Hariharan Panda 24EMBA1613

Marketing Analytics

Question 1
An e-commerce platform experimented with three types of discount strategies (flat
discount, buy-one-get-one, and coupon-based) across multiple regions. They want to
assess which strategy led to the highest conversion rate. Explain how One-Way and N-
Way ANOVA can be used to analyze performance and what strategic insights can guide
future marketing campaigns.

Using One-Way and N-Way ANOVA to Evaluate Discount Strategy Performance

An e-commerce platform tested three discount strategies—Flat Discount, Buy-One-Get-

One (BOGO), and Coupon-Based Discounts—across different regions. Their objective is to
identify which strategy produced the highest conversion rate and under what conditions.
Statistical tools such as One-Way ANOVA and N-Way ANOVA help systematically
evaluate performance differences.

1. Application of One-Way ANOVA

Purpose

One-Way ANOVA helps determine whether conversion rates differ significantly across
the three discount strategies, independent of any other factors.

How it works

 Groups = Discount Types

(Flat Discount, BOGO, Coupon)
 Dependent variable = Conversion Rate
 ANOVA tests:
Is the difference in mean conversion rates across strategies statistically significant?

Example Interpretation

 If p < 0.05, it indicates that at least one discount strategy’s conversion rate is
different from the others.
 Post-hoc tests (Tukey) identify which strategy outperforms the rest.

Strategic Insight from One-Way ANOVA

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Helps decide the best universal discount strategy if the company wants a single
nationwide campaign.
 If Flat Discount significantly outperforms others:
→ Company may prioritize simple, upfront pricing strategies.

If Coupon-based performs worst:

→ Reduce or redesign coupon campaigns to avoid friction.

2. Application of N-Way ANOVA (Two-Way or Multi-Factor ANOVA)

Purpose

N-Way ANOVA helps analyze not only the main effect of discount strategy but also how
performance varies together with other factors, such as:

 Region
 Customer segment (new vs repeat)
 Device used (mobile vs desktop)
 Season/time period

This allows evaluation of:

 Main effects (impact of each factor independently)

 Interaction effects (e.g., discount strategy × region)

How it works

 Factors = Discount Type + Region (and/or others)

 Dependent variable = Conversion Rate
 ANOVA tests whether certain discount strategies work better in specific regions or
customer groups.

Example Interpretation

 If interaction between Strategy × Region is significant:

→ Some regions respond better to BOGO while others prefer Flat Discounts.

Strategic Insight from N-Way ANOVA

This helps in localized or personalized campaigns, such as:

 Region-A: BOGO works best → Use BOGO in marketing pushes here

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Region-B: Flat Discount has highest conversion → Prioritize Flat Discounts

 New customers responding well to Coupons → Use them in onboarding campaigns
 Mobile customers preferring Flat Discounts → Push mobile-only flash discounts

N-Way ANOVA allows precision marketing instead of “one-size-fits-all.”

3. Strategic Insights to Guide Future Marketing Campaigns

A. Move Toward Data-Driven Personalization

 Use discount strategies that statistically outperform others in specific regions or

customer segments.
 Avoid strategies that consistently underperform (e.g., complicated coupons).

B. Optimize Budget Allocation

 Allocate higher discounts to markets where the ROI is strongest.

 Reduce spending on ineffective discount mechanisms.

C. Improve Customer Experience

 If coupon strategies have low conversion → indicates friction; simplify redemption

processes.
 If BOGO works best for bulk buyers → Target high-value or high-frequency
customers.

D. Seasonal and Regional Tailoring

 If N-Way ANOVA reveals seasonal interactions:

→ Use BOGO during festive seasons and Flat Discounts during off-peak periods.

E. Strategic Testing Cycle

 Use ANOVA results as input for A/B testing, followed by further segmentation
analysis.

Conclusion

 One-Way ANOVA helps determine which discount strategy performs best overall.

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 N-Way ANOVA reveals context-dependent performance, such as regional,

demographic, or device-based differences.
 Together, they guide precision discount strategies, better budget allocation, and
improved conversion rates by tailoring campaigns to customer behavior patterns.

Question 2
A telecom company is designing data plans for various user types: heavy users,
moderate users, and light users. Explain how linear and non-linear pricing models can
be implemented. Discuss the advantages and drawbacks of each approach in balancing
profitability with customer satisfaction.

