Visual management, kindergarten style

Toyota's lean operations system emphasizes making work and data visible inside the workplace. It part what's been meant by the term "visual management". It's an idea you can see in practice in any kindergarten. And no one has more fun with it than them!

When I introduce a new product how much should I make? Here's an idea: zero.

You've got a brand new product. You've never sold it before. In fact, no one has ever really sold anything quite like it. Also, it's not really one new product. It has many variations—color, size and so on—so it's more like dozens of new products.

Here's your problem. How many of each kind should you make?

It's a classic manufacturing dilemma. The typical approach is to guess. No one calls it guessing of course. You dress up the process to make it seem like you're not guessing. You have math and formulas. You have spreadsheets. You title the spreadsheets "forecasts" which is another word for "guesses" but sounds way more scientific. And you are wrong. Always. You make too many of some versions, too few of others.

Another approach is to not make any—at first. You wait until one is ordered, then build it to order. Sure, you have to have all the components on hand so those aren't built to order, but most of the components are probably shared between the different variations so it's not that big of a deal. The big benefit is that you never have the wrong level of finished goods inventory. It's always zero. The only products that exist are the ones that are already paid for.

Apple introduces their watch this week. Build-to-order is the approach that some believe they are taking. There are demo versions of watches out there but, when you buy one, the order goes to China and someone in a pink dust suit starts making your watch. 

Will Apple keep doing this forever? I don't think so. For one, FedExing a single watch at a time out of China is expensive. Another reason is that build-to-order creates a long lead time for the customer—days, if not a week or more (though Apple doesn't seem to care much about long lead times when they introduce products, to some extent it appears to be their strategy). If Apple does build to order they're probably doing it to learn about the demand. Once they see which versions sell in what quantities they can begin to build inventory ahead of time.

The urban density math of snow plowing

I caught this piece today that said 25% of Detroit's streets were plowed a day after the city's biggest snowfall since 1974. Whether or not that's good or bad here's another way to look at it.

Detroit has 1,884 miles of residential side streets and about 700,000 residents. That means every resident has to pay for about 14 feet of plowing.

New York City has 6,000 miles of streets and 8.4 million people. Even though that measure includes all roads, not just residential side streets, the fact that there are so many more people means big per capita savings. Each New York City resident only has to pay for 4 feet of plowing—over a 70% savings.

It's just a reminder of the uphill battle depopulated cities like Detroit have. The people leave but the streets don't go away.

Detroit can't copy New York's density but they might want to borrow one of its tricks that can create savings. The Big Apple owns very few plow trucks. Instead they just mostly stick plows on garbage trucks, like in the picture above. 

Silicon Valley wants to deliver your food

Restaurants and small food shops have always been flustered by delivery. On the one hand they could help customers—and find more of them—if they took orders online and delivered. On the other hand there's the problem of how to price delivery, the logistics of delivery and the problem of setting up an online order system and making sure its inventory is accurate.

In the last year there's been a wave of new Silicon valley start-ups that try to help with the last part—the online order system. 

The most prominet are Grubhub and Seamless. They take orders for restaurants. The restaurants figure out how to make the food and deliver it. Grubhub and Seamless take a cut that's probably around 20%. Speaking personally, I've used Seamless a lot in Brooklyn and it's very good. The benefit to a restaurant here is that they only have to figure out the logistics part of delivery. They can put all or just part of their menu online—and make it available at times that make sense to them.  Take Prime Meats in my neighborhood, a fancy restaurant that's full almost every night. They are on Seamless but in order to prevent overburdening their kitchen they initially showed up on Seamless only between 5 and 7pm, when they were slow.

In the novelty arena, you can also order pizza on a smartphone, albeit in a ridiculous way. There's a one button app that, when you push it, delivers pizza in 30 minutes. From somewhere. Anywhere.

In a more interesting twist, Square, the payment processing software company, is buying a food delivery company called Caviar. This is the only Silicon Valley firm I know trying to do the delivery part of the delivery business. Presumably the idea here is that a restaurant can buy their POS system from Square and the delivery software—and delivery logistics team—will come along with it.

