Emami to unlock Untapped Potential
Emami holds a dominant position in multiple high-growth categories. The Emami TP Intelligence Framework is designed to leverage this market leadership into smarter, more profitable growth.
Navratna
Cool Oil & Talc
Seasonality: Summer, High Heat Periods
Target: Broad (Rural & Urban)
BoroPlus
Antiseptic Cream & Lotion
Seasonality: Winter, Dry Seasons
Target: Broad (Family-oriented)
Zandu & Mentho+
Pain Relief Balms
Seasonality: Winter, Cold/Flu Season
Target: Broad (Rural & Urban)
Kesh King
Ayurvedic Hair Care
Seasonality: Year-round
Target: Hairfall-conscious consumers
Fair and Handsome
Men's Fairness Cream
Seasonality: Year-round, Pre-festive
Target: Men (Urban & Rural)
Dermicool
Prickly Heat Powder
Seasonality: Summer
Target: Broad (Urban & Rural)
Why a ₹400 Crore Spend Can Deliver Suboptimal ROI?
The Challenge
Emami's annual trade promotion spend of ~₹400 Crores is a significant growth lever, but the current ecosystem lacks the required depth of visibility and data-driven decision-making. The result is suboptimal ROI, potential channel conflict, and an inability to proactively plan promotions based on predictive insights.
Expected Business Outcome
A conservative 5-10% improvement in promotion effectiveness, translating to ₹20-40 Crores in either direct savings or reinvested for incremental growth, alongside improved forecast accuracy and transforming spend from a "cost of doing business" into a precision-guided growth engine.
The Intelligent TPO Platform Solution
We propose an end-to-end solution that moves Emami from reactive analysis to predictive and prescriptive optimization. This is not a dashboarding project; it is an AI-powered decision-making system. It will enable Emami to:
- See the Past: Unified visibility into performance.
- Understand the Present: Monitor promotions in real-time.
- Predict the Future: Simulate "what-if" scenarios.
- Prescribe the Best Action: AI-driven recommendations.
Cyclical Trade Promotions Process
A superior process generates higher quality data, which in turn fuels more accurate analytics and leads to smarter, more profitable decisions. As Emami develops confidence in the Tech and Business Consulting capabilities of Searce, we would be excite to automate a lot of this process.
Emami Trade
Promotions Cycle
Key Stakeholders and Their Roles
Successful transformation requires seamless collaboration. Here’s how key roles interact with the TPO system across the promotion lifecycle.
Diversity & Scale make Trade Promotions complex
Emami's success hinges on managing a fundamental dichotomy: driving seasonal peaks for its high-margin "Power Brands" while nurturing consistent growth for its year-round brands. This requires a differentiated approach for the vast General Trade and the fast-growing Modern Trade channels.
Framing the Promotions Calendar as per the AOP
Timing is everything. Emami's promotions are meticulously aligned with seasonal demand and cultural events.
Planning, Building and Operationalizing
Emami can break down the long journey at multiple points to ensure the direction and pace is right. While planning, its important to have the Trade Promotion Management be the 1st goal and Optimizations later. The same would apply during the Development cycle that Data Foundations are step 1, followed by more rigorous analysis, followed by more creative automations. In terms of Maturity, this would lead to Level 1 to Level 4.
Trade Promotion Management
- Planning & Budgeting: Hierarchical financial planning with real-time budget tracking.
- Promotion Calendar & Execution: Visual planning with configurable tactics and automated approval workflows.
- Claims & Deduction Management: AI-powered claims matching and dispute resolution to reduce revenue leakage.
- Reporting & Analytics: Dashboards visualizing KPIs like Sales Lift and ROI with drill-down capabilities.
Trade Promotion Optimization
- Predictive Forecasting: AI engine predicts baseline sales and incremental uplift for promotions.
- Simulation Sandbox: "What-if" environment to model and compare scenarios, optimizing plans before launch.
- Analytical AI Agent: A key differentiator using a conversational interface (NLP) to analyze data and provide answers, democratizing data analysis for all users.
1. Data Foundation
Single Source of Truth
What actually happened?
2. Diagnostic & Predictive AI
The Brains of the Operation
Why & What will happen?
3. Simulation Engine
The Planner's Cockpit
What if we try this?
4. Execution & Performance
Closing the Loop
What is the best action?
Top 50 Questions Beyond the "What actually happened?"
Unlocking the true effectiveness of trade promotions requires moving beyond simple sales tracking. Here are the critical questions leading FMCG brands are asking, and how AI provides the answers.
Baseline & Incrementality
Promotion Mechanics & ROI
Channel & Customer Dynamics
Supply Chain & Operations
Competitive & Market Factors
Long-Term Brand Health
What does it take to be able to answer such questions?