Telecom Pricing Models: Linear vs. Non-Linear Pricing

A telecom company wants to design differentiated data plans for heavy, moderate, and light
users. Pricing design can follow linear or non-linear models, each offering different trade-
offs between profit maximization and customer satisfaction.

1. Linear Pricing Model

Definition

Linear pricing charges customers a constant price per unit of data used.
Example: ₹10 per GB for all users, regardless of usage level.

Implementation for Telecom Plans

 Single unit price for data (e.g., ₹10/GB).

 Customers pay based on exact usage.
 Bills fluctuate:
o Light user (5 GB): 5 × ₹10 = ₹50
o Heavy user (50 GB): 50 × ₹10 = ₹500

Advantages

1. Simplicity and Transparency

 Easy for customers to understand.

 Builds trust—no hidden charges or tiers.

2. Usage-Based Fairness

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Customers pay only for what they consume.

 Low-usage customers remain satisfied due to lower bills.

3. Predictable Marginal Revenue

 Every GB generates a guaranteed, constant revenue stream.

Drawbacks

1. Price Sensitivity and Bill Shock

 Heavy users face high monthly charges, causing dissatisfaction and churn.
 Fluctuating bills reduce long-term customer commitment.

2. Limited Price Discrimination

 Telecom cannot fully monetize heavy users or subsidize light users.

 Less flexibility to offer competitive bundles.

3. Revenue Instability

 Revenue varies widely based on consumer usage patterns.

2. Non-Linear Pricing Model

Definition

Non-linear pricing charges different prices per unit at different usage levels, typically
through bundles, tiers, or two-part tariffs.

Common telecom implementations:

 Tiered plans (e.g., 10 GB, 50 GB, Unlimited)

 Two-part tariff (fixed monthly fee + variable per-GB charge)
 Declining block pricing (price per GB decreases as usage increases)
 Bundle pricing (flat fee for fixed GB)

Implementation for User Types

Light Users (1–10 GB)

 Small, low-cost plan (e.g., ₹199 for 10 GB)

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Higher per-GB cost beyond the cap (e.g., ₹20/GB)

Moderate Users (10–50 GB)

 Mid-tier plan (e.g., ₹399 for 40 GB)

Heavy Users (50+ GB)

 Unlimited or large-volume plan (e.g., ₹699 unlimited or 200 GB)

Advantages

1. Effective Price Discrimination

 Extracts higher revenue from heavy users (who are willing to pay more).
 Attracts light users with affordable small plans.

2. Revenue Stability

 Fixed monthly fees ensure predictable cash flows.

3. Customer Satisfaction Through Choice

 Users select plans that match their consumption profile.

 Unlimited plans reduce "bill shock."

4. Encourages Data Usage

 Bundles and declining prices encourage greater data consumption, boosting

engagement.

Drawbacks

1. Complexity

 Too many plan options can confuse customers.

 Difficult for customers to estimate whether they will exceed their caps.

2. Hidden Charges and Fair Use Policies

 Extra charges beyond data caps may cause dissatisfaction.

 Unlimited plans may throttle speeds after fair-use limits.

3. Potential Profit Loss

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 If heavy users choose unlimited plans, telecom may face:

o Network congestion
o Lower per-unit revenue

4. Higher Administrative Costs

 Managing and monitoring multiple plan tiers increases operational cost.

3. Balancing Profitability and Customer Satisfaction

A successful telecom strategy often uses hybrid non-linear pricing, such as:

A. Three-Tier Structure

 Light user plan: Low price, limited GB

 Moderate plan: Balanced price and volume
 Heavy plan: Unlimited or large GB at higher price

This satisfies diverse needs while maximizing price discrimination.

B. Add-Ons

 Purchase extra data at moderate incremental pricing.

 Reduces dissatisfaction among users who exceed caps.

C. Loyalty Benefits

 Data rollover for moderate users

 Discounts for heavy, long-term customers

These maintain customer satisfaction while preserving profitability.

Conclusion

Model Strengths Weaknesses Best For

Linear No price discrimination; Homogeneous customer

Simple, fair, transparent
Pricing heavy-user churn base, small markets

Non-Linear Maximizes revenue; Complex; risk of Large, diverse markets

Pricing offers choice; stable cash dissatisfaction if poorly

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Model Strengths Weaknesses Best For

flow designed with varied user needs

A telecom company should adopt non-linear, tiered plans with options for add-ons and
throttled unlimited offerings to balance maximum profitability with high customer
satisfaction across all user types.

