Off the Shelf provides a platform for Retailers and Consumer Packaged Goods companies to discuss and gain insights on the pressing problems, trends and solutions.

Main

December 18, 2014

Digital wallet and self-check-out - are they the answer to shorter check-out queues.

Market Size projections for Mobile Wallet Range from $41 B (per Forrester Report in 2013) to $64B (emarketer report in 2014) in US by 2017.  There is an even more impressive report card for Global Mobile Wallet market with estimates ranging from $ 191 B for 2017 (ABI Research for NFC Mobile payments) to $ 1602 B (Transparency Market Research, Oct 2014) by 2018 with a CAGR of 31% from 2012-18.

As we debate the correctness of various estimates, the writing is clear - these are well and truly exciting times for Mobile Payments in general and Digital Wallets in particular.

While there is a huge potential to transform the payment space, even at current levels of maturity the market is crowded and fragmented (at least in the US) with many players aiming to take a slice of the pie - see image below on who the players are and what are the various types of offering  (US Context)

DigitalWallet Ecosystem.jpg























Image1: Overview of Different Types of Wallets and Players


One would imagine that with the plethora of options,the awareness and adoption rate will be quite high, however the reality is quite contradictory - there is a high awareness (~ 80% in US) but a dismal adoption rate (< 20% in US) (src: Yankee Group 2014 report)

So what is really the problem - does the wallet not provide the convenience or benefits? Are consumers worried about security and data privacy?

Let's now consider the benefits of a Digital Wallet. Many proponents of the digital wallet will argue that usage of wallet results in shorter check out queues. While there are really no proven data points to quantify the actual saving, on the contrary there are some examples where different digital wallets have failed miserably on usage and performance that they had to be discontinued eventually ex: Starbucks Failed Square Wallet Rollout in 2012. Also, more importantly there's nothing particularly inconvenient about traditional card payments at POS / checkout. Given this kind of a scenario, the Mobile Wallet as a replacement of Credit Card with marginal improvement in checkout times will not be a great business case.

So now let's look at critical success factors that will help in adoption and in providing better benefits to customers. These factors fall under 3 different heads:

1. Superior Customer Centric Design -Foundation of any good mobile wallet offering
2. Address Customer and Merchant Concerns

Key Concerns.jpg






Image2: Key Concerns


3. Provide Value Adds

Value Adds.jpg







Image3: Key Value Added Functionality


In short rather than limiting the benefits of the wallet to just faster checkout times (yes this is absolutely needed but it is the bare minimum), the focus has to be more on looking at the digital wallet as a customer experience enhancement platform with built in components to address concerns and provide value adds so as to enable smooth payments by the wave or tap of a phone.

The success of any digital wallet program will depend on the extent to which it scales across the 3 dimensions of design, handling concerns and providing value adds to customer (see image below)

Attributes of a Good Wallet.jpg
















Image4: Attributes of a Good Digital Wallet Solution

June 3, 2014

EDI Transforming Business Models

For most of us, EDI is a communication system to exchange formatted data between disparate applications, however it is time to realize that the technology has potential to deliver much beyond than what it was intended to. Since early 20th century commerce had seen industries leveraging the opportunities provided by Express mail and Telefax to grow network. Healthy and far-reached network implies smart way of handling goods and providing services, optimization of cost and labor, efficient processes and customer delight. Business growth, technological advancement, and ever changing customer needs demanded for more swift and complex information exchanging networks.

The growing concern was addressed by UN recommended EDIFACT and ANSI recommended X12 EDI standards that set to establish uniformity in business transactions defined for each industry or operation segment. The principal vision was to establish standardization and integrity of the business documents being exchanged amongst trading partners. However EDI proved to be a state of art technology in many areas starting from finance to processes. Following is an attempt to understand the way EDI has contributed to the evolution of many processes, creation of opportunities, and efficient management of resources in a supply chain.

 

Continue reading "EDI Transforming Business Models" »

May 12, 2014

Power of Crowdsourcing

Crowd sourcing is a fairly new phenomenon that has shaken the basics of various industries, business models and enterprise decisions. Right from getting inputs to product designs to utilizing pictures from Instagram for running marketing campaigns this practice of leveraging collective intelligence is now becoming  main-stream.

