Retailers must centralize their data to thrive

Discussion
Photo: Getty Images/Wavebreakmedia
Jun 18, 2021

Successful retailers know that every decision — from workforce planning to space and assortment management to inventory optimization — requires a foundation of high-quality data. If granular data about stores is not made available to the central planning team, however, retailers will be missing critical information on which to base both chain-wide and localized decisions. Worse, if that data is inaccurate or inaccessible, forecasts will be faulty, leading not only to incorrect inventory management but also insufficient staffing and ineffective store layouts.

Although most of today’s retailers have at least a basic understanding of the need to collect and centralize their local data, there are those that either do not have the resources or have not invested in the technology to make it happen. This has resulted in a two-tier playing field of retail development:

  • Retailers investing in technology and concentrating on centralizing and really using their data to see significant growth. They are able to execute their business strategy across all stores while improving both customer service and inventory management.
  • Others are still relying on individual store managers to make local decisions using disconnected spreadsheets (or even pen and paper) and wasting valuable local staff time, straying from their brand strategy and missing opportunities to increase efficiency.

When business decisions are made in a disconnected manner at different points in the supply chain, there are going to be inconsistencies. This can lead to major inefficiencies, overstocks or out-of-stocks, end-of-season residual stock and lost sales. When retailers bring together the data from each store and all online channels into a single AI-driven forecasting solution, their centralized planning teams are able to review and streamline the full supply chain from end-to-end.

It is more important than ever that retailers, as they move to omnichannel models, collect data from all channels and use it to manage the flow of goods through the supply chain to the point of customer interaction. This not only allows for more consistent decision making but improves workforce management and space planning. Retailers that are unwilling or unable to centralize their data will continue to waste time and money while struggling to keep up with fluctuating demand — and they are going to be left behind.

DISCUSSION QUESTIONS: How critical is having centralized data to the ongoing success of meeting customer demand? Has the COVID pandemic accentuated the importance of having the right data in the right hands?

Please practice The RetailWire Golden Rule when submitting your comments.
Braintrust
"We’ve been preaching 'one version of the truth' for data in retail forever. There is zero justification for any other model but centralized data."

Join the Discussion!

22 Comments on "Retailers must centralize their data to thrive"


Sort by:   newest | oldest | most voted
David Naumann
BrainTrust

Having centralized, accurate inventory data is imperative to maximize inventory profitability and customer satisfaction. Enterprise real-time inventory visibility enables retailers to make smarter fulfillment decisions for online orders based on cost of delivery and margin optimization. Inventory visibility is also critical to ensure product availability for online orders and if the inventory is extremely accurate, retailers can eliminate safety stock for available to promise inventory.

Paula Rosenblum
BrainTrust

I don’t understand how this is even a question in 2021. Of course data should be centralized. And where is it not, and why not?

Liza Amlani
BrainTrust

I have worked with many large global retailers where data is not centralized and is used by parts of the business where others have no access to it all. This is happening today with design driven brands that should be relying MORE on data. Insights driven from data are only used by marketing and not for product creation and merchandising. Things are slowly changing but definitely not fast enough.

Ken Morris
BrainTrust

We’ve been preaching “one version of the truth” for data in retail forever. There is zero justification for any other model but centralized data. The Internet of Things (IoT) for retail should include RFID, electronic shelf tags, heat/cold sensors, digital signage—everything. And it should all be both feeding and listening to info from the centralized database. The centralized data model supports omnichannel retail in every aspect. Without it, customers, suppliers, store managers, and HQ are just guessing. Realtime centralized retail has to be embraced to support the customer journey of today.

Michael Terpkosh
BrainTrust

It is extremely critical for retailers to “connect the dots” between their data sources. For too long, retailers have allowed different management functions (merchandising, marketing, operations, etc.) to use their data in their management silos. The result is decisions made for the good of the silo, not for the holistic good of the business and the customer. Only when data and management silos are broken down will there be truly insightful data understanding and decisions. Retailers miss many great insights into the business and customers when they don’t connect the data dots. This can be the difference between continued business success and failure.

Kai Clarke
BrainTrust

The better question here might be, how can retailers not centralize their data and still maximize their growth? The requirements of logistics, inventory management, and any other data required to optimize a retail model feeds on disparate data from all available sources.

