Linkedin Share

Friday, November 22, 2013

Is Item Level RFID The Joke That Won't End?

I was directly involved with RFID in my early days at Microsoft in 2003. We were convinced RFID was going to take off in retail and in short order and so much so we invested in an RFID server built on BizTalk. I wonder what happened with that server?  Wal-Mart was mandating RFID tagging from their suppliers which would then make RFID ubiquitous in only a few years and life would be beautiful.  I lost track of RFID for a number of years after that simply because none of what was envisioned ever happened, at least not on the projected scale. I moved on to this concept of customer-centric retailing with the focus on clienteling and analytics. That was the next big thing, the customer! I have always believed in the concept of item level RFID, but unfortunately the ROI must not have been there or more retailers would have implemented.

Let’s take a look at where we are today. Yes item level RFID still has a pulse mostly in high valued goods and items requiring traceability, but short of that “I don't see the dogs eating the dog food”!  So why has item level RFID adoption been so slow? To quote Patrick Javick, vice president of industry engagement at GS1 US, “Many of the benefits are now quantifiable: up to 99 percent inventory accuracy, inventory labor productivity increases by 96 percent, out-of-stock reductions of as much as 50 percent, and sales increases have been documented from 2 percent to 20 percent.” In the summer of 2013 “within a span of six weeks, two of North America’s three largest mid-market department store chains made big announcements regarding the pace of their chain-wide rollouts of RFID. And the aftershocks are still being felt, both domestically and abroad.” John-Pierre Kamel, Principal at retail consultancy RFID Sherpas LLC. Aftershocks, really? I guess it is not slow after all with aftershocks and tsunami's happening all around us.. Larry Arnstein, vice president of business development for Impinj states, “it is well established that store inventory accuracy typically hovers between 60 percent and 80 percent.

Okay so for me it is starting to make sense depending on how all the quantifiable and non-quantifiable costs/benefits add up. First it appears that apparel and high fashion is the best suited retail market for this solution. American Apparel has deployed and even touted their results in their 2012 SEC filing, “..RFID (solution).. Increase sales, reduce costs and increase liquidity.. also RFID (deployments)..have allowed us to reduce our unit inventory levels..”! In 2013 American Apparel CEO, publicly stated. “June represents our 25th consecutive month of positive store sales growth. I am excited with the 7% increase in comparable store sales in June particularly since it is on top of the 19% increase achieved in June of 2012. Others are apparently implementing item level RFID like Macy’s, Kohl’s, Lord and Taylor who have been identified in the press.

There is a case for simple math to take over and help guide this decision or is there? After you add up all the RFID costs and then subtract them from current sales (I conservatively assume flat revenue) and  projected savings in inventory  from better planning, logistics,  and store level stocking you end up with a number either positive or negative. Alright I did discount the notion of increased sales by not having out of stocks, so add that too if you like. 

Where I see the real driver of need for RFID item level tagging is with serving the demanding omni-channel requirements. How are you going to be customer focused providing great experiences with an inventory accuracy of 60 to 80%? I see many opportunities for retailers to turn well-meaning customer serving initiatives into customer satisfaction issues. You must have near 100% accuracy of your inventory if you are expecting to source from your stores for your omni-channel experience. If you say you have it in the store and I drive over to get it and you do not have it in your hand waiting for me you are toast! This need for knowing real-time what you have and very specifically where it is may be one of the biggest underpinnings of being a successful retailer in this new customer-centric world.


One more point, I know that item level RFID is a very complex discussion starting from source and through consumption. I hope to receive much dialog and feedback! This is customer-centric retailing. 

Wednesday, November 6, 2013

Are We Ready For Virtual Dressing Rooms?

This is yet another technology which I am seeing more and more, but I am still not clear on the appetite of the consumer to use. The concept is for you to input (manually or automatic) your detailed body measurements into one of these systems and you can then see how you will look virtually when a certain garment, style and size is mapped to your body. Wikipedia states, a fit technology may be categorized according to the problem that it resolves (size, fit or styling) or according to the technological approach. There are many different types of technological approaches, of which the most established and credible are:
- Size recommendation services
- Body scanners
- 3D solutions
- 3D customer's model
- Fitting room with real 3D simulation
- Dress-up mannequins/mix-and-match
- Photo-accurate virtual fitting room
- Augmented reality
- Real models

Wow! As you can see there are all sorts of approaches being presented to retailers as an answer to shopping online or even for in-store customer experience? Here are a few solutions you might want to check out, Fits.me, Facecake“Swivel”, TryLive and  Styku. I find this all pretty neat except what results does it drive? Based on non-verified market data, results are pretty awesome! Conversion rates by overcoming customer hesitation show an increase of 57% with reduction in garment return rates of 35%. – InternetRetailer This is huge increases in revenue and cut in costs. The reduction in returns must be more than worth the investment alone! Wait, what does it cost? “On average, data compiled from about 50 clients of retail consulting firm Kurt Salmon show that online consumers return 20% to 30% of orders of apparel and other soft goods.” -Internet Retailer

So is everyone out there jumping on the bandwagon? Is it too early to decide which technology and solution is best? How hard is it to implement and more importantly manage? Will online companies like Amazon just now getting into apparel “wipe the floor” with brick and mortar apparel retailers? Is there a play for this technology as part of the customer experience helping in fact to drive consumers to stores to try on clothing in front of these new “magic mirrors” like Bloomingdales is doing?

I have a question that puzzles me? I remember an old statement when we planned for MIS internal budgets back at IBM, “first we will measure with a micrometer and then we slice with an ax after we see what choices we are given from above” we would joke. What does this mean? Well we can enter in all of these measurements  and with tools like Microsoft’s Kinect and get it down to a micrometer, but you still only have a few buying choices like small, medium and large? Companies like Levi’s who make as many sizes and styles that you can think of might do well with this technology but for the most part it is the manufacturer of the garment that is the problem. Size medium in one style or manufacturer is different than the other. So each garment has to be cataloged and then precisely mapped to an individual's style and measurements by some algorithm.  This is where the “secret sauce” resides! If I could have garments made on demand like some men and women suits are done with a tailor, but augmented with this virtual world that would be something!


Bottom line is this is going to happen and apparel retailers you need to start figuring out which of these technologies and solutions you will use.