Home Depot, McDonald’s Pushing Non-Traditional Kiosk Trials
Written by Evan SchumanHome Depot and McDonald’s are both in the middle of non-traditional kiosk trials. McDonald’s is on its fourth such trial, after having concluded that the first three simply didn’t work well. Not too many retailers would opt for a fourth trial after three unsuccessful attempts.
The non-traditional Home Depot kiosk trial is based more on the units themselves—small mobile units, some as tiny as 5-inches tall—and the size of the chain’s planned kiosk commitment: Well north of $100 million for full deployment, according to the technology consultant handling one of the trials.
The Home Depot kiosks are sort of a cross between a kiosk—with its capability to show related products and other multimedia—and an old-fashioned queue buster, which allows barcode scans, checkouts and payment to happen in the middle of an aisle.
But some of the Home Depot trial kiosks add some CRM twists, with preferred customers being given loyalty cards with embedded passive RFID tags. The cards alert associates when that customer enters the store, allowing them to be greeted by name in the middle of an aisle. All products discussed and considered can potentially be entered into the customer’s CRM profile, along with items purchased literally via the kiosk.
One of the Home Depot kiosk trials—happening initially in two stores starting Dec. 1 and code-named Unicorn—equips each store with 20 kiosks, according to Scott Minor, services VP for InfoLogix, the company handling this particular Home Depot trial. Minor said that another vendor is handling a different kiosk trial in parallel with InfoLogix, with Home Depot looking to compare the results.
InfoLogix has created three different kiosks for the Home Depot trial (a 5-inch, a 10-inch and a 12-inch) and each store is given an assortment of units to choose from. Some will work better for different customer needs. Only the 10-inch and 12-inch versions, for example, can handle multimedia, Minor said. The units all have 2-GBytes RAM and the hard-disks range from 5-GBytes to 20-GBytes. The kiosks can handle traditional payment and debit cards, but no checks, cash or contactless payment cards. The kiosk communicates with a small wireless portable printer to generate receipts in the aisle.
For this trial, the devices can’t read RFID item-level data, but Minor characterized such a capability as one of Home Depot’s "future stated directions."
Minor wouldn’t reveal the cost of the trial, but did say that his company—and not Home Depot—was eating all of the cost. Ahhhh, sometimes being a Fortune 22 retailer—with $85 billion in annual revenue—has its advantages.
The way the trial works is simple. The store identifies 150 of its top customers, in this instance professional homebuilders, who might go into a Home Depot an average of twice a day.
Each of those hand-chosen customers is given the special orange-color loyalty card with that active RFID tag and 2-MByte memory for CRM data about that customer. There is one reader over the entrance that such customers typically use, Minor said, as well as readers in each aisle. When the location-based system detects the card and that customer’s exact location, an SMS alert goes to associates. One associate grabs a kiosk and meets that customer in the aisle, greets the customer by name (hopefully he hadn’t borrowed his boss’s loyalty card or that will be one blown bit of personalization), helps find items and then assists in the self-checkout process.
This is another example of the trend of chains using kiosks to help—as opposed to replace—associate interaction.
Going from orange loyalty to orange marmalade (you think it’s easy transitioning from Home Depot to McDonald’s?), McDonald’s kiosk strategy is quite different.
McDonald’s has done several kiosk trials, but all ended disappointingly, and it was because of an insufficiently customer-friendly interface, said Jeff Smith, the CEO of Clarity Consulting, which is handling the current McDonald’s kiosk trial.
The initial trials leveraged the same content and programming from the chain’s POS system, which was designed to be used by trained employees. It was apparently not sufficiently intuitive for consumers, Smith said.
The older trials displayed only parts of the McDonald’s menu and did not sufficiently factor in regional menu differences, he said.
One example Smith cited of a difference not considered: Sales tax. In some areas—such as Chicago—the tax laws differentiate between a hamburger, fries and soda purchased at the drive-through and the same meal purchased from the counter. Apparently, some laws see the counter-purchased version as more of a taxable extravagance. How would such laws view a kiosk purchase? The soda would be considered an extravagance, either way.
Then there are the global issues along with the various languages spoken in parts of the U.S. Smith said the current trial focuses much more on those issues.
Another change is the ability to alter an order on the fly. Let’s say a customer orders a Coke and wants to change it to a Sprite. In the old system, Smith said, the Coke would have to be deleted and the customer would have to start over again to add the new beverage. In the current trial, that drink can be changed directly on the screen in one move.
The initial McDonald’s stores for this kiosk trial are in Australia and Germany, Smith said.
As for why McDonald’s agreed to do yet another kiosk trial, Smith said it was nothing that his company said. What convinced McDonald’s was the demands of many franchisees, some of whom wanted to do kiosk trials on their own. If it’s to be done, McDonald’s concluded, it might as well be done consistently and through corporate, Smith said.
October 16th, 2008 at 6:54 pm
I firmly believe that to have kiosks work effectively, companies need very clean and accurate data. Most retailers are challenged in this way, and I’d be surprised if we didn’t hear at least one big “ooops†from each retailer that tries to roll out kiosks. Sort of like when I was with a dotcom in the late 90’s and our product team put the sku for an IBM server in for a Palm Pilot cover, so when someone purchased a leather Palm Pilot cover for $12 bucks or so, they’d get a server instead.
Since most retailers are still getting very bad data from their suppliers, they either need to have a great product data quality solution or expect some really interesting results from kiosk orders…
October 20th, 2008 at 9:15 pm
Effective retail kiosks should connect effectively with the POS system,so that to the POS software they appear as another lane. However the kiosk interface should be entirely different from a POS interface. It’s interesting that MacDonald’s have now tested and proved this.
It’s true that data in the POS generally needs some improvement for use on the kiosks. Generally the improvements are in better product descriptions and photos, rather than product skus, which are already in use on the POS.
October 21st, 2008 at 8:07 am
Editor’s Note: Good point, Cath Sample, but I would make a very nitpick. To the best of our knowledge, McDonald’s has yet to have “proved this.” The trial is under way but hasn’t yet delivered final results. They may indeed prove it, but we’re not aware that it has happened yet.