With the success of this practice manifesting in many walks of life, we can take a look at some of the examples to see how well this model has been adapted.

a) Quirky - Adopted Crowd sourcing into their Product Development process and called it Quirky Social Product Development. Individuals submit product ideas which is then curated by the community, reviewed by a product evaluation team before its designed, engineered and mass produced. A percentage of resulting sales from the product is then shared with the person who proposed the product idea. With projected revenue of 54 Million USD by 2014, Quirky is quietly becoming the Mass Market consumer product manufacturer.

b) Wal-Mart: In 2011-12, Wal-Mart started with the Get on the shelf initiative, where it invited public to submit great product ideas and then through a review process and thrown open for voting. Finalists were chosen based on the votes and rating and their products were available online and in stores. Human Kind in 2012 and Elvis Presley Home Bedding Collection in 2013 were chosen as winners (see https://getontheshelf.walmart.com)

c) InstaCart:  Grocery service provider that leverages the power of crowd sourcing to delivery grocery to homes within an hour of order getting placed. Company began operations in San Francisco but has rapidly scaled up to cover cities like Boston, Chicago, NYC and has also expanded partnership with retailers like Whole Foods, Shaw's, Market Basket, Harvest Coop and more. With revenues rising more than 2X times in the last few months, there is a great growth story for the company.

There are numerous other examples of L'oreal, Coach and other fashion retailers seizing this opportunity to put together marketing campaigns and/or product images that borrow on photos and amateur images from social media sites like Instagram and Pinterest.

Real benefits in Crowd Sourcing? Clearly the benefits are not very hard to see

  • For brands that weave in crowd sourced images for product display or marketing the benefit is plain and simple - imaginative, original and authentic
  • Crowd sourcing helps disrupt the existing business model and creates new opportunities for startups.  A classic example is that of Instacart that has successfully launched same hour Delivery for Grocery orders where larger players were struggling with same day deliveries.
  • For companies like Quirky, it is about accelerating innovation and cutting overheads and operating costs associated with typical New product Introductions.
  • For Walmart it's about finding the next Billion dollar of sales by identifying those high potential new products and getting it on the shelf
  • For players like Utest it is unlimited access to an expanded set of skills and sometimes hard to find skillsets.

Problem Areas: While we certainly want to focus on the benefits and the key drivers of this model, there are few problem areas that needs to be addressed as well

  • The first and foremost would be that of Intellectual Property. Who owns the IP in case the idea is transferred from the founder to companies like Quirky or Kickstarter. Can there be non-exclusive license transfer instead of completely transferring the IP rights?
  • How to sufficiently compensate the founders? Is it a onetime lump sum or gain sharing or a commission on sale?
  • How do we ensure quality of product or service? Take the case of Instacart - for some orders if items are not found in a store then those items might not be fulfilled leading to lower customer experience.
  • Lack of Confidentiality
  • Lack of Proper Communication

Key Considerations: Finally for retailers, service providers and manufacturers trying to leverage the power of Crowd Sourcing the following key points needs to be considered to ensure a very effective process and resultant output

  1. A rating mechanism for the participants in Crowd Sourcing to identify their significance of past and present contributions, social influence, originality of design or ideas. Ex: In case of Instacart rating parameters could include order fill rate and on time delivery by personal shopper. Higher the rating, the better the compensation or assignment for future orders.
  2. Search Mechanism (using an extensive algorithm) for systematically sifting through enormous amount of data, photos, videos, ideas and products and also considering the rating to identify those opportunities that aid future growth, address under served  customers or geographies. 
  3. A workflow driven technology platform that will help the collaborative nature of work right from submission of ideas to reviewing, evaluating, allowing public to register vote and eventually identifying the top picks.
  4. A palatable profit sharing mechanism to reward the Founders (of idea) based on the outcome - sales of the product or service.
  5. Re-defining the various roles in the Internal Organization - Product Incubation, Marketing or service teams to adapt to the new model and to deliver creative outputs (campaign, service, product) both from within and outside the organization. 
Clearly these are interesting times for the various participants in this model - be it the contributors/founders of ideas, new age companies that aid to translate these ideas to sellable products or the retailers that borrow upon the end products for final sale. With more funding and with proven case studies on sustainability and success this model will become a huge source of disruption to as-is business models in the coming years. As they say, the collective intelligence of community (local or global) can very easy surpass the intelligence of corporate giants.

April 8, 2014

To Pause (POS) or Not

In a recent study done by University of Arizona and Demandware it was highlighted that Traditional POS systems at Brick and Mortar stores are not able to keep pace with the expectations of digital consumers and the retailer's response in the form of omni-channel strategies.

