October 24, 2015
TO: Doug McMillon, Neil Ashe
Wal-Mart Stores, Inc.
702 SW 8th Street
Bentonville, Arkansas 72716 USA
SUBJECT: How to beat Amazon
The recent NY Times article points out the giant task you have to become more relevant in the online business.
If you want to make significant progress building your online business and have a serious impact on Amazon you need many things: more products, a better website, slick mobile app and price competitiveness. But that alone won’t draw people to start using Walmart. The one thing that will make a difference is low cost same day delivery.
To do this you need to think outside the box. So here is a totally out of the box idea: It may seem crazy but it's worth considering.
You have 70% of the population living near a Walmart store so shipping directly from your stores is one part of the solution. You have way more coverage than Amazon. The other is getting your customers to do the delivery for you. How would this work....
1) Let's start with the obvious. All your warehouse and store inventory must be available to the web on a real time basis so that you can make the decision of shipping point. If what a customer wants is in a store close to them and your algorithm indicates that you could ship from the store with the same day option then that is offered to the customer.
2) The next KEY STEP is that workers in the store who are dedicated to online shipping must QUICKLY pick the items from the shelf for the order so store customers don't take the stock. You can't wait until later and risk that the stock you promised is now gone.
3) Once an order has been accumulated and complete it is boxed and bar coded and put in the shipping area which is a new location near the front of the store.
4) Now the fun part starts. Imagine that you have developed a cool and fun Uber type app that your customers can download to their smart phone. By accepting the terms of the app you will know their address and they will have agreed to "potentially" deliver packages to nearby neighbors for a delivery fee. Perhaps $3.00 would be enough to entice them but that would have to be tested.
5) So now you have open orders to be shipped and you have customers in the store who may live close to the delivery address and have opted in to the delivery program. With constant cycling of the algorithm you are looking for customers who live close by who are in the store. Obviously you will have to fine tune things based on just how close the customer is to the delivery point (route wise) and how many deliveries have they accepted in the past, etc. And if you have two or more customers then rank them, etc.
6) Once you have a potential delivery person you text that customer with a message, send an email and popup a notification. The basic message is "Would you like to earn $3.00 delivering a package to a neighbor" and give them a map showing them the route from the store to the delivery and then to their house. (Just like Uber does when you get the ride summary email). Plus it would be cool to calculate the extra time involved. They click on YES or Thanks, but not today. If they click yes then you thank them and tell them to stop by the pickup area before leaving the store. Using GPS you monitor the fact that they are still in the store but if they go away from the store and never pick up the package you have to handle that option. Maybe alert them so they can elect to come back or to cancel.
7) When the customer comes to the pick up area they activate the app and press "Ready to pickup" button which displays a 2D bar code on the screen that your employee scans. This tells your employee what order to pick from the rack. Your employee scans the package as they give it to the customer. Now the package is "out for delivery" and you basically know when it will arrive, assuming the delivery customer goes directly to the ordering customer. You could at that point email and text the ordering customer with delivery notification.
8) When the delivery customer reaches the customer location they indicate on the app that it has been delivered. When that is clicked you capture the GPS location to verify that it matches customer location. Plus you can look at the route taken, how long, etc. Just like Uber does.
OK, that's the Utopian summary but it obviously has some huge issues to deal with.
A) You can't even consider launching this without fully developing the software and testing the hell out of it with beta testers. Premature launching of a buggy app will kill it before it can get any traction. Considering the complexity of this you would have to devote a considerable number of very talented developers to get this done. But hey, you need to do something bold. You might invest $10m on this and throw it away.
B) The big unknown is will customers do it and will there be problems with missed deliveries or locations customers don't want to drive to, etc. Clearly if you get enough interest you will have to rank delivery customers and screen out those who don't perform. Before investing in any software you should put together a very slick Powerpoint presentation of the whole concept with mock up of the screens and concepts and test that extensively with focus groups throughout the nation. Testing it with customers who are known frequent shoppers. I would develop a mobile trailer that is very comfortable and could seat maybe 12 people. You drive it to different stores and entice customers who have a smartphone to participate in a 30 minute presentation for a $25 gift card. If the focus groups are a bust then it's probably not worth doing. But sometimes people cannot conceive of how something would work. But the PP could get it across clearly. If they like it then go dark and take the plunge to develop it. Get the participants to sign NDA’s as part of getting the gift card.
C) The other thing is that in the beginning you will not have enough delivery volunteers to handle the load so at the end of the day you will have to make sure the promised deliveries happen using your own delivery trucks or paying a local cartage. This will probably be a requirement even if the program is a great success because there will always be something that nobody opted in to deliver. But if you can get 80% delivery it would be very cool and cost effective. Clever analysis of end customer address could guide one to not assign it to a store because there are no delivery customers signed up who are close enough.
Some people may see this as an opportunity to earn some money so they might just park outside the store waiting for a delivery opportunity. Those people could be flagged as dedicated drivers and regardless of where they live they are basically saying that I want to deliver packages and could go anywhere within a reasonable radius from the store. Let them indicate just how far they are willing to go. I could see many retirees who would love the opportunity to get out of the house and earn some money. It’s just like an Uber driver only with no passengers to deal with.
OK, that’s it. I hope somebody actually reads this. Maybe it’s crazy but if by chance you decide to go forward please use our Walmart store in Placerville CA as the first beta site. I have had a long and successful career in software development and management and while I am now retired I would love to participate if there was a role in beta testing.
Regards,
Tom Dillon
Retired Chief Operations Office, Netflix 1998 - 2006
Placerville, CA