DeAnn Campbell
BrainTrust

Centralized data is the lifeblood of retail, how else can companies execute the omnichannel strategies that consumers demand? How else can brands and retailers benefit from pop-ups, manage brand partnerships, or understand what impacts profit margins most? The bigger challenge will be redefining their key performance indicators to reveal how each channel impacts the others, and re-engineering business models accordingly.

Jennifer Bartashus
BrainTrust

In an era of big data the importance of accessible and integrated data sets is critical. Centralization of data is absolutely necessary to remain competitive, and the need for data transparency through the entire supply chain was one of the takeaways from the pandemic. From sourcing to stocking shelves, all parts of an organization need to have access to relevant data, which leads to efficiency and ultimately lower cost to serve.

Oliver Guy
BrainTrust

There is no doubt that COVID-19 has highlighted just how siloed retailers can be when it comes to their processes, systems and data. Meeting the ever increasing expectations of consumers in terms of providing a harmonized retail experience requires the free-flow of data. Data related to inventory, product, customer and orders is all critical.

Market discussion relating to supply chain control towers leading toward an autonomous supply chain is futile until data silos are eliminated. Such initiatives are 100 percent reliant on the free flow of accurate, precise and timely data associated with inventory, orders, product and customers.

Dave Wendland
BrainTrust

Data can be either the enabler or the crippler. For retailers (and/or brands) to be as effective as possible, the single source of truth and centralized data is paramount.

Suresh Chaganti
BrainTrust

A single view of the data is critical. But physical centralization is not the only way. The critical aspect is to assure consistency and quality of data capture. If pen and paper are used, replace them with tablets. If free text fields are used, minimize them and use drop downs or a list of values.

The key is you must make the data usable before you can make it useful.

One thing for the practitioners to watch out for is the “boiling the ocean” effect. Centralization projects sound good in theory, but the constraints are not just technical. Who owns the data, who has the right to update the data? Who needs to be consulted or just informed? In short, the whole data governance dictates the success of data projects.

Matthew Pavich
BrainTrust

Having robust, centralized, accurate data is absolutely critical for retailers in today’s retail landscape. Every aspect of business including supply chains, operations, pricing, assortment planning and loyalty programs rely on a robust foundation of data. The more data that is available – the more levers a retailer can pull to achieve strategic objectives and the more AI can accelerate that success with better forecasting and results.

Mohamed Amer
BrainTrust

Amazon Web Services S3 and Microsoft Azure Data Lake Storage have blurred the centralized versus decentralizing data question. However there is no one size fits all solution. Data will continue to be centralized and decentralized based on the use cases. The data explosion in mobile retailing, 5G networks, and the Internet of Things (IoT) will increase the need to get more analysis and decisions at the edge, even for machine learning that is typically centralized. Federated data techniques for analytics and AI will also continue to bring value in situations where data cannot be centralized. A corollary question here is how you should organize your analytics team- centralized, decentralized, or hybrid.

Cathy Hotka
BrainTrust

The topics in today’s instant poll are real, but are also fixable. The retail companies that win are working on them right now.

Mark Heckman
BrainTrust
Certainly, having one version of the shopper across all shopping channels is the gold standard and the necessary goal for cogent analysis, decision making and shopper communications. One of several reasons why this condition does not exist, as intuitive as it may be, is retailers often have external partners and these partners manage a slice of their shopper data, not all. So data and reporting comes from disparate sources. Another scenario is one in which stores or regions manage their own shopper data and marketing and merchandising decisions and they have never been made centrally, but rather locally, so there has been no urgency for centralization. Consolidating the data is easy enough from a process standpoint, but connecting and updating multiple sources of shopper data can get gnarly and expensive if done ad hoc. The solution lies in a commitment from the retailer leadership to have a plan to create a real-time updating process that collects and organizes the data centrally. As others have posited, in the omnichannel world we live in, not having a… Read more »
Trevor Sumner
BrainTrust

Is this really a question? Yes. Centralized. Organized in the cloud. Available to other systems via API. Exposed to the web in real-time. AI powered data insights. Welcome to the now of retail.