The reasons are not hard to see; today's POS systems are no better than siloed systems that have stores as boundaries and cannot look outside to support many of the retailer's omni-channel aspirations. Thus the tools and software available as part of traditional POS are falling short of supporting some of the key initiatives like

·        Single view of customer (or the 360o customer view),
·        Cross channel inventory visibility
·        Looking up inventory in nearby stores or
·        Assisted Selling and Product personalization
·        Personalized Offers

Also added to that these traditional POS devices have a cyclical re-haul  which is to happen in the next 2-3 years -perfectly coinciding with the thinking of several CIOs to see if they should replace the Traditional POS systems.

So what are the alternatives that CIOs or Leaders have -

Option1: Extension of current POS to bridge the Gap

One easy way out for retailers is a natural extension of current capabilities and investments to support future needs. Some POS vendors like NCR, Epicor and Oracle already offer capabilities like Assisted Selling, targeted up-sell and cross-sell and mobile POS.

This option is characterized by (View image) leveraging existing investments to a large extent, lower cost, Faster Time to market, Incremental Change management and clearly Short Term Focus.


Option2: Overhaul of the current system in favor of a common customer transaction suite

A compelling alternative would be that of extending the best in class digital commerce architecture to support store operations. This solution will help connect all channels and implement an integrated solution that will provide a common view of customers, products, inventory, price, promotion and order information. This will also help address other in store opportunities like guided selling, and locate nearby store inventory to save the sale.

This approach is generally characterized (View imageby a Flexible Architecture that can scale to support future needs, Significant Initial Investment, a Big Bang Approach to replace existing POS systems, Training and change management initiatives that are needed to support such a big bang and finally has a Long Term Strategic focus.

In the not so distant future we might as well hear from the best in class digital commerce platform vendors like hybris, IBM WCS, ATG and Demandware about the emergence of a new class of enterprise software - we can call it the 'Customer Experience Management Suite' or 'Integrated Customer Transaction Platform' covering all customer touch points (from stores, desktop, mobile, tablet and Call Center) and providing truly seamless experience. 

However replacing the Traditional POS systems cannot happen overnight nor is it feasible. Some of the broad questions below need to be answered before we can arrive at meaningful answers and way forward


  1. Risk perception of Retailers: What if it means putting all the eggs in a single basket for retailers and if they perceive a higher risk in doing that then that's a huge impediment to deal with.
  2.  Complement or Supersede - Given that traditional POS is entrenched and is a proven model, should retailers look to complement investments in Traditional POS with that from the new customer platform?
  3. Applicability for all types of Retail: Will the new POS platform be suited for all types of retail businesses? Definitely the opinion seems to be divided here - will it be more suited for High value transaction businesses compared to high volume transactions
  4. Security Concerns:  The security and PCI Compliance for traditional retail POS further augmented by Video Surveillance puts it ahead of Mobile POS and make the current offering fairly robust with very few incidence of violations. Unless the new platform is able to match up the security aspects of incumbent it is going to be a hard bargain.
  5. Reliability: Given that the new Mobile POS will run on Wireless network, questions around matching the reliability and availability of existing POS system needs to be addressed
  6. Tenders: Again Traditional POS is designed to accept multiple tender types like Cash and card at a bare minimum. Support for multiple types of tenders needs to be built in
  7. Purpose of Use - Self Service or Assisted: How should mobile POS (based on digital commerce architecture) be used in stores?  Should it be left for self-service as standalone kiosk for customers to checkout and complete payment or should it be used by the Store Associate as assisted selling tool? Or should it be a combination thereof?
  8. Business Process Re-Design / Standardization: Right from re-defining the roles of cashiers and store Associates to designing new business process ranging from providing common inventory visibility, uniform chain level pricing, common promotion definition & delivery, Guided selling, Deliver to home and many more - there is a whole lot of Business process and Role alignment that needs to be thought through and designed.
  9. Training needs: Identification of Change management and Training aspects needs to be done upfront. This will help address various aspects like for example, in an assisted selling scenario (using a new mobile POS platform) determine when exactly to approach the customer to assist and close the transaction? Approaching too early might result in wasted opportunity as customer might have other items or approaching too late will mean wasted opportunity. The roles of cashiers needs to be re-evaluated- How to reskill the current army of cashiers and deploy them as trained store associates to help in clientiling or assisted selling
  10. Choice of Technology: Several options exists for retailer that needs a very careful consideration
      1. Platform Options
        • Extension of current Platform with value added use cases
        • Re-haul of existing platform in favor of Table POS
        • Re-haul of existing platform in favor of an integrated Customer transaction platform driven by digital commerce architecture
      2. Other considerations:
        • SAAS vs On-premises
        • Single tier vs n-tier
        • Rich Website like UX vs traditional POS like UX / features
        • Global Platform vs Localization Needs (store specific, Brand specific)
        • Point to point integration vs Open Standards based Integration with ERPs