This solution is "more than mere dashboards" because of these specific, interconnected components that drive from data to decision.
Demand Forecasting & Baseline Generation
Creates a robust baseline, accounting for seasonality, holidays, and trends.
What It Is:
This model predicts the sales of a product in a specific region and channel as if no promotion were running. This predicted sales figure is the "baseline." It is the most fundamental building block of TPO.
Why It's Critical:
Without an accurate baseline, it's impossible to measure the true success of a promotion. The incremental lift is calculated as: Actual Sales - Predicted Baseline Sales. A poor baseline leads to an incorrect ROI.
Practical Example for Emami:
The model predicts that, without any promotion, Emami would sell 100,000 units of Navratna Cool Talc in North India in May. This 100,000-unit figure is the benchmark. If Emami runs a promo and sells 130,000 units, the true incremental lift is 30,000 units.
Causal Uplift Modeling
Isolates the true impact of the promotion from other confounding factors.
What It Is:
This model isolates the causal impact of the promotion itself, filtering out the "noise" from other simultaneous events (like an advertising campaign or a heatwave). It answers: "How many extra units did we sell because of this specific promotion and for no other reason?"
Why It's Critical:
Correlation is not causation. This model prevents misallocating budget to ineffective promotions that were just "lucky" to run during a period of high natural demand.
Practical Example for Emami:
Emami runs a BOGO offer on Kesh King in Maharashtra while also running a national TV ad. The model determines that sales increased by 25%, but concludes that the TV ad was responsible for a 10% lift, and the BOGO offer was causally responsible for the remaining 15% lift.
Price Elasticity Modeling
Understands how consumers react to price changes for different SKUs.
What It Is:
This model quantifies how sensitive consumer demand is to changes in price for a specific product, in a specific channel.
Why It's Critical:
This is the key to optimizing discount depth. It helps Emami avoid offering a 30% discount when a 15% discount would have achieved 90% of the same volume lift, thus saving margin.
Practical Example for Emami:
For Zandu Pancharishta, the model finds an elasticity of -1.8. This tells the planner that a 10% price reduction is predicted to increase sales volume by 18%. They can now simulate the P&L to find the optimal discount.
Cannibalization Modeling
Predicts the impact of a promotion on a "basket" of related products.
What It Is:
This model measures the "side effects" of a promotion, such as how promoting one product decreases sales of other products in the portfolio.
Why It's Critical:
It ensures a holistic, portfolio-level view of profitability. A promotion on a single SKU might look successful on its own, but if it decimated the sales of a higher-margin product, the company as a whole lost money.
Practical Example for Emami:
A 25% discount on large BoroPlus cream might increase its sales by 50,000 units but decrease sales of the smaller BoroPlus cream by 15,000 units. The final ROI calculation must sum the profit/loss from all these effects.
The Trade Promotion System
A fully custom-built system designed to integrate seamlessly with Emami's core SAP and DMS platforms. This provides an unparalleled, end-to-end view of the promotion lifecycle, transforming siloed data into a strategic asset for predictive planning and profitable growth.
Dashboard
Financial Settlement & Reconciliation
Deductions & Claims Funnel
Accrual Management
Total Accrued
₹1.25 Cr
Forecasted for Month
₹0.45 Cr
Actualized this Month
₹0.80 Cr
Pending Reversal
₹0.15 Cr
Post-Event Analysis (PEA): BoroPlus BOGO
A deep dive into the true drivers of promotion performance.
True ROI: 15.2%
Calculated based on incremental volume only for accurate profitability.
Simulation & Forecasting
"What-If" Promotion Simulator
Predicted P&L
Incremental Lift
0 units
Promotion Cost
₹0
Predicted ROI
0.0%
Prescriptive Recommendation
Optimal Calendar Generated
To maximize annual profit for D-Mart, our engine suggests replacing the Q4 20% TPR with a 'Display + 10% TPR' combo.
- Predicted Profit Increase: +₹12.5 Lakhs
- Volume increase of +8% within budget.
Scenario Comparison: Maximum Volume vs. Maximum Profit
Intelligent TPO Assistant
Ask any question, simulate scenarios, and get AI-powered recommendations in seconds. This is your interactive growth engine.
Modes
Standard Explorations
Data Sources
Daily SAP Sales Reports
DMS Promo Campaigns
SAP Promo Campaigns
Behind the Scenes: How the Engine Works
"Example: what would our sales have been for Navratna Oil in North India if we had run no promotion at all?"
