4-page Case Study
Posted: 5/30/2012
3598
Rate This Evidence:

Hy-Vee Hy-Vee Boosts Performance, Speeds Data Delivery, and Increases Competitiveness

Hy-Vee operates a growing chain of employee-owned supermarkets in eight states in the midwestern United States. To boost its competitiveness, the company sought to increase its data warehouse performance, so it could deliver store-level purchasing data more quickly to its business analysts and managers. Hy-Vee also wanted to expand the capacity of its data warehouse. The company implemented a new Microsoft data warehouse solution that combines software with cost-effective hardware. The solution has increased query performance by up to 100 times, and Hy-Vee can now provide timely data to analysts every morning. The company has also improved its item-level reporting capabilities, and it can better understand its customers’ purchasing habits. Additionally, Hy-Vee can scale for the future and free up resources for its IT staff.

Situation

Iowa-based Hy-Vee is a growing grocery store chain that operates 235 employee-owned supermarkets in eight states in the midwestern United States. These full-service stores feature bakeries, delicatessens, banks, florists, pharmacies, and coffee shops. The company also operates 100 gas stations with convenience stores. In 2011, Hy-Vee had total sales of U.S. $7.3 billion, which ranked it among the top 25 supermarket chains and among the top 50 private companies in the nation.

*
* The SQL Server 2008 R2 Parallel Data Warehouse solution helps us recognize changes in our customers’ buying habits. We can then respond to those changes before our competitors do, and that gives us a huge business advantage. *

Tom Settle
Assistant Vice President
Data Warehousing
Hy-Vee

*
Like most retailers today, Hy-Vee must constantly find new ways to compete in order to increase sales and gain new customers. “Competition is our main challenge, especially because people now have so many more stores to choose from than they used to,” says Tom Settle, Assistant Vice President, Data Warehousing, Hy-Vee. “Also, customers’ needs and wants are changing all the time, so we need to try and keep up with changes in their buying habits and respond to those changes.”

To do that, the company analyzes customer data every day. This data, collected from all Hy-Vee locations, is stored in an 11-terabyte data warehouse based on Microsoft SQL Server 2008 R2 data management software. Each day, more than 800,000 rows are imported into the warehouse—a number that is growing rapidly. “We collect data on every customer transaction, so we have information on the items they buy, the time of day they buy, how they paid, and many other things,” Settle says. “It’s very important for us to understand all the details of customer purchases, because it gives us insight into the categories we need to focus on.”

For the Hy-Vee marketing analysts who need to create new in-store promotions, it is critical that they have the latest information each morning at 6:00. However, processing the data on time each day was not always possible with the previous hardware. “Every morning at 3:00, all the data from our stores is gathered in the data warehouse, and we only have a three-hour window to process that data,” says Settle. “However, the process was not at all consistent. It simply took too long to load the files, and query times were too slow. We need to get that data to our employees for analysis first thing in the morning. If they don’t have it on time, they don’t have the most updated data for analyzing promotions.”

Additionally, Hy-Vee wanted to expand its data warehouse. “We had about two years of data on the old hardware, and we had plans to build that out to seven years of data,” Settle says. “It’s difficult to really identify trends using only two years of information. We need at least five years of data to do that properly. We simply needed more space in the warehouse.”

Ultimately, this combination of business challenges led Hy-Vee to look for a new data warehouse solution in January 2011.

Solution

Although Hy-Vee had considered solutions from other vendors in the past, the company wanted to stick with Microsoft technology for its new data warehouse. “We wanted to take advantage of our existing SQL Server knowledge base,” says Settle.

*
* HP and Microsoft Services working together made it very easy for us. We only had to call one number for assistance on hardware or software questions, and everything was resolved very quickly as a result of that collaboration. *

Tom Settle
Assistant Vice President
Data Warehousing
Hy-Vee

*
Hy-Vee decided to implement Microsoft SQL Server 2008 R2 Parallel Data Warehouse, a comprehensive data warehouse solution that combines SQL Server technology with low-cost industry-standard appliances. The solution delivers a massively parallel processing (MPP) architecture that provides extremely fast performance and scalability up to hundreds of terabytes of data.

In addition, Parallel Data Warehouse offers the simplicity of an appliance and added value through integration with Microsoft business intelligence (BI) tools. “We had already been using Microsoft BI tools such as Microsoft SQL Server 2008 R2 Analysis Services and SQL Server 2008 R2 Reporting Services, and that was a big advantage of implementing Parallel Data Warehouse,” says Settle.

