Slide 1

Slide 1 text

Dharshini M.B. Anu, Sri Madhavi Kanna SPAR International B.V. How is SPAR unlocking business value through Advanced Data Analytics?

Slide 2

Slide 2 text

SPAR Data Team

Slide 3

Slide 3 text

All will benefit from united co-operation Adriaan van Well and 14 other like minded people founded DESPAR in 1932 Founding Principle D S E P A R D oor E endrachtig S amenwerken P rofiteren A llen R egelmatig

Slide 4

Slide 4 text

2021 – A Year of Growth *Constant Currency Values +3.3%* Sales Growth €41.2 bn Global Sales 13,623 Stores Globally

Slide 5

Slide 5 text

Multi-Format Retail Strategy

Slide 6

Slide 6 text

Creating business insights to support the profitable growth of SPAR country organisations and SPAR International through the provision of an added value data service. SPAR Data Services – Better Together

Slide 7

Slide 7 text

SPAR Data Services - Principles Regional and International Business Insights Supplier Collaboration Simple and Easy

Slide 8

Slide 8 text

Business Intelligence Business Insights Ad Hoc Analysis /Insights Business Reporting Predictive Analytics 1 2 3 4 5 From Reporting To Analytics

Slide 9

Slide 9 text

Interactive Dashboards User self-service dashboards to give insights and allow investigation to identify and build on growth opportunities Predictive Analytics Bringing deeper insights into shopper missions, basket insights and customer segmentation, with comparisons to regional and international trends Project Analysis Project specific analysis for new stores, new format development or new concepts, with insights from other SPAR Markets Specialist Analytics Targeted data sets for use by Partners, integration into reporting or specialized external analysis SPAR Data Services Bringing strategic, development focused insights to Partners from regional and international insights on category, supplier and store trends

Slide 10

Slide 10 text

• Solution • Data upload in a secured way to Designer cloud • Online cloud storage • Connecting data to BI tool using ODBC drive Architectural Challenges Solved with Designer Cloud • Challenges faced • Different platforms & databases • Data volume • Regular & Continuous data collection

Slide 11

Slide 11 text

Currency conversions with daily volatility Multiple Suppliers for the same product Different Manufacturer Brand descriptions Data Challenges Categories and Sub- Categories with different naming conventions

Slide 12

Slide 12 text

• Collect the data from SPAR country organisations at Transaction Level • Data Cleaning and Data Standardisation • Data Preparation for the Analysis • Final Deliverable to the SPAR country organisations, through Power BI Methodology SPAR country organisation SPAR country organisation SPAR country organisation Data Preparation Country organisation Dashboard FMCG Dashboard Customized Analytics for Country organisations Specialized External Access

Slide 13

Slide 13 text

Data Collection Country organisation onboarding is complete only after we have completed all 5 of these steps and we are able to receive the data successfully on a monthly basis. Success 01 Onboard Onboard country organisation after checking their data format 02 Storage Compartment Data Sharing through S3 bucket 03 Initiate Receive initial data dump from country organisation 04 Verification Check if data is in the right format or not 05 Periodic Updates Automation of receiving files on periodic basis

Slide 14

Slide 14 text

Data Cleaning • Data wrangling is carried out at SPAR country organisation level • Periodic refreshes are done to update the data lake with the latest available information

Slide 15

Slide 15 text

Data Manipulation - Trifacta APIs Text translation APIs For country organisations where official language is not English, translation APIs convert local text into English Currency Conversion API Trifacta APIs pull latest conversion rates which are used in converting all types of currencies into Euros

Slide 16

Slide 16 text

Data Preparation - Text translation

Slide 17

Slide 17 text

Data Manipulation - Nielsen Standardization

Slide 18

Slide 18 text

Data Lake Creation Single Source Data Democratization Advanced Analytics • Aggregating all partners data into one platform creates one single source of data • Views built on this data lake allows all the users to compare the performance of all partners across regions • Central data creates a pathway to more advanced analytics and thus enabling the full potential of data

Slide 19

Slide 19 text

Interactive Dashboards

Slide 20

Slide 20 text

Business Insights through Dashboards

Slide 21

Slide 21 text

Promotion Analysis

Slide 22

Slide 22 text

Basket Analysis Supporting Growth with regional and international insights on category, supplier and store trends • Analysis of average basket value over past 12 months • Comparisons to region, with category level insights • Can be filtered live by: – Date range, region, brand or manufacturer • Charts and Data can be exported • Auto filter when a field is clicked Partner *Dummy data has been used in this visual

Slide 23

Slide 23 text

Currency Conversion

Slide 24

Slide 24 text

Data Insights in Retail Industry Descriptive Analysis Shopper Mission Analysis Customer Segmentation, Cross promotional Analysis Price Elasticity . Loyalty Program RFM & Churn Analysis Missions by time of day Social Media (Google Review Analysis) Store Clustering

Slide 25

Slide 25 text

Understanding the reason for a shopper’s journey to a store helps marketing and operational team for planning in advance Shopper Mission Analysis