The Granular Foundation
To build a reliable baseline, we move beyond aggregated reports to create a "single source of truth" by combining granular, time-series data from your core systems.
Master data (Product/Customer Hierarchies) and transactional data (Primary Sales, Pricing, Promotions).
★ Critical Secondary Sales data (distributor to retail) and stock statements.
POS data, public holidays, festivals, and even weather data.
Traditional vs. AI-Powered View
Traditional (Aggregated)
AI-Powered (Granular)
Traditional views hide risk. The AI view reveals the ground reality across hundreds of districts.
From Statistics to Machine Learning
A simple average is highly inaccurate. We use advanced models to analyze the data and create a reliable baseline. Use the controls to see how the AI builds this baseline from different components.
This is the key differentiator. A traditional dashboard can't create this baseline, making it impossible to calculate true incremental lift and ROI.
Presentation to Management: The Interactive Dashboard
We translate complex models into clear, actionable insights through a suite of interactive dashboards, turning data science into a tool for exploration and decision-making.
1. The Baseline vs. Actuals View
A powerful time-series chart where the visual gap between actual sales (solid) and the AI baseline (dotted) instantly represents the total volume sold on promotion.
2. Decomposing the "Why"
This dashboard builds trust by making the model transparent, showing the different components used to build the baseline: trend, seasonality, and holiday impact.
3. Quantifying the Uplift
★ The final step. This focuses on the financial outcome, providing the clean, fact-based input needed for accurate ROI calculations.
The Army of Custom-Designed AI Agents 🤖
In the past, we had Human Analysts spending hours and even days to process data to draw insights. With new Agentic AI approaches, it is possible to have Specialist AI Agents work in tandem to perform complex tasks that require - reasoning, planning, execution, edge cases handling, coordination and more.
User Query
"Analyze the ROI of the 'Monsoon Bonanza' promotion."
Root Agent
Interprets intent and delegates tasks to the Core Agents Layer
The Core Agent Layer
Agent Toolkit
Core Agents use these specialized tools to execute tasks. Click a tool to see details.
Tool: Database Agent (NL2SQL)
Translates natural language into SQL queries.
This tool bridges the gap between human language and structured databases. A Core Agent (like a Business Analyst Agent) uses this tool to convert a user's question into the correct SQL code to get an answer from Emami's DMS platform.
Input Prompt: "What was the sales uplift and trade spend for the 'Navratna Cool Talc' promotion in Uttar Pradesh?"
-- Tool translates the prompt into the following SQL
SELECT
p.brand,
SUM(s.trade_spend) AS total_spend,
AVG(s.sales_uplift_percentage) AS avg_uplift
FROM
promotions s
JOIN
products p ON s.product_id = p.id
JOIN
regions r ON s.region_id = r.id
WHERE
p.brand = 'Navratna Cool Talc' AND r.state = 'Uttar Pradesh'
GROUP BY
p.brand;Result from DMS:
[
{
"brand": "Navratna Cool Talc",
"total_spend": 4500000,
"avg_uplift": 15.7
}
]Tool: Visualization Agent
Generates specifications for interactive charts.
A Core Agent uses this tool to take structured data and transform it into a visual format. It generates precise instructions (often in JSON) that a front-end library (like Plotly) can use to render an interactive chart for a TPO dashboard.
Input Data: (JSON for multiple promotions)
// Tool generates a Plotly JSON specification
function create_roi_chart(data) {
const brands = data.map(item => item.brand);
const uplifts = data.map(item => item.avg_uplift);
return {
"data": [{
"x": brands,
"y": uplifts,
"type": "bar",
"marker": {"color": "#2dd4bf"}
}],
"layout": {
"title": "Average Sales Uplift % by Brand",
"paper_bgcolor": "#1e293b",
"plot_bgcolor": "#1e293b",
"font": { "color": "#cbd5e1" }
}
}
}Rendered Chart:
Average Sales Uplift % by Brand
Tool: Report Generation Agent
Synthesizes insights into structured reports.
This tool takes all the gathered data and visualizations and combines them into a human-readable summary. A Core Agent would use this as the final step to present findings back to the user.
Generated Report Snippet:
Promotion Analysis: 'Navratna Cool Talc' in Uttar Pradesh
The promotion for Navratna Cool Talc in Uttar Pradesh resulted in an average sales uplift of 15.7% against a total trade spend of ₹45,00,000.
Recommendation: The ROI on this promotion is strong. Consider reallocating budget from lower-performing promotions to replicate this campaign in adjacent regions during the next planning cycle.
System Generated Response
Delivers a complete, business-savvy insight to the user.
12-Month AI-Powered TPO Transformation Program
An interactive timeline of key deliverables and business outcomes.