Unveiling the secrets of FMCG success: Harnessing data-driven sales predictions in an ever-changing consumer environment

The FMCG market has experienced substantial global growth over a significantly long period, by 2010, the industry had created 23 of the world’s top 100 brands and had grown total return to shareholders (TRS) almost 15 percent a year for 45 years—performance second only to the materials industry. However, its trends are constantly evolving due to dynamic consumer behaviour.
The field of FMCG analytics is revolutionizing data operations within organizations and the focus from Product is now shifting rapidly towards Consumers.
23 October 2023
KAL
chaitanya
Chaitanya
Akula

Senior Manager, Data Strategy & Engineering

kandarp
Kandarp
Vyas

Principal, Data Strategy & Engineering

Introduction

The ever-evolving consumer landscape, shaped by notable occurrences like the Covid-19 pandemic, the Ukraine war, and subsequent Russian sanctions, along with the spectre of inflation, has brought about a remarkable transformation in how consumers engage in the buying process.
These changes are far from uniform, exhibiting systematic disparities across regions and countries, as well as within product categories, effectively distinguishing between essential and non-essential goods. Particularly vulnerable are countries heavily reliant on energy and oil supplies, as they bear the brunt of substantial retail price increases.

Problem Statement

Traditional forecasting techniques are backward looking predominantly depending on historical data and often fall short when it comes to capturing the complexities of changing consumer behaviour.

1. Inadequate consideration of external Factors: Overlooking the influence of external factors (macro trends, social dynamics, technological advancements, geopolitical events, recessions, natural disasters, and pandemics) and focussing on historical data and internal factors (price, promotions and market share) can lead to inaccurate forecasts and missed opportunities.
2. Changing consumer preferences: Consumer preferences are dynamic and diverse, requiring flexible forecasting models to account for changing lifestyles, cultural shifts, sustainability concerns, digital transformation, and health and wellness trends.3. Complex interactions and dynamics: Consumer behaviour is a complex interplay of various factors, and traditional forecasting techniques often oversimplify these interactions, while consumer behaviour-specific methodologies offer a more nuanced understanding of the intricate cause-and-effect relationships that shape preferences and buying patterns.


High Level Solution

An effective forecasting solution incorporates consumer behaviour-specific methodologies, employing advanced analytics, market research, data science, and consumer psychology. By assimilating consumer trends from diverse sources, this forecasting solution enables informed decision-making for customized marketing strategies and offerings.

Retail & media: Modelling data and marketing mix retail volumes, values, price, distribution, promo, media spends, GRPs.

Macro data: External macro measures to understand the economic condition GDP, PDI, UR, inflation, prod. index.

Survey: Survey data that reflect evolving trends and any behavioural changes.

Consumer: Panel data for understanding consumer behaviour, penetration, consumption, frequency, spend per buyer.

Mobility, pandemic: Covid fluctuations in mobility, stringency, new waves impacting the consumer behaviour, mobility, stringency, active cases, deaths.

Secondary research: Additional factors that impact seasonal categories or any unusual promos temperature data, big players exiting.

Multi Stage Robust Model

Consumer behaviour: Consumer trajectory basis social / search, Mobility, Pandemic Impact, Survey data.

Benefits evolution: Benefit Growth trends for core and new benefits.

Training & cross validation: With test and train data, increasing window validation is performed to retain the time series nature and have stable fit.

Ensembled model: Ensembling technique that combines several linear & non-linear base models to produce one optimal predictive model.

Solution Details

Our solution offers a multitude of unique features that set it apart from traditional forecasting approaches.

KAL
 
Data Agnostic
The solution can perform forecasts using multiple data sources:
- Retail panel data
- Consumer panel data
- Ecomm sales data
- Secondary research

KAL

Explanatory

The solution decomposes factors affecting the forecast and provides reasons for growth.
 

KAL

Consumer Focused

Incorporates consumer behaviour by understanding purchase dynamics leveraging data and insights from:
- Consumer panel
- DX consumer trends
- Mobility
- Consumer confidence

KAL
 
Provides Simulator

There is an in-built simulator where users can change input variables and observe growth scenarios.

KAL

Reliable & Scalable

Uses a mix of linear and non-linear modelling technique to provide best fit results. It provides scalability in:
- Managing multiple cells simultaneously
- Provides quick refresh options

Through our data-agnostic approach, unwavering focus on accuracy and reliability, consumer-centric methodology, and expert consulting services, our solution offers a comprehensive and robust forecasting solution that goes beyond conventional methods. 

Business Benefits

Through our solution, we have empowered our clients to make informed decisions, achieve their strategic goals with confidence and take proactive actions in the following areas:

• Efficient supply chain management: By anticipating demand patterns with precision, our forecasts helped streamline production, inventory management, and distribution processes for a global personal care major. This minimized the risk of overstocking or stockouts, leading to improved operational efficiencies during disruptions by 20%.
• Effective resource allocation: Reliable forecasts allowed our client, a leading home care products manufacturer to allocate resources effectively. This included planning production capacity and procurement of raw materials. This contributed to cost savings and operational effectiveness.
• Enhanced marketing and promotions: Our explanatory results enabled our client, largest cleaning & laundry supplier to plan targeted marketing campaigns and promotions. By understanding consumer demand patterns, they were able to identify peak sales periods, seasonal trends, and specific customer preferences. As a result, it helped increase promo ROI by 12% and increased brand awareness by 100-200 bps.
• New product introductions: By understanding market dynamics and consumer preferences, our solution provided pockets of opportunity and potential demand areas for new offerings, this helped a leading FMCG player plan NPD and launch products effectively in a 100+ Bn USD market. 
• Competitive advantage: With precise demand predictions and predictions for various macro scenarios, our client a home care major were able to anticipate market trends, respond to changing consumer preferences, and stay ahead of competitors. This allowed for agile decision-making, quicker response times, and presented the clients with qualified opportunity of 2Bn USD globally.
• Financial planning and risk management: Using the forecasts to project sales revenues, assess profitability, and determine budget allocations. Moreover, reliable forecasts aid in identifying potential risks and uncertainties, allowing companies to develop contingency plans, manage market volatility, and mitigate potential disruptions.


Want to learn more? Contact us.