From Store to Big Data Storage: Interview with George Shaw of RetailNext on Retail Big Data

Photo: George Shaw of RetailNext

George Shaw, Head of Research and Dev at RetailNext

Big data tends to be widely discussed in IT circles, whereas uptake elsewhere has been more gradual. Compared to other sectors, how nuanced is an appreciation for big data in retail?

It’s not as nuanced as it used to be. The RetailNext platform is a “big data” analytical platform, but it’s very much built by retailers for retailers. The last couple of years, we’ve seen a trend in the convergence and integration of data needs from across the retail organization. Loss prevention, marketing, retail operations – all the functions are realizing the benefits of making data-driven decisions, and as such, there is an emphasis on investing in solutions that can be seamlessly integrated and scaled. It’s not an “IT thing” anymore. Rather, it’s a key strategic and tactical decision-making tool across all functions.

Can you discuss the emerging role of mobile geospatial big data for retailers? Are there unique opportunities for “big box” retail, malls, downtown districts and regional shopping districts?

There are some really interesting technologies and movements around “smart cities,” but the caution here is that a “connected community” isn’t necessarily “smart.” There’s got to be an inherent value.

Related to that, there’s a funnel-type of way of looking at smart cities, going from the very broad and macro, like people movement and traffic, down to the very specific and micro, an individual at an office or government building, at a restaurant or club, at a retail business.

Opportunities abound, regardless of the service model, as location analytics is critical for customization and personalization of service offerings. We’ve seen the impact customization and personalization has made in the delivery of online services and experiences. It’s coming to the offline world very rapidly.

We have done an unpublished interview with StreetLine, whose big data sources include street-level sensors and parking analytics to go with 267M “parking events.” How might apps like RetailNext inter-operate with StreetLine and similar applications?

The RetailNext APIs are pretty robust, and they’re designed to seamlessly integrate, inbound and outbound, to systems. In the end, its about collecting and aggregating relevant data streams, synthesizing and analyzing the data, and facilitating better decision-making.

In our space, retail, different clients have different needs. So, a requirement for the RetailNext platform is to be flexible. And, of course, it’s necessary to be scalable as well.

How can retail big data integrate with weather and other external events? Is this complex event processing (CEP) a la Tibco StreamBase, Microsoft Streamsight or IBM Infosphere Streams?

Weather is a big driver related to the performance of stores. A good example was the relentless winter storms that much of the Eastern United States suffered through early in 2014. Bad weather drives down retail traffic, and less traffic often translates to decreased sales.

As such, the RetailNext’s inbound API allows for the inclusion of weather data, just another data stream to help draw a more complete picture. The same inbound API allows integration into staffing systems, too, another highly correlated variable to key store performance metrics.

The more comprehensive and complete the data set, the more accurate the analyses can become, and that results in better decision-making.

What other connected devices in the emerging Internet of Things does RealNext envision being useful to retailers?

It’s a good question, and there’s no question that the Internet of Things will have a big impact on how all businesses interact with customers, retail included. Right now, of course, it’s the mobile phone and tablet that has such an important role in a both a consumer’s shopping journey and in what more and more retailers are doing on the floor with retail sales associates.

There are a lot of interesting scenarios with connected cars and the “smart city,” all of which funnels down to the smart business at the end of the value proposition.

In-store, there are interesting concepts around shopping carts and other ways of moving merchandise, and then, of course, there are the products and SKUs within the products themselves.

What’s going to be important is to collect, analyze, present and report data that is not only actionable, but that supports and drives the retailer’s business model. Data for data’s sake is an unnecessary cost. Data to drive decisions, differentiation, customer acquisition, experience, retention and loyalty, and performance – that’s an investment with a clear and positive ROI.

How might the emergence of in-store analytics (including mobile) interact with the physical layout of stores? Are these possibilities that might partly offset the declining appeal of brick-and-mortar retail?