Question 3: Netflix wants to shift its focus from short-term subscriber growth to long-
term customer value management. Calculate and interpret the Customer Lifetime Value
(CLV) for high-value subscriber segments (e.g., premium plan users or long-term loyal
viewers). How can this metric guide Netflix’s strategies for user acquisition, retention,
and personalized content investment?

Customer Lifetime Value (CLV) for Netflix: Calculation, Interpretation & Strategic
Application

Netflix wants to shift from short-term subscriber growth to long-term customer value
(LTV) management.
CLV is a crucial metric enabling Netflix to understand which subscriber segments generate
the most economic value over time and how resources should be allocated for acquisition,
retention, and personalized content strategy.

1. CLV: Formula and Framework

A standard subscription-based CLV formula:

CLV=ARPU×Gross MarginMonthly Churn Rate\text{CLV} = \frac{\text{ARPU} \times \

text{Gross Margin}}{\text{Monthly Churn
Rate}}CLV=Monthly Churn RateARPU×Gross Margin

Where:

 ARPU = Average revenue per user per month

 Gross Margin = Profit margin after subtracting content/serving costs
 Churn rate = % of users leaving per month

(For strategic clarity, numbers below are illustrative, not actual Netflix internal data.)

2. CLV Calculation for High-Value Segments

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Segment A: Premium Plan Users (e.g., 4K, multi-screen)

Assumptions (hypothetical but realistic):

 Monthly ARPU = $20

 Gross margin = 60%
 Monthly churn = 2% (0.02)

CLVpremium=20×0.600.02=120.02=$600CLV_{premium} = \frac{20 \times 0.60}{0.02}

= \frac{12}{0.02} = \$600CLVpremium=0.0220×0.60=0.0212=$600

Interpretation

A premium user is worth $600 over their lifetime, meaning Netflix can spend up to $600 on
acquisition and retention efforts without harming profitability (though in practice it would
spend a fraction of this).

Segment B: Long-Term Loyal Subscribers (3+ years)

Assumptions:

 Monthly ARPU = $15

 Gross margin = 60%
 Monthly churn = 1% (0.01)

CLVloyal=15×0.600.01=90.01=$900CLV_{loyal} = \frac{15 \times 0.60}{0.01} = \frac{9}

{0.01} = \$900CLVloyal=0.0115×0.60=0.019=$900

Interpretation

Long-term loyal users provide the highest CLV. Their value comes from low churn, not
high ARPU.
Netflix should invest heavily in maintaining this stability.

Segment C: Price-Sensitive Basic-Plan Users

Assumptions:

 Monthly ARPU = $8
 Margin = 60%

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Monthly churn = 4% (0.04)

CLVbasic=8×0.600.04=4.80.04=$120CLV_{basic} = \frac{8 \times 0.60}{0.04} = \

frac{4.8}{0.04} = \$120CLVbasic=0.048×0.60=0.044.8=$120

Interpretation

Basic users hold much lower lifetime value; retention costs must be proportionally lower.

3. Strategic Interpretation of CLV Insights

A CLV-based perspective helps Netflix shift from volume-driven growth to value-driven

growth.

A. User Acquisition Strategy Guided by CLV

1. Prioritize high-value segments

Since:

 Premium user CLV = $600

 Loyal user CLV = $900

Netflix should:

 Target high-CLV households, families, 4K device owners

 Use predictive models to identify viewers likely to stay longer
 Run premium-focused onboarding campaigns

2. Reduce acquisition spending on low-CLV segments

 Focus on low-cost marketing for basic plan users

 Test ad-supported tiers to raise ARPU cost-effectively

B. Retention Strategy Informed by CLV

1. Allocate highest retention budgets to low-churn, high-loyalty users

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Examples:

 Special renewal offers for multi-year subscribers

 Exclusive previews or beta content
 Premium-only features (mobile downloads, multi-language support)

2. Reduce churn through personalized engagement

CLV increases significantly when churn drops even by 1%.

Retention tactics:

 Predictive churn analytics

 Personalized “continue watching” nudges
 Smart push recommendations based on viewing history

3. Strategies for premium users

 Priority content releases in 4K

 Early access to global launches
 Premium UI/UX improvements

High-value viewers should feel “valued and differentiated.”

