Analysis Of POS Data – A Journey From Distrust To Collaboration
POS stands for point of sale. The time-space moment when the cashier presses the “Total” button on his register and the bill whirrs forth from the spool. Or, in the virtual arena, after the short wait as your credit card details are confirmed and the “Your transaction is confirmed” pops up on your screen. Each of these transactions generates a row, or perhaps a few, in a database and becomes a drop in the ocean of POS data. In the retail world, POS stands in the pantheon of ROI, EDLP, GM, OTB, BOM, etc. The concept behind the abbreviation is so embedded in a retail activity that anyone with a brief association does not need further elaboration.
But what concept underlines POS data – recording each transaction and item in the store in the retailer’s register? In today’s world with global retailers selling millions of items each day to hundred of thousands of customers across thousands of stores spread over the globe, recording and storing for protracted periods each of these transactions means massive investment in IT, workflows designed around the feedback loops emanating from this data, staff and warehousing costs. Why go through all the pain and millions of dollars in investments?
Obviously, the gains are bigger than the costs – and the retailers have realised that the benefits are many-fold. But when the first cash register was made by James Ritty at the end of the nineteenth century, the benefit sought from the activity was very single-minded and simple. Once the employee rang up the amount for the transaction (the device being totally mechanical) and pushed the “Total” key, the cash drawer opened for that instant and the employee could put the cash inside and return the change.
A bell also rang at the moment the drawer opened (the bell to become a staple for any drugstore scene in classic Hollywood much as the sound of the swinging doors in a spaghetti western) and alerted the owner that money was changing hands. Mr Ritty made the cash register to monitor and control employee theft.
(In the age of surveillance cameras then, the amount of investment POS data management requires would hardly be justified by this narrow motive anymore. )
With the addition of the receipt spool soon, generating a receipt for each transaction, the cash register also served in maintaining the account of the sales and reconciling cash flows at the end of the day. Whatever cash was there in the register at the start of each day, net of the sales and refunds should remain in the register at the end. Of course, there might be some manual errors in inputting prices and calculating the taxes, but many errors in recording transaction details by hand were reduced. But beyond the accounting and monitoring facility, the cash register offered little help elsewhere in tracking sellthrus, sales performance and inventory tracking that a modern POS system does.
The next big innovation in the cash register was the development of McDonald’s microprocessor-controlled cash register systems in 1974. With its limited menu, it was possible for McDonald’s to assign a key to each of the items in the menu and store the price and tax information in the system. This eliminated any scope for manual error in price input and made reconciling accounts more accurate and easier. The other benefit came in making the task of the cashier easier and faster turnaround at each register: from scanning and punching prices each time, all the cashier had to do was to punch the item number and the units bought and with a press of the “Grill” button, he could move to take the next order while the first was still in process. The registers were looped in a local area network and any device could be used to handle the server end at a time. With data backup, even the risk of device failure was reduced.
As hardware and technology improved in leaps and bounds, so did the capabilities offered by POS systems in understanding the business beyond single transactions to a comprehensive multi-dimensional view of the business –top selling lines, best selling flavours and sizes, store performance, inventory – to name a few. The automated processing of these reports every day gave the retailer hitherto unprecedented insight into understanding how customers are buying into his proposition and optimizing it accordingly.
The landmark partnership of Walmart and P& G in the eighties ushered in a new paradigm of vendor-retailer relationships. The linking of Walmart’s POS systems with P& G’s supply-chain system through EDI enabled both of them to continuously track, forecast and replenish the sales of P&G’s products in Walmart stores and hence maximize their sales while keeping the inventory in the system, both Walmart’s and P&G’s, at the minimum.
This was a giant leap forward from the previous distrust and opacity that had underlined the relationship between the two giants. At the end of that historic white water ride, Sam and Lou decided that the opportunity was too large for both partners to waste. Today some of the retailers, led by Walmart and P& G, trust their key partners enough to share the sales of entire categories, including their competitors’, and manage their entire category business for them, including assortment selection, placement and presentation.
And so was the journey of the POS system and its data analysis, from its genesis as the ringing Watchguard of the distrusting store owner to the juncture where retailers and vendors collaborate to maximize the performance of their associated business and provide the customer with the best choice with the maximum service level and the minimum fuss.
conclusion
Big data analysts are at the vanguard of the journey towards an ever more data-centric world. Being powerful intellectual resources, companies are going the extra mile to hire and retain them. You too can come on board, and take this journey with our Big Data Specialization course.
