into four component-mixes: the product mix, the terms of sale mix, distribution mix and communications mix. 2.

Apr 20, 2021 · Marketing: The activity, set of institutions, and processes for creating, communicating, delivering and exchanging offerings that have value for customers, clients, partners, and society at large. (Marketing Management 15e, Kotler and Keller, 2016) Marketing Management is the art and science of choosing target markets and building profitable .File Size: 720KBPage Count: 30Explore further(PDF) Marketing Mix of 4P'S for Competitive Advantage .www.academia.eduMarketing Mix of 4P’S for Competitive Advantageiosrjournals.org(PDF) The Evaluation of Marketing Mix Elements: A Case Studywww.researchgate.netMARKETING MIX THEORETICAL ASPECTSgranthaalayah.comTHE 4 P’S OF MARKETING MIXwww.angle180.comRecommended to you b

The four phases of a Marketing Mix Modeling project are: 1. Data collection and integrity: Collaborate with your Marketing Mix Modeling vendor to decide which data needs to be included. 2. Modeling: Test the models against your checklist. Ensure your in-house analytics team is involved. 3. Model-based business measures: Interpret the model .

9. attlefield – Karaoke Mix Jordin Sparks 10. What About Now – Karaoke Mix Daughtry 11. Waking Up In Vegas – Karaoke Mix Katy Perry 12. Let It Rock – Karaoke Mix Kevin Rudolph ft. Lil Wayne 13. Halo – Karaoke Mix Beyonce 14. Shake It – Karaoke Mix Metro Station 15. etter In Time – Karaoke Mix Leona Lewis 16. Light On – Karaoke .

Marketing Strategy of the supermarkets 6 II. Theory The research uses the theory of “marketing mix” and merchandising’s theory. 1. Marketing Mix According to Doyle (2002) 5, the “marketing mix” is the central task of marketing professionals. The set of

1. Understand what a marketing manager does. 2. Know what marketing strategy planning is—and why it is the focus of this book. 3. Understand target marketing. 4. Be familiar with the four Ps in a marketing mix. 5. Know the difference between a marketing strategy, a marketing plan, and a marketing program. 2–2

Nielsen MultiSource Aggregation and Distributional Alignment (MADA) process. MADA is a proprietary methodology for assessing national distributions, which begins with data from the annual Nielsen demographic update and is informed by additional data from the Nielsen

Marketing mix modeling uses advanced statistics (typically regression analysis) to quantify the impact of each marketing element on KPIs such as incremental sales volume, revenue, new customers or patient admissions. Results can be compared with spending to determine ROI, and future marketing spend can be used with model results to develop .

factors which are categorised under the 4 P‟s to decide marketing mix for a product. Keywords: Marketing Mix, Product, Price, Positioning, Promotion, Competitive Advantage I. INTRODUCTION Marketing mix is the comb

Journal of Management and Marketing Research Brand equity, marketing strategy, Page 4 marketing strategy (marketing mix elements), or marketing efforts as antecedents of brand equity, and operationalized the marketing mix as: (1) price, (2) store image, (3) distribution

UNIT: - I BASIC CONCEPTS IN MARKETING MANAGEMENT STRUCTURE 1.0 Introduction to Marketing 1.1 Definition of Marketing 1.2 Evolution of Marketing 1.3 Marketing Concept 1.4 Role of Marketing 1.5 Strategic Marketing Planning 1.6 Scope of Marketing 1.7 Approaches of Marketing 1.8

May 05, 2011 · 3022 Broadway . Uris Hall, Room 604 . New York, NY 10027 . dn75@columbia.edu . May 5, 2011 . Abstract . We review accounting principles related to the reporting of marketing activities and evaluate their implications for marketing research and practice. Based on our review, we argue thatFile Size: 393KBPage Count: 50Explore further(PDF) Strategic Marketing and Marketing Strategy: Domain .www.researchgate.net(PDF) Marketing Management - ResearchGatewww.researchgate.net5 Marketing Management Orientationscommercemates.com5 Marketing Concepts: Marketing Management Philosophieswww.iedunote.comBasic Marketing Principles - Mercer Universityfaculty.mercer.eduRecommended to you b

Marketing Mix Modeling . A brand's volume sales constantly changes & it changes for a reason Marketing Mix Modelling is a mathematical approach that explains how each factor drivers sales & share Year 2007 Year 2010 What is Marketing Mix Modelling? Sales (Brand metrics) Print, OOH

marketing management is the concept of the marketing mix (Figure 1). The marketing mix is not a theory of management that has been derived from scientific analysis, but a conceptual framework which highlights the principal decisions that marketing manager’s make in configuring their offerings to suit customers’ needs. .

Know what marketing strategy planning is— and why it will be the focus of the book. 3. Understand target marketing. 4. Be familiar with the four Ps in a marketing mix. 5. Know the difference between a marketing strategy, a marketing plan, and a marketing program. 6. Understand what customer equity is and why marketing strategy planners seek to

3 CLA MixDown / User Guide Introduction CLA MixDown glues a mix together and brings life to it. Insert it on your mix buss and mix through the plugin or open it on a mix in progress to add depth and impact. Four faders provide control over the color, dynamics, and drive of the mix.

formulate a winning Marketing Mix, this Marketing Management course is about how to set that Marketing Mix in motion. Application of the appropriate marketing tools for each situation and putting Marketing to work towards the g

International Marketing (3) Sales Management (3) Marketing of Financial B2B Marketing (3) Marketing Strategy (1.5) Media Planning ( 1.5) (3)Brand Management (3) Retail Marketing ((3) Marketing Engineering (1.5) Rural Marketing (3) Innovations in Marketing (3) Customer Relationship Management (3) Sales Promotion (3) Marketing for Entrepreneurs

14 D Unit 5.1 Geometric Relationships - Forms and Shapes 15 C Unit 6.4 Modeling - Mathematical 16 B Unit 6.5 Modeling - Computer 17 A Unit 6.1 Modeling - Conceptual 18 D Unit 6.5 Modeling - Computer 19 C Unit 6.5 Modeling - Computer 20 B Unit 6.1 Modeling - Conceptual 21 D Unit 6.3 Modeling - Physical 22 A Unit 6.5 Modeling - Computer

monitors, and flexible seating to accommodate small group, large group, and individual work. The classroom has a maximum capacity of 36 students. Figure 1 shows the classroom before and after redesign, and Figure 2 shows three views of the new ALC. Participants Faculty and students who had taught or taken at least one