C. Content Investment Strategy Based on CLV

CLV reveals which segments justify investment in certain content categories.

1. High-CLV users guide major content investments

If loyal users mainly consume:

 Crime thrillers
 Big-budget dramas
 International originals

→ Netflix allocates larger budgets to these genres because they retain high-value users.

2. Segment-based content personalization

 Premium users often watch visual-rich formats (4K HDR films, documentaries)
 Low-CLV users may prefer regional, mobile-friendly, small-budget content

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Netflix can map CLV × Genre to optimize ROI.

3. Reduce spending on segments with low CLV elasticity

Basic users contribute less value → content spend should be optimized to cost.

4. High-Level Strategic Recommendations

1. Build CLV-Based Segmentation

Cluster users by:

 Tenure
 ARPU
 Engagement score
 Churn risk

2. Adopt a “Retention-First” Strategy

 Convert short-term binge-watching spikes into steady engagement

 Release content episodically for long-term retention

3. Personalize everything

 Homepages
 Recommendations
 Notifications
 Offers

4. Launch upselling pathways

 Encourage basic users to upgrade to premium

→ Even a 10% upgrade can significantly raise total CLV.

5. Improve onboarding for new customers

Higher early engagement → lower early churn → higher CLV.

Conclusion

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

CLV shows that not all Netflix users are equally valuable, and strategies must be carefully
targeted.

Segment CLV Strategic Focus

Loyal users $900 Retain, reward, protect

Premium users $600 Upsell, differentiate experience

Basic users $120 Low-cost acquisition and retention

By using CLV, Netflix can:

 Shift away from subscriber-count obsession

 Invest in high-return customer segments
 Reduce churn through personalization
 Optimize content production and acquisition

Overall, CLV enables Netflix to build a profitable, loyal, and deeply engaged global
subscriber base.

Question 4: A D2C (Direct-to-Consumer) brand ran a paid ad campaign across

Instagram, YouTube, and Google. They now need to evaluate its effectiveness. Use
digital advertising metrics like CTR, CPA, CPM, ROAS, and Brand Lift to evaluate the
campaign. Recommend optimization strategies based on your analysis.

Evaluating a D2C Brand’s Paid Ad Campaign Using Digital Advertising Metrics

A D2C brand ran ads across Instagram, YouTube, and Google. To measure effectiveness,
the brand must assess reach, engagement, efficiency, and revenue outcomes using
standardized digital marketing metrics.

1. Key Digital Advertising Metrics

1.1 Click-Through Rate (CTR)

Formula:

CTR=ClicksImpressions×100CTR = \frac{\text{Clicks}}{\text{Impressions}} \times

100CTR=ImpressionsClicks×100

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Purpose: Measures ad relevance and whether the message resonated with the target
audience.

Interpretation:

 High CTR (e.g., >2.5% on Google Search) → Strong ad copy, relevant audience
targeting
 Low CTR → Poor creative, weak call-to-action, irrelevant targeting

Platform insights:

 Instagram: High CTR indicates compelling visuals.

 YouTube: CTR depends heavily on thumbnail + first 5 seconds.
 Google Search: CTR reflects match between search intent and ad copy.

1.2 Cost Per Acquisition (CPA)

Formula:

CPA=Total SpendConversionsCPA = \frac{\text{Total Spend}}{\

text{Conversions}}CPA=ConversionsTotal Spend

Purpose: Shows how much the brand pays to acquire one customer.

Interpretation:

 Low CPA → campaign efficient

 High CPA → poor targeting or weak landing page conversion

Platform tendencies:

 Google Search often has low CPA due to high intent

 YouTube may have higher CPA as it's awareness-driven

1.3 Cost Per Mille (CPM)

Formula:

CPM=SpendImpressions×1000CPM = \frac{\text{Spend}}{\text{Impressions}} \times

1000CPM=ImpressionsSpend×1000

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Purpose: Measures cost of reaching 1000 users; ideal for awareness analysis.

Interpretation:

 Low CPM → efficient reach

 High CPM → costly inventory, hyper-targeting, competitive season

Platform benchmarking:

 YouTube: Lowest CPM (broad awareness)

 Instagram: Moderately high CPM (premium visual inventory)
 Google Search: Not CPM-driven; focuses on intent

1.4 Return on Ad Spend (ROAS)

Formula:

ROAS=Revenue from AdsTotal Ad SpendROAS = \frac{\text{Revenue from Ads}}{\

text{Total Ad Spend}}ROAS=Total Ad SpendRevenue from Ads

Purpose: Measures financial return.