It is now fairly evident that it needs a long drawn process where the pros and cons need to be weighed in before a final decision. For the near future we can safely assume that new POS (Integrated Customer Transaction Platform) at best might at co-exist with the enhanced versions of Traditional POS and not end up replacing it altogether.

However in the long run once the benefits from the new systems are realized, when the risk-reward equation has been hashed out and after enough pilots validate/fail to validate the hypothesis and long after store associates are adequately trained and when Mobile Payments become main-stream - we can for sure say that the new POS platform (with much lower incremental cost) will have sufficiently replaced the existing ones.

In my view we are 5 years from that point of inflection, but that's good enough a time for the likes of established Traditional POS players like Epicor, Oracle, NCR SAP, Microsoft and Retalix to take rearguard action and have a compelling solution offering to stifle progress of Digital Technologies in their own turf. Whichever way the pendulum swings, it's a very interesting space to observe and act in the next 18-24 months as it is bound to have far reaching impact on the future and landscape of POS technologies and solution offerings.

September 30, 2012

"Mobile Apps For Faster Shopping Sprees"

 "Google Indoor Maps" has opened a whole new opportunity to retailers whereby they can enable their customers in getting easy and quick access to the products that they're looking for. In brief, "Google Indoor Maps" allows a person walking inside a store to navigate his travel within this indoor location just like what she(/he) would have done while driving a car using a GPS

Integrating these maps with Retailer's own mobile application (app) and tagging a store's aisles by "Product Categories" on these Google Indoor Maps for a large format store, going down to category of products available by Brands may be just the beginning of the thought as to how Retailers allow their patrons to directly reach out for the product that they are looking for rather than wandering through store departments or searching for a store associate which for some shoppers may be time consuming or even frustrating.

One of the best supporting feature on Google floor maps, is the ability to guide the user by individual floor's plans, whereby a large multi-level retail stores can also be covered easily and therefore we feel that some of the immediate exploits can be in the large scale Store or say Super Store

Let us try to understand how one such Retailer app may make the life a Retailer's clientele much simpler and the shopping experience much better. Assume the scenario that our loyal Customer "Louise" is entering such a Large Super Store for her weekly purchases which runs across several departments of the store like fresh produce, dry grocery, apparel, cleaning supplies, bath-ware, electronics, sports goods and the list goes on. Soon after she has parked at the store, Louise logs-in to the Retailer's own native Mobile app on her smart phone. The Retailer's app upon invocation on her smart phone, detects her geographical location and the store that she is visiting today via Location Based Services (LBS) wherein this specific store's latest tagged maps can be pulled and displayed on Louise's phone's screen

To make the whole shopping trip faster, Louise has keyed in the shopping-list beforehand in the app and a route map is prepared for her upon her check-in into the store via this app. This route map is based on the latest movement of shelves/racks in the store.  Such a guided walk cuts down the Louise's walk through the aisles a short, easy and a confortable one

Louise was looking for a shirt for her son, but the size small does not appear on the shelf today... does the store have it? No problem... the check would be a quick one by Louise quickly getting to know this via her mobile app. Moreover, if it is not available in store right now, the app prompts her with an easy and quick site to store order which she can pick up during her trip next week 

Another use for this app+google maps eco-system can be to integrate with the floor maps and publish current vacancies/next available time slots/expected wait times in Large Store's sub stores like ophthalmologist shops, saloons etc.

 

This Large Super Store's sub stores are very frequently publishing a status of a vacant customer spots/seat available or unavailability of the same on the floor map which when viewed by Louise, will give her an idea whether she needs to do the shopping first or go to the sub store for a quick visit to the hair salon

While these are just some of the initial thoughts, when pursued actively this specific technology can be utilized in umpteen ways to boost the store sales and to guarantee customer satisfaction. Overall, sky is the limit when one starts documenting the concept of such a solution/product. Customer purchase/return history and loyalty points can be utilized to highlight offers/deals when customer is approaching a specific aisle or when she has been looking for a specific product for some time

This article has been contributed by Ashutosh Kaushal - Senior Consultant (Sterling Commerce - Infosys Ltd). You can reach Ashutosh at Ashutosh_Kaushal@infosys.com.