In-store analytics play a huge role in store design. Through video analytics, Wi-Fi, beacons and Bluetooth Low Energy (BLE) and other methods, we can measure and map how customers go through the store. Plus, we can use historical data to simulate traffic patterns that would result from changing display configurations and store layout.

Analytics also help retailers and even manufacturers measure the effectiveness of displays and merchandising, measuring traffic flow, engagement with the display and the length of time at the display and, with POS data, the effectiveness of conversion.

Lastly, A/B testing is extremely popular, allowing a retailer to experiment and determine the most effective means of providing product, service or both.

Do you see a threat from giant internet firm initiatives, such as Google Glass or Facebook Oculus VR? They seem to want to control and own emerging data streams.

Personally, I look at giant firms like Google and Facebook as opportunities more than threats, and those data streams are increasingly important. RetailNext currently works with retailers in understanding how their shoppers are interacting with search engines, social sites, their own site and competitors’ sites while in-store. And more and more often, we’re seeing shoppers engage their social networks in a sort of real-time polling asking whether they should buy a garment, a pair of shoes or whatever. Stores are actually using technologies to help empower shoppers to do just that.

Can the analytics in RetailNext be applied to virtual environments like multiplayer games, Second Life, Active Worlds and others – for simulation, staff training, retail layout, product placement negotiation with retailers and product design?

It depends. The analytics can be presented in a variety of ways, in a variety of environments. A very important consideration is change management. Sometimes, just the data and the analytics alone face an acceptance hurdle because it’s such a drastic change to the “ways things are done around here” at a retailer. Remember, at many retailers, the business of retail has been more of an “art” than “science” from the very beginning.

At RetailNext, we provide some clients with predictive analytics based on simulations of how a store redesign, for example, will impact the way traffic flows through the store and how customer engagement will change at floor and wall displays. To this point in time, the communication of those simulations have been folded into the traditional communication and training mechanisms already in place at the client.

Are you concerned about privacy and security pushback from consumers? For some, retail big data may be their first practical window into retail data utilization.

Consumer and shopper privacy is a concern in the industry and with retailers, as it should be. It’s important to note that RetailNext provides data analytics in the aggregate, and individual shoppers cannot be individually identified through passive detection devices.

There’s also active systems, like opt-in Wi-Fi, that allows retailers to engage in two-way dialogue with their shoppers, oftentimes through loyalty programs and the like. Opt-in programs will be the key to delivering personalization at physical stores.

Retailers recognize the issue of privacy and how important it is to many shoppers, and they are designing service and experience models around those issues. It’s in the long-term best interest of retailers to approach service and experience in a way that builds trust with their shoppers and helps them become retailers of choice.

What standards – de facto or emerging – are you most keenly following?

I’m keeping my eyes on a lot of innovations and the potential applications for retailers. In addition, a great deal of our funding goes to internal R & D, where we’re constantly looking to push out our technology advantage and differentiation even further.

Regarding the implementation of technology to existing business opportunities, there are a couple of companies, including Tyco, who are developing RFID technologies for real-time inventory control. It’s a nifty bit of technology, and it’s a wonderful example of marrying technology to retail operations in a way that has demonstrable benefits to both retailers and their shoppers.

Another very interesting technology I’m keen on is Sociometric Solutions’s social sensing technology that generates data on social interactions and engagement. The data developed by the interactions between sales associates and shoppers has the potential to dramatically impact how shoppers and sales associates relate to one another, and potentially revolutionize a new in-store customer experience.

Those are just two of the innovations that I’m following closely right now. There are a lot of interesting new technologies being developed, as well as new applications for existing technologies, so it’s an exciting time in the space.

✓ From Stores to Store: See how Syncsort tools can integrate retail data into big data storage.

Mark Underwood

Authored by Mark Underwood

Syncsort contributor Mark Underwood writes about knowledge engineering, Big Data security and privacy.

Leave a Comment