Slide 26

Slide 26 text

Shopper Mission Grab & Go I’ running in for some grab&go I need fresh deli items Fresh Deli 15% Unique Receipts Share 25% Turnover Share 14% Volume Share 40% Unique Receipts Share 20% Turnover Share 20% Volume Share Average Volume 28,86 Average spend 29,19 Euros Average Volume 14,76 Average spend 17,27 Euros I forgot to buy something Mixed Mission 6% Unique Receipts Share 2% Turnover Share 4% Volume Share Average Volume 20,72 Average spend 26,91 Euros Average Volume 17,31 Average spend 24,79 Euros Average Volume 38,75 Average spend 56,63 Euros 20% Unique Receipts Share 45% Turnover Share 40% Volume Share 11% Unique Receipts Share 3% Turnover Share 6% Volume Share Family Shopping - Fresh I doing a big family shopping 8% Unique Receipts Share 5% Turnover Share 16% Volume Share Average Volume 18,09 Average spend 14,26 Euros Pasta – Small Basket I need few item with pasta I’m filling up my pantry Family Shopping- Pantry Shopping

Slide 27

Slide 27 text

Outcome of Shopper Mission Analysis The missions which gives maximum receipts, maximum turnover and similar insights can be derived analytically Example: Impulse missions may have higher number of transactions but with lower turnover The mission analysis supports for store clustering which in turn helps for well defined store design The store design is enhanced based on the buying pattern of the shoppers.

Slide 28

Slide 28 text

Learning more about the level of impact which a price change has on the sales of the product Price Elasticity Analysis

Slide 29

Slide 29 text

Price Elasticity Helps In Estimating Impact of Price Changes in Demand Introduction Usage Benefits Price elasticity is the measuring of the impact of a price change on the sales of the product Elasticity values help in classifying products into price sensitive and price insensitive categories Price Optimisation and Margin Maximisation are two key benefits from price elasticity analysis

Slide 30

Slide 30 text

Relationship Between Price Changes and Demand Fluctuations Methodology  Model uses Price and other important variables to explain the variations in demand  In the model, we have used Weekend/Holiday, Promotion and Weather as external variables  By looking at historical changes in price and its impact on the demand, model will try to estimate the elasticity of the product  We express |E| on a scale from 0 (inelastic) to 5 (highly elastic). Q = Sales Units; P = Price of Product

Slide 31

Slide 31 text

Product Results Summary Product Description Elasticity Weekend Promo Holiday Weather Sales Category Sub Category Brand XXX 5.00 (1.39) 0.00 (0.36) 0.08 726,337 XXX XXX XXX XXX 4.40 (1.12) 0.81 (0.06) 0.05 4,266 XXX XXX XXX XXX 4.01 (1.21) - (1.11) 0.12 5,352 XXX XXX XXX XXX 3.07 (4.38) - 0.84 0.54 30,470 XXX XXX XXX XXX 2.87 (6.10) 53.92 1.16 0.11 155,892 XXX XXX XXX Product Description Elasticity Weekend Promo Holiday Weather Sales Category Sub Category Brand XXX 0.73 (4.61) 4.09 (1.47) 0.20 161,443 XXX XXX XXX XXX 0.62 (0.84) 1.10 0.48 0.00 5,899 XXX XXX XXX XXX 0.57 (0.12) 0.27 (0.02) 0.01 8,162 XXX XXX XXX XXX 0.51 (0.51) 3.44 0.01 0.02 16,980 XXX XXX XXX XXX 0.28 (0.01) 0.06 (0.05) 0.01 6,461 XXX XXX XXX Elastic Products Inelastic Products Elasticity of 4 indicates, when a price is changed by 1% the unit sales of product is expected to change by 4% Promo flag of 54 indicates, that, in a day, this product sells 54 units more under promotion compared to regular sale Weather 0.54 indicates, an increase in temperature by 1O increases the sales by 0.5 units Weekend -4.61 indicates, this product sells 4.61 units less during weekends compared to weekdays Holiday -1.47 indicates, this product sells 1.47 units less during Holidays compared to working days Illustrative

Slide 32

Slide 32 text

Loyalty Programme Analysis Understanding customer needs and their expectation of products and services

Slide 33

Slide 33 text

• This programme is a reward for consumers who are repeat customers • Detailed analysis of customers (Frequent buyers, High spenders, Deal seekers) • Gives a clear picture about the worth of the loyalty programme in place and the percentage allocation of profit to the loyalty programme Loyalty Programme Analysis

Slide 34

Slide 34 text

Loyalty card analysis & Benefits Customer Retention Rate Understanding Customer Needs Customer Segmentation Analysis Report Focused Promotions of Products

Slide 35

Slide 35 text

Working with Designer Cloud Cloud Strage Reproduceable recipes Intuitive Data Wrangling API Usage (Translation, Currency) Reliable Support Easy connection: ODBC Drive Job Scheduling Easy to refresh with latest data Designer Cloud

Slide 36

Slide 36 text

Dharshini M.B. Anu Sri Madhavi Kanna SPAR International B.V. THANK YOU