 

Leveraging Social / Consumer Genome for Merchandising

Most retailers traditionally leverage sales data from POS terminals to analyze buying behavior. In some cases, loyalty card data is also used to determine appropriate assortment decisions. These data sources and their corresponding analysis have proven reasonably helpful, though they don't convey the whole picture. With the recent explosion in social and consumer related data on the web, there is a wealth of information that Retailers should be exploiting and incorporating into their merchandising decisions, primarily around assortment and space.

Social/Consumer genome related data provides rich insights into needs, wants and buying behavior of individuals. This data when combined with demographic and geographic data can provide a good map of consumer wants and needs for a given market for a set of product categories.

A major input into merchandising decisions / assortment plans is to determine consumer buying behavior to identify what products sell and what potential products could sell to increase sales. Based on this analysis, assortment decisions of inclusion/exclusion or allotment of space are provided. The means to identify this buying behavior was typically the use of POS data or data from Nielsen/IRI that provided good basis for WHAT was being purchased. When married with Demographic data, there was a good proxy for WHY the products were being purchased. Even when rigorous correlation and clustering analysis is carried out, the determination of buying behavior and the reasons for the same were proxies at best.

Now with the availability of consumer genome or social genome information, the analysis of WHY purchases are being made and what is being purchased with identification of latent and express needs becomes even more accurate as there is clear expression of wants and needs. This will significantly enhance the quality of assortment and merchandising decisions as the degree of error/approximation is reduced.

There are however some pitfalls to the use of this data. We cannot solely rely on this data as web usage and consumer/social genome information may not fully represent buying needs and wants for the entire market population. This data usage has to be married to traditional sales analysis to augment the decision making process.

There are no tools / application products in the market place that provide truly integrated capabilities. The holy grail for optimized merchandising would be to integrate social/consumer genome data effectively into the traditional clustering analysis and thereby into the assortment planning process. This might be a challenge for some of the retailers who struggle with traditional approaches. Asking them to adopt more advanced analytical approaches to incorporate social/consumer genome data would be a challenge.

The key would be to devise suitable technology/process platform augmented with robust analytics shared services that can leverage necessary data to enable optimized merchandising decisions.

This article has been contributed by Amitabh Mudaliar (Group Engagement Manager - RCL Infosys). You can reach Amitabh at Amitabh_M@infosys.com.

Cashing in on "Order to Cash" through BPO shared services

 The Order to Cash process is at the heart of every business and I have often been tasked with asking clients and prospects to consider moving their Order to Cash processes into a Shared services delivery model with a third party service provider.  Over the years this attempt of mine has evinced myriad expressions from clients ranging from the "Have you lost it?" look to the "You just showed me business process nirvana" look. The industry where I experienced the most contrasting reactions was the CPG industry and I wanted to share some interesting learning from the experience of working with a few clients in walking the talk and sharing in their Order to Cash transformation journey over the years.

To provide some context, the concept of Business process outsourcing in the CPG industry is not a new one. This industry was one of the earliest adopters of outsourcing transactional and back office business processes such as Accounts Payables, Accounts Receivables, Payroll processing and Procurement. The primary driver for outsourcing across all these processes was labor cost arbitrage followed by process efficiencies from consolidation; harmonization and continuous improvement. These drivers have become the cornerstone of all business process outsourcing relationships and have for long settled into becoming the marketing tagline for the BPO industry.

With this legacy of perceived value from BPO relationships, when one has to go down the same journey with Order to Cash processes, the value equation does not necessary stack up. The most obvious reason for this skew in the case of O2C processes is Risk. To cite an instance provided by one of my clients - The entire annual cost savings delivered by a BPO relationship can potentially be eroded by one order from a large retailer being inadvertently dropped or shipped to the wrong address due to the contractual penalties involved. And one defect out of a million is acceptable even by "Six Sigma" standards!

So two questions come up, the "Why" and the "How"? The "Why" is easy. Given that the Order to Cash process impacts not just the G&A metric but also the Cost of Goods Sold, Working capital efficiency, Days Sales of Inventory and Gross Revenue metrics, a successful shared services strategy undertaken by engaging the right partner could catalyze business transformation impacting the highest levels of the organization.