In March 2011, the Central Region Parallel Data Warehouse team, the Hy-Vee account team, Microsoft Technology Center (MTC) staff in Minneapolis, and the customer worked with HP to implement a proof of concept (POC) for a new Parallel Data Warehouse solution, based on HP hardware. The POC was conducted at the Minneapolis MTC, executed on remote Parallel Data Warehouse appliance hardware, and delivered by HP to the MTC in Chicago. During the POC, Hy-Vee analyzed its current architecture and its future needs, and looked at the options that could meet them.

In addition, Hy-Vee was able to take advantage of Premier Mission Critical Support from Microsoft Services. Through this program, Microsoft offers personalized support coverage, including around-the-clock problem resolution and a single number to call for questions about HP and Microsoft solutions. “HP and Microsoft Services working together made it very easy for us,” says Settle. “We only had to call one number for assistance on hardware or software questions, and everything was resolved very quickly as a result of that collaboration. The entire implementation was smooth and easily managed, and we were up and running quickly.”

Another Hy-Vee technical partner, Sogeti, helped automate the new solution’s extract, transform, and load process.

Beginning in August 2011, Hy-Vee started loading three years of point-of-sale (POS) data into the new data warehouse. The company is in the process of expanding that to seven years’ worth of data. The new Parallel Data Warehouse processes all of the company’s transaction data and loads it into an operational data store. The solution then processes the data into fact tables used for building online analytical processing (OLAP) cubes in SQL Server 2008 R2 Analysis Services. Using this data, Hy-Vee purchasing and category management teams can look at the performance of specific grocery items to determine trends in sales and gross profits.

*
* During the proof of concept, basic queries that previously took 20 minutes only took 20 seconds using SQL Server 2008 R2 Parallel Data Warehouse. This solution dramatically improves our query performance and nightly processing times.  *

Tom Settle
Assistant Vice President
Data Warehousing
Hy-Vee

*
Hy-Vee executives and analysts also use the solution for ad hoc queries to create analytical reports in SQL Server 2008 R2 Reporting Services. The queries are run against the Parallel Data Warehouse to take advantage of the performance improvement gained from the solution’s MPP architecture.

The new solution, which went into production in January 2012, is being used by 35 employees at Hy-Vee corporate headquarters. By the end of 2012, about 500 employees, including top executives and store managers, will use the solution for reporting.

Benefits

With the SQL Server 2008 R2 Parallel Data Warehouse solution, Hy-Vee has increased query performance by up to100 times and can now consistently deliver data on time each morning. In addition, the company has new reporting capabilities, can better understand its customers, and is more competitive as a result. Hy-Vee also has a more scalable solution, and its IT employees have more time to focus on other aspects of the business.

 Boosts Query Performance by 100 Times, GetsCritical BusinessData to Analysts Faster

Taking advantage of the MPP architecture in SQL Server 2008 R2 Parallel Data Warehouse, Hy-Vee has been able to perform data warehouse queries up to 100 times faster than before. “During the proof of concept, basic queries that previously took 20 minutes only took seconds using the SQL Server 2008 R2 Parallel Data Warehouse,” says Settle. “This solution dramatically improves our query performance and nightly processing times.”

As a result, Hy-Vee can get critical store-level data to managers and analysts when they most need it. “Previously, we couldn’t get important data to analysts on time each morning,” says Settle. “But now, with the SQL Server 2008 R2 Parallel Data Warehouse solution, we can process the data faster and consistently deliver that data every morning at 6:00.”

In addition, because the solution processes more data faster, analysts can view purchasing data from the day before. “Using the previous system, analysts were working with data that was two weeks old, so it was difficult for them to react to trends,” says Settle. “Now, they can view yesterday’s sales data each morning. So if we’re in the middle of a promotion for a certain product, analysts can come into the office in the morning and analyze how that item has been selling, and they can order more products if they need to.”

Offers New Item-Level Reporting Capabilities

Using the new solution, Hy-Vee can reload its entire data history into the data warehouse in less than one day. “In the past, it would have taken us weeks to reload the data,” says Settle. “With the faster loading times, users are much more willing to run queries and create reports, because there’s more data to work with.”

Additionally, the solution’s combination of faster querying and new reporting capabilities gives Hy-Vee employees the opportunity to quickly identify and search for new types of data. “With this solution, we can analyze ‘basket-level’ data, which means every single piece of information about a customer’s purchase in a store,” says Settle. “For example, we can look at the average number of items and average dollars a customer spends on a visit, and we can look at the average promotional item dollars spent during that visit. We never had this level of analysis before, and we can measure more relevant data trends as a result.”

