Andrews Distributing quenches thirst for sales growth with analytics

Boosting sales for brewers and retailers by predicting consumer behavior at the point of sale

Published on 26-Aug-2013

"Our goal was to give our sales team a competitive advantage by providing them with actionable insights at the retail level, enabling them to help our retailers create great beer experiences for their beer-lovers." - Matt Canon, Business Planning and Analysis Manager, Andrews Distributing

Customer:
Andrews Distributing

Industry:
Wholesale Distribution & Services

Deployment country:
United States

Solution:
BA - Business Analytics, Business Integration, Business Performance Transformation, Business Resiliency, Empowering People, Enabling Business Flexibility, Enterprise Modernization, BA - Predictive Analytics, Supply Chain Management, Transforming Business

Overview

Since 1976, Andrews Distributing has been a leader in the wholesale tier of the Texas beer distribution industry. With 1,100 employees, four distribution centers and a fleet of more than 500 vehicles, Andrews delivers more than 250 brands of beer and other beverages from 32 suppliers to a network of over 10,000 retailers across Texas.

Business need:
Andrews Distributing saw an opportunity to offer greater value by complementing its highly efficient distribution model with analytics services – helping retailers stock the right products to maximize sales and profits, and giving manufacturers more accurate insight into consumer demand.

Solution:
As part of the IBM® SMART Program, Andrews worked with Southern Methodist University to develop a predictive analytics solution that mines data on 1,000 SKUs sold at 10,000 retailers, and reveals new insights about beer category sales in different retail environments.

Benefits:
Achieved a 35 percent compound monthly growth rate for a new beer by targeting specific retail accounts for distribution. Formed new partnerships with two breweries by demonstrating the value of a data-driven strategy. Improved resource efficiency by 75 percent using SPSS models instead of spreadsheets.

Case Study

When your distribution business is already a market leader, how do you extend your advantage?

As a distributor, Andrews Distributing recognized that its own success was linked to the success of its supply-chain partners – not only the retailers that sell its beers, but also the suppliers that brew them. If the company could find a way to boost retail sales, the whole market ecosystem would benefit.

Andrews realized that its position in the supply chain gave it a unique opportunity to gather data on the entire North Texas retail market. If it could mine this data for new insights into consumer behavior and purchasing patterns, it would be able to help its partners market and sell their products more effectively.

What Andrews needed were the tools and expertise to make sense of its data. Working with professors and students from Southern Methodist University as part of the IBM Statistics and Mining in Academic Research and Training (SMART) program, the company built a solution that is helping to transform its business model, complementing its traditional distribution model with a range of innovative analytics services.

Setting the scene

Since the repeal of prohibition in 1933, the three-tier structure of beer distribution in the state of Texas has remained the same. To manage and regulate the industry optimally, the Texas Alcohol and Beverage Commission (TABC) oversees the process where suppliers may only sell to retailers indirectly, via an independent wholesaler. And since 1976, Andrews Distributing has been a leader in the wholesale tier.

With 1,100 employees, four distribution centers and a fleet of more than 500 vehicles, Andrews delivers more than 250 brands of beer and other beverages from 32 suppliers to a network of over 10,000 retailers across Texas.

Matt Canon, Business Planning and Analysis Manager at Andrews Distributing, describes the market in which the company operates: “We supply retailers of all kinds, from mom-and-pop stores to large retail chains to bars, clubs and restaurants. It’s a complex market, not only because there are so many types of retailer, but also because the retail environment itself changes depending on location and demographics. A small store in an economically depressed area won’t sell the same types of product as a similar store in a more affluent area.

Sarah Davies, Business Analyst at Andrews Distributing, adds: “On top of the complexity of the retail landscape, the beer market itself is undergoing some very significant changes. Thanks to constant innovations in beer styles and flavors from our major brewing partners such as MillerCoors and from craft and specialty brewers, the number of brands and SKUs we distribute is rising all the time – and it’s becoming more and more difficult for individual retailers to work out the optimal product mix for their stores.”

Canon continues: “Because we work with so many retailers and distribute beers from so many suppliers and brands, we’re in a unique position within the North Texas market: the data from our invoices can tell us very precisely which brands are being sold in which stores, which SKUs are most popular in which locations, and what volumes each retailer is selling. We realized that if we could find a way to turn this data into valuable insights for our retailers and suppliers, we could add a lot more value to the whole supply chain.

“We have a powerful beer portfolio, and customizing it for our retailers’ customers creates tremendous value. A retailer that offers an optimal assortment of products can transform the way its customers perceive it. Instead of being just a standard liquor store or restaurant, it can become a destination for beer-lovers far and wide.”

Turning data into dollars

The senior management team at Andrews, led by the company’s President, Mike McGuire, places a strong emphasis on harnessing technology to develop new opportunities for the business, and asked the business planning and analysis team to investigate the options for a data-mining and predictive modeling solution.

“The mission of the business planning and analysis team is to continuously improve upon Andrews’ analytic tools, resources, and capabilities,” says Canon. “Our goal was to give our sales team a competitive advantage by providing them with actionable insights at the retail level, enabling them to help our retailers create great beer experiences for their beer-lovers.