Interpretation:

 ROAS > 3x = healthy

 ROAS < 1x = loss-making campaign

Platform trends:

 Google Search → Highest ROAS (high intent)

 Instagram → Moderate ROAS (visual discovery leads to conversion)
 YouTube → ROAS typically low but boosts top-of-funnel demand

1.5 Brand Lift Metrics

Evaluated through surveys and digital lift studies (Google/Meta).

Measures improvement in:

 Brand awareness

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Ad recall
 Consideration
 Purchase intent

Interpretation:

 Strong brand lift on YouTube + Instagram → Good for upper-funnel brand

building
 Weak brand lift → Creative not memorable or targeting too broad

2. Hypothetical Example Summary Table

Platform CTR CPA CPM ROAS Brand Lift Interpretation

Instagram 1.2% ₹350 ₹240 2.0x High Good awareness, moderate conversions

Excellent recall, weak direct

YouTube 0.45% ₹500 ₹120 1.2x Very High
conversion

Google Search 4.5% ₹180 NA 4.0x Medium Strong intent, most profitable

3. Insights From Performance

Instagram

 Decent CTR → strong visual appeal

 CPA moderate → landing page may need optimization
 Good brand lift → strong storytelling

YouTube

 High brand lift → good for awareness

 Low ROAS & high CPA → unsuitable for immediate sales
 CTR low → creatives need improvement

Google

 Best CTR & lowest CPA → high purchase intent

 Highest ROAS → strongest channel for immediate conversions

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

4. Optimization Strategies

A. Improve CTR and Relevance

1. Refine audience targeting

o Lookalike audiences on Instagram
o Custom intent audiences on YouTube
2. Improve creative messaging
o Shorter YouTube hooks
o Clearer CTA on Instagram
o Add keywords dynamically in Google Ads

B. Reduce CPA (Increase Conversion Efficiency)

1. Landing Page Optimization

o Faster loading time
o Clear CTA (“Buy Now”, “Limited Offer”)
o Fewer form fields
2. Retarget High-Intent Users
o Instagram retargeting carousel for viewers who visited product pages
o Google remarketing for site visitors

C. Increase ROAS (Profitability Focus)

1. Shift budget to high-performing channels

o Increase spend on Google Search
o Reduce lower-performing broad YouTube spend
2. Campaign segmentation
o Segment by product, region, and audience
o Use SKAGs or tightly themed ad groups on Google

D. Improve Brand Lift and Recall

1. Creative storytelling on YouTube

o Short narratives, emotional content
o Strong brand mnemonic
2. Consistent visual identity
o Same colours, fonts, product visuals across platforms

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

E. Allocate Budgets According to Funnel Role

 YouTube → Awareness (10–20% budget)

 Instagram → Discovery + consideration (30–40%)
 Google Search → Conversions (40–50%)

5. Conclusion

Using CTR, CPA, CPM, ROAS, and Brand Lift, the D2C brand can comprehensively
evaluate performance:

 YouTube is strongest for brand building

 Instagram supports mid-funnel discovery
 Google Search drives direct conversions and profitability

A systematic optimization using better targeting, creative refinement, and funnel-aligned

budget allocation will significantly improve both profitability and long-term brand growth.

Question 5:
A soap brand invested heavily in TV ads and wants to measure the lingering impact of
advertising over time and its return. Explain how the Adstock model can be used to
analyze decaying ad impact.

Measuring Lingering Advertising Impact Using the Adstock Model

When a soap brand invests heavily in TV advertising, its impact does not occur only at the
moment the ad is aired. Instead, ad exposure influences consumers’ memory, brand recall,
and purchase behaviour over time. The Adstock model is a widely used econometric tool that
helps quantify this decaying yet persistent advertising effect.

1. What is the Adstock Model?

Adstock refers to the residual effect of advertising that carries forward into future periods
even when no new ads are shown.

It captures two key advertising realities:

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

1.1 Advertising has a carry-over effect

 Consumers do not instantly forget a TV commercial.

 Brand recall and persuasion persist over days or weeks.