James Tenser
BrainTrust

When dealing with decisions based on aggregate consumer demand, it makes intuitive sense to centralize data. But store level merchandising depends on granular analysis, not averages or chain-wide norms. That means all ordering and many promotion decisions should be made and implemented at the “edge” — as close as possible to actual shoppers. Each store must be uniquely optimized.

This is not an argument against a “single version of the truth” which remains an essential goal for retail managers. Of course data integrity and democratized access across the organization are critical necessities.

Pooling all an organization’s data into a single lake can well enable this. Care must be taken, however, to preserve the specificity that enables valid store-level and sometimes even shopper-level decisions.

Certainly retailers must tear down silos and provide mechanisms for decision makers to access all data relevant to their missions. This may be accomplished via an AI-enabled query system, even if certain data remain in their existing repositories.

Brian Cluster
BrainTrust
It’s an absolute necessity now. The pace of business in retail accelerated in the past year, by having one single consolidated view of suppliers, products, locations, and even employees; retailers have everyone running on the same version of the data — making it a more efficient organization. Product data and customer data are the most common types of data that need to be centralized but other types of data such as supplier and location data are also important. From a supplier perspective, different departments need to be on the same page to understand contact information, terms, ownership relations, and much more. Having one single source of supplier data can aid in approving and onboarding new suppliers and eliminating any delays in going to market. From a location/site perspective, there is a lot of changes that have happened in the marketplace in terms of delivery options (BOPIS/BORIS/Curbside) and fulfillment capabilities as more store locations are used for local eCommerce fulfillment. As these changes have been tested and implemented in the past year, it is absolutely critical… Read more »
Peter Charness
BrainTrust

Yes data needs to be centralized and organized in a quality fashion, but it also has to be accessible and useful. If the rules, access and consumption tools of these central stores are not conducive to end user needs, then the organization owns a Data Prison, and not a useful resource.

Gib Bassett
BrainTrust
Even being a data warehouse and platform advocate, I can’t help but think it best to respond in this manner, especially in the era of cloud and the ability to scale up or down based on your requirements. Of course there are many elements critical to the needs of meeting customer demand, and having data, insights and operations aligned to anticipating and serving customers is among the most important. Retailers and all industries have been chasing the value behind centralized data for a very long time, but only those able to pair technology investment with greater organizational and process changes have been able to succeed. That’s because new insights and prescribed actions fall down when layered into faulty process models and old norms. Few retailers have appetites for large transformational data projects with uncertain benefits and timelines, so point solutions remain the status quo for many. These offer some value but won’t likely lead to best-in-class results and organizational change takes time and is difficult. Best advice would be to lead with goals for improvement… Read more »
Brent Biddulph
BrainTrust

Previous definitions of “centralized data” used to mean a retailer had to make a significant investment (including armies of IT folks) into an Enterprise Data Warehouse (EDW). Not any more. Rather, that’s not even good enough and cost prohibitive for retail leaders.

Instead, leveraging new technologies (e.g. cloud infrastructure, IoT edge devices and real-time data movement) to capture new data sources, blend with legacy EDWs and operational data stores is now required to enable the data science (predictive analytics, AI, robotics) to compete effectively, cost efficiently and with agility.

It’s not good enough to “centralize data” in a single physical device like an EDW anymore, rather “centralized data access” with streaming capabilities in an “enterprise data ecosystem” supporting a myriad of legacy and emerging business use cases.

That’s the new Chasm.

John Orr
BrainTrust

Retailers have always had the need for speed and visibility of their people and business. The issue has been that many of the ERP, Talent, workforce management, and payroll providers have done a dis-service to retailers by offering disparate, acquired, merged, fragmented systems cobbled together that didn’t support what retailers need. Those times changed in 2010 and today Gartner’s magic quadrant, Nucleus research, and others have noticed. Streamlining and consolidation of systems provides one version of the truth and in real-time. The derivative value is multiple in offering more options for employees, improved accuracy, and visibility across all the domains of HCM.

wpDiscuz
Braintrust
"We’ve been preaching 'one version of the truth' for data in retail forever. There is zero justification for any other model but centralized data."

Take Our Instant Poll

What retail challenge does centralized data have the biggest impact on?

View Results

Loading ... Loading ...