“At that time, we had very little experience of data-mining technologies, but we were able to form a partnership with Southern Methodist University under the IBM SMART program. This was a win-win situation for all parties involved: we got access to some of the best academic resources and brightest students in SMU’s data-mining program, while the university benefited from using us as a real-life test case for their research. At the same time, the students gained valuable experience of working in a business context: in fact, we were so impressed that we hired one of them at the end of the project.”

“The project was a learning opportunity for us on how to create a systematic approach to the right beer brands in the right accounts for our retailers’ consumers. With an insights-based, data-driven approach to decision-making, we knew we would be able to build our credibility as trusted advisors for our retailers.”

Data-mining becomes business-as-usual

Davies, having graduated from SMU, now works at the heart of the company’s predictive analytics team, and has played a huge role in making data-mining part of business-as-usual at Andrews.

“Working with Andrews as part of the SMART program was a great experience, but working there full-time is even more exciting,” she comments. “We’ve now developed a cluster model in IBM SPSS Modeler that allows us to run monthly analyses of consumer interactions with different types of retailers, and find out which SKUs sell best in each kind of outlet.

“From the results, we generate dashboards that give each retail account a set of individual recommendations about how they should change their product mix to boost sales. The process is mostly automated, so it’s about 75 percent faster than using traditional spreadsheet-based approaches.”

New insight into retail product mix

Some of the insights revealed by the solution are already having a profound effect on retailers’ sales strategies. For example, Andrews is able to analyze the product mix of the best performing stores in each of its clusters (a cluster might be “liquor stores in Dallas”). Using this information, it can then look at how the mix in less successful stores differs, and suggest improvements (“you are underperforming on imported beers for your consumer base”).

Andrews is also now able to identify stores that are highly “adaptable” – good at introducing new brands and SKUs – which has proven to be a characteristic of the most successful retailers.

“From a sales perspective, this new level of insight transforms our position in the supply chain,” says Canon. “We’re not just the guys who deliver the beer – we’re able to play more of a consultancy role, and have much more constructive conversations with store managers about how to make the best of their businesses. Increasingly, our retail partners now rely on us to help them get it right for their consumers.”

Mike McGuire, President of Andrews Distributing, comments: “At a strategic level, developing our capabilities to execute the optimization of assortments is very important to our position as category leader and trusted advisor to the retailer. Moreover, since the model is aimed at the retailer’s customer, this capability becomes a cornerstone of our customer-centric approach.”

More successful product launches

Insight from Andrews also pays dividends for the suppliers, particularly in terms of maximizing the impact of new product launches and allocating marketing spend more efficiently. Introducing the right new brands in the right package sizes to the right stores can significantly boost the total sales of a new product, and help it build momentum in the wider market.

“For example, during the launch of Redd’s Apple Ale, we used a data-driven analytical strategy to target accounts for distribution,” says Canon. “This year, between January and July 2013, we have grown Redd’s volume at a compound monthly growth rate of 35.5 percent. I would say Redd’s Apple Ale is off to an impressive start.”

Reputation for innovation and passion for analytics

Suppliers are impressed by Andrews’ data-driven strategy and passion for analytics, and two new brewing companies have recently formed partnerships with the company as a result.

“Increasing sales at the retail level means increasing sales at the supplier level too, so everyone benefits,” says Canon. “The ability to provide these analyses quickly and easily sets us apart from the competition and makes us a more attractive proposition for our brewing partners.”

Davies concludes: “Best of all, this is really just the beginning for data-mining at Andrews. Now that we have the platform in place, there are so many areas of opportunity – forecasting, ordering, inventory on hand, pricing and discounts to name just a few. IBM SPSS Modeler gives us the tools we need to deliver the insights that matter to our retail and supplier partners, and to our own business.”

About IBM Business Analytics

IBM Business Analytics software delivers data-driven insights that help organizations work smarter and outperform their peers. This comprehensive portfolio includes solutions for business intelligence, predictive analytics and decision management, performance management, and risk management.

Business Analytics solutions enable companies to identify and visualize trends and patterns in areas, such as customer analytics, that can have a profound effect on business performance. They can compare scenarios, anticipate potential threats and opportunities, better plan, budget and forecast resources, balance risks against expected returns and work to meet regulatory requirements. By making analytics widely available, organizations can align tactical and strategic decision-making to achieve business goals.

For more information

For further information please visit ibm.com/business-analytics

Products and services used

IBM products and services that were used in this case study.

Software:
SPSS Modeler

Legal Information

© Copyright IBM Corporation 2013. IBM Corporation, Software Group, Route 100, Somers, NY 10589. Produced in the United States. August 2013. IBM, the IBM logo, ibm.com and SPSS are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the web at: www.ibm.com/legal/copytrade.shtml. This document is current as of the initial date of publication and may be changed by IBM at any time. Not all offerings are available in every country in which IBM operates. The client examples cited are presented for illustrative purposes only. Actual performance results may vary depending on specific configurations and operating conditions. THE INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS” WITHOUT ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING WITHOUT ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ANY WARRANTY OR CONDITION OF NON-INFRINGEMENT. IBM products are warranted according to the terms and conditions of the agreements under which they are provided. The client is responsible for ensuring compliance with laws and regulations applicable to it. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the client is in compliance with any law or regulation.