1.2 Advertising impact decays over time

 Memory weakens unless reinforced with fresh ads.

 This is modeled using a decay rate.

Adstock therefore transforms raw advertising GRPs (Gross Rating Points) into a measure of
effective advertising pressure over time.

2. The Adstock Formula

The basic Adstock formula is:

Adstockt=Advertisingt+λ×Adstockt−1Adstock_t = Advertising_t + \lambda \times

Adstock_{t-1}Adstockt=Advertisingt+λ×Adstockt−1

Where:

 Advertising_t = TV ad spend or GRPs in current period

 λ (lambda) = decay factor, 0 < λ < 1
 Adstock_{t-1} = residual advertising effect from previous period

Interpretation of λ (Decay Rate)

 If λ = 0.5 → half the ad impact remains next period

 If λ = 0.8 → strong memory, slower decay
 If λ = 0.2 → fast decay, weak memory

For fast-moving consumer goods (FMCG) like soap, the decay rate is often 0.4–0.6,
because purchase cycles and memory windows are short.

3. How Adstock Helps the Soap Brand

3.1 Measures Lingering Impact of TV Ads

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

Instead of assuming ad effects disappear immediately, Adstock shows how influence

continues over time.

Example:

 Week 1: GRP = 100

 Week 2: GRP = 0, but with λ = 0.5 → Adstock = 50
 Week 3: GRP = 0 → Adstock = 25

Even without new advertising, the brand still benefits from declining consumer
memory.

3.2 Helps Identify the Optimal Frequency of Ads

The brand discovers:

 Are ads being shown too frequently (wasting money)?

 Or too infrequently (losing recall between bursts)?

Adstock helps calculate the point of diminishing returns.

3.3 Links Advertising to Sales More Accurately

Regression models combine Adstock with sales:

Salest=α+β×Adstockt+ϵSales_t = \alpha + \beta \times Adstock_t + \epsilonSalest

=α+β×Adstockt+ϵ

This helps quantify:

 How much each unit of Adstock (effective advertising) increases soap sales
 Whether TV ads are generating a positive return on investment

3.4 Allows ROI Calculation Over Time

Instead of evaluating ROI only in the campaign period, the brand evaluates:

 Short-term ROI = immediate sales uplift

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

 Long-term ROI = prolonged benefits from lingering Adstock

This is crucial because TV ads often have delayed recall and influence.

3.5 Helps Budget Reallocation

Insights from Adstock show if pressure is strong enough to:

 Reduce TV spend without losing effectiveness

 Shift budget to digital channels
 Run shorter but more frequent bursts (pulsing strategy)

4. Practical Example for a Soap Brand

Let’s assume:

 Week 1 GRP: 200

 Week 2 GRP: 100
 Week 3 GRP: 0
 Decay factor λ = 0.6

Adstock Calculation

Week GRP Formula Adstock Value

1 200 200 + 0.6×0 200

2 100 100 + 0.6×200 220

3 0 0 + 0.6×220 132

Interpretation

 Even with 0 advertising in Week 3, effective advertising pressure is still 132.

 TV ads continue to influence consumers even in dark periods.

This helps the brand understand the lagged impact on sales.

DIGITAL ASSIGNMENT Marketing Analytics

Hariharan Panda 24EMBA1613

5. Strategic Insights for the Soap Brand

A. Optimize Timing

Use bursts or pulsing rather than continuous heavy spending.

B. Improve Media Planning

Plan GRP waves so Adstock does not fall below optimal threshold before a sale cycle.

C. Enhance Budget Efficiency

Avoid overspending when Adstock is already high — saved money can go to digital.

D. Measure Campaign Fatigue

If Adstock stays high but sales plateau, creative refresh is needed.

E. Estimate Long-Term ROI

Adstock enables the brand to show management how TV spend yields compound benefits
over multiple weeks.

6. Conclusion

The Adstock model is essential for FMCG brands like soap manufacturers because it:

 Captures continuous and decaying advertising impact

 Provides a more realistic measure of ad effectiveness
 Links TV ads to sales through time-lagged effects
 Helps optimize media budget, creative rotations, and campaign timing
 Enables more accurate ROI evaluation

Thus, the Adstock model becomes a powerful analytic tool for maximizing the long-term
value of expensive TV advertising investments.

DIGITAL ASSIGNMENT Marketing Analytics