Helps Hy-Vee Better Understand Its Customers and Be More Competitive

With faster query response times, analysts can run queries they have never been able to run before. As a result, Hy-Vee can get a better sense of its customers. “With the SQL Server 2008 R2 Parallel Data Warehouse solution, we can now understand what our customers are buying and when they’re buying it, so we can make better decisions on what we’re advertising, what prices we advertise at, and even whether there are items we need to discontinue in the stores,” says Settle.

This kind of knowledge helps Hy-Vee compete better. “The SQL Server 2008 R2 Parallel Data Warehouse solution helps us recognize changes in our customers’ buying habits,” says Settle. “We can then respond to those changes before our competitors do, and that gives us a huge business advantage. In this marketplace, we really need to understand our customers and how they’re changing,” he continues. “Because we can react to their changes faster, we’ll be more competitive.”

*
* With SQL Server 2008 R2 Parallel Data Warehouse, we can easily scale the system as we add new data. *

Tom Settle
Assistant Vice President
Data Warehousing
Hy-Vee

*
Also, because the expanded Hy-Vee data warehouse now contains more than three years of data, the company’s executives can make better decisions. “Previously, we didn’t have the numbers to back up decisions,” says Settle. “We had an inadequate amount of data, and not the right type of data. Now we have the right information, so we can make the best decisions.”

Scales to Meet Future Data Growth

The new solution also gives Hy-Vee the scalability it will need. “With SQL Server 2008 R2 Parallel Data Warehouse, we can easily scale the system as we add new data,” says Settle. “The type and amount of data we collect will grow substantially over the next 10 years, and we’re very confident that this solution will be able to handle that growth.”

This is especially important for Hy-Vee because it plans to expand its data warehouse even further to include data from other areas within the business. “The POS data is only the first piece of this solution. We have four other warehouses that we want to pull information from, including human resources and social media. These areas are growing quickly, and this is the right solution to handle that growth.”

Frees Up IT Resources

Because Hy-Vee managers and analysts can now do self-service reporting, the company’s IT department has been freed up to concentrate on other critical projects. “We have a lot of growth planned for our data warehouse, and we can spend more of our time managing that growth, because our users can do their own ad hoc reporting,” says Settle. “Instead of extracting data for users, we can spend more time bringing the data into the warehouse itself.”

Overall, the new solution gives Hy-Vee the technology to build on for the future. “We need to be on the leading edge of technology,” says Settle. “With SQL Server 2008 R2 Parallel Data Warehouse, we gain that ability, and we are able to better understand our customers and compete. It’s definitely the right solution for us.”

Microsoft Server Product Portfolio

For more information about the Microsoft server product portfolio, go to:
www.microsoft.com/servers

For More Information

For more information about Microsoft products and services, call the Microsoft Sales Information Center at (800) 426-9400. In Canada, call the Microsoft Canada Information Centre at (877) 568-2495. Customers in the United States and Canada who are deaf or hard-of-hearing can reach Microsoft text telephone (TTY/TDD) services at (800) 892-5234. Outside the 50 United States and Canada, please contact your local Microsoft subsidiary. To access information using the World Wide Web, go to:
www.microsoft.com

For more information about Hy-Vee products and services, visit the website at:
www.hy-vee.com

Keyword: PDW

This case study is for informational purposes only. MICROSOFT MAKES NO WARRANTIES, EXPRESS OR IMPLIED, IN THIS SUMMARY.
Solution Overview



Organization Size: 59000 employees

Organization Profile

Hy-Vee, headquartered in West Des Moines, Iowa, is a full-service grocery store chain with 59,000 employees.


Business Situation

To better compete, Hy-Vee wanted to improve its data warehouse performance to give its employees more customer purchasing data faster.


Solution

The company deployed a new solution based on Microsoft SQL Server 2008 R2 Parallel Data Warehouse, SQL Server 2008 R2 Analysis Services, and SQL Server 2008 R2 Reporting Services.


Benefits

  • Boosts query performance by up to 100 times and gets critical business data to analysts faster
  • Offers new item-level reporting capabilities
  • Helps Hy-Vee better understand its customers and be more competitive
  • Scales to meet future data growth
  • Frees up IT resources


Hardware
  • HP

Software and Services
  • Microsoft SQL Server 2008 R2 Enterprise
  • Microsoft SQL Server 2008 R2 Analysis Services
  • Microsoft SQL Server 2008 R2 Parallel Data Warehouse
  • Microsoft Services
  • BMC - Mission Critical
  • Microsoft Technology Center
  • Microsoft SQL Server 2008 R2 Reporting Services

Vertical Industries
Food & Beverage

Country/Region
United States

IT Issue
Data Warehousing

Languages
English

Partner(s)
HP Sogeti

RSS