The tough question however is the "How"? I often relate that to converting an "Art" into a "Science". The "Art" here refers to the finesse and proficiency that Customer service representatives have honed on their jobs for a number of years. The "Science" here refers to replicating that same level of service or sometimes even improving it but with a completely new team that's located thousands of miles away in a global delivery model.  Sounds like mission impossible right? Not really. It's been done before and involves the following best practices:

·         Create the right operating model: The tried and tested Front Office-Back Office model is a great starting point. It allows for a risk mitigated approach to the end state sourcing mix given the need to balance risk and value in a customer centric setup for the CPG industry.

·         Create the right performance measurement methodology: The days of adopting operational metrics such as cycle time, average handling time, Abandonment rate, First Contact Resolution etc to govern outsourcing relationships are a thing of the past. These metrics are now considered Business As Usual (BAU). Consider including business metrics and supply chain performance metrics such as cost per order/invoice, Case Fill Rates (CFRs), On Time Delivery (OTD), Truckload capacity utilization, Days Sales Outstanding (DSO) into the performance measurement framework

·         Create a segmented service framework: Most CPG businesses operate in the 80:20 model where 20% of customers account for 80% of revenues. Given this mix, a one size fits all shared services model will fail to deliver differentiated service experience to top tier customers. Consider a three tier service framework involving Transactional shared services, Business support services and Enterprise support services layers

·         Choose the right global delivery model: Most top tier BPO providers today have global presence with established Centers of Excellence. Adopt a hub and spoke global delivery model preferably aligned to customer tiers as well as the segmented service framework

·         Adopt a phased transition approach: A prudent transition plan is the foundation to successful Order to Cash shared services adoption. Consider creating a phased transition plan where phasing is based on customer tiers, markets, process complexity or a combination of these parameters

While there are several other design considerations, the above five are what I could call the essential ingredients of Order to Cash BPO solutions for the CPG industry. With these best practices implemented, companies can start to cash in on their Order to Cash processes through BPO shared services led transformation.

This article has been authored by Sushanth Ananth (Manager, Client Services - Retail, CPG, Logistics and Life Sciences, Infosys Ltd). You can reach Sushanth at Sushanth_ananth@infosys.com.

 

 

 

July 20, 2010

RFID Middleware - Smart Devices?

For any organization to adopt RFID (radio frequency identification), the biggest challenge is to integrate their existing enterprise applications with the data collected from RFID devices. Neither do these applications have the capability to hold and process EPC data, nor do they have the capability to communicate with devices and hardware. While the first issue of holding and processing EPC information is something that I've addressed in my white paper here, this blog is targeted towards discussing the issue of integrating enterprise applications with devices.

This purpose of integrating existing legacy applications with RFID is enabled by deploying a thin piece of software called as the RFID middleware. Simply put, the job of the RFID middleware is to collect data from RFID devices, eliminate duplicate or junk data, convert the data to a format which the enterprise application understands, and feed the data to the enterprise application.

The market for RFID middleware has been tapped by many players in the space, though it is dominated by product vendors who consider their RFID middleware as an excuse to sell more licenses of their existing products by making it dependent on their application servers, web servers, database servers, so on and so forth. Device vendors realized this to be a challenge for enterprises in terms of cost, and came up with an alternative - provide an embedded version of the middleware on the firmware of the device itself (making it a "smart" device) and let it communicate with enterprise applications directly. But is this a favourable approach?

There are several returns of having an embedded RFID middleware, e.g., it reduces the TCO (total cost of ownership) because it is not dependent on any other licensed products like application-servers or databases. However, are there several drawbacks as well, for example, it over-burdens the device with additional processing responsibility, it doesn't allow filtering of duplicate data to be done across devices (since a middleware instance is specific to a device and can't access data from another device), and most importantly, it increases the dependency of the organization on the vendor - which is not a scalable strategy in the long run, especially for an enterprise which spans across multiple geographical locations, in each of which the vendor might not have presence and customer-support.

Therefore, a more recommended approach is to use an RFID middleware which brings about the best of both worlds - the processing capabilities of a server-based RFID middleware and the low-cost, light-weight behaviour of an embedded middleware.

I've done more in-depth analysis of this in a technical view-point which was published on RFID Journal under "RFID Expert Views", accessible here. It provides more insights into some of the unforeseen challenges that an embedded RFID middleware on a "smart" device might bring about for an enterprise. If you have any thoughts to share on this topic, I'd be glad to take your comments on the same.