This is happening with grocery, too. I just spent time at Bi-Rite and talked with the GM Patrick (ex-Zingerman's Deli manager) and learned about Instacart, which Bi-Rite just joined. With Instacart you place your grocery order online and they find someone to go get it from you. It's not an employee of the grocery store, it's not an employee of Instacart, it's just some shmo who signed up to be a grocery store picker. (They call them pickers, just like we do for people who pick items for boxes on our production line). Like with Grubhub and Seamless, Instacart takes a cut.

How these all play out will be interesting. Short-sighted merchants, or ones that do their own order and delivery, may look at the cut these companies take and say they don't need to pay someone else for something they can do on their own. The problem there is they will be shut out of network effects. The more merchants sign on to Seamless the more common it'll be for customers to shop there. If you're a merchant and you're not there, you'll loose out. Merchants will be saving cost to give up sales, which is rarely a good move.

What's the downside for the customer? On the restaurant side there seem to be very few negatives. Seamless doesn't mark up for delivery so why not order online and get the same food you could have driven to pick up brought to you for free? On the grocery side, I can see inventory being a hard nut to crack. Right now Instacart has no database connection between what's for sale online and what's in stock at the merchant. If something is sold out, the merchant has to remember to go to Instacart's website and mark it sold out. Will that happen? Sometimes, but not always. That will mean upset customers. Instacart gives leeway to their pickers to choose subs or call the customer to see what they'd like, but either answer is a flawed fix, one that will frustrate customers and hurt sales.

Recent Reading

What if UPS and FedEx were like Uber and Lyft, where regular people do the deliveries, not employees? She started a company to see if that can work. 

A 40 fruit tree. Hat tip to Spike. 

Ever wonder what a camel broker eats each day? Hat tip to Val.

What do lean operations and New York bodegas have in common?

New York sometimes gets a bad rap for a mythical, brusque, who-cares kind of service, but speaking personally, I've almost never experienced it. What I've had instead usually combines professionalism (on-time, accurate, everyone says "sir" a lot) with individual customization. Take this instance.

Recently I stopped by the closest bodega to my house for a couple cans of beer as I do from time to time. They have a big selection and I always hope they'll have Heineken in a can but they almost never do. So I chatted with the guy who appears to be kind of like an owner — you know, he was the kind of person who acted like one whether he was or not —and asked if he'd take suggestions for beers. "Of course!" he said and agreed with me, "Heineken in a can is so much better!" He got the attention of the other guy in the shop and told him, "Let's get a couple cases tomorrow." This is 10pm at night. Next day, they were there.

That's what a corner store used to feel like. Or so I imagine. Frankly, I've' never experienced many corner shops in Michigan. In one way it's weird that there's this level of personalization in the biggest city in America. On the other hand, there are a lot of factors about shopping in New York that, when thought through, make its personal service seem not so odd.

In New York, everyone shops in the immediate vicinity of where they live because it's a pain to travel long distances hauling stuff on foot. You don't have a car and you have to carry everything yourself so you shop in small batches. This is a key factor that, like small batches in lean operations elsewhere, leads to beneficial and unexpected results. Because you shop in small batches you'll often be in the same shop several times a week.

The shops are different than many other cities, too—smaller, often run by adults, not the teenagers and college students you see working in big box stores elswhere America. The staffs don't turn over very much. I'm not a guy who is super chummy with everyone when I shop but it's telling that I know the names of the person who runs the laundromat, the florist, the wine shop, the cafe, several restaurants, the bodega owner and probably a few others I'm forgetting. In Ann Arbor I knew maybe two of the names of the people who ran their shops—Bob Sparrow the butcher and Mike Monahan the fishmonger. (They were there when I shopped.) Ann Arbor is a town a fraction of the size of New York but, to me, felt far more anonymous

Anyway, this is not meant to be a plug for New York. I wanted to point out a couple things. One is that smaller shopping batch sizes, one of the principles of lean, lead in this case to greater personal contact. That personal contact ultimately, in my case (and I know I'm not alone), led to better and more customized service. That service was not administered by a survey or a some other process. It was a question, an answer, and and act—all done immediately, just in time, on demand. It's a powerful way to run a business. These are great lean skills, hard to replicate and some of the reasons that, in spite of CVS and Rite Aid and other national chains trying to make a dent in the commerce here, small owner-operated bodegas in New York thrive.

Same day delivery? How about same hour delivery?

eBay now has $5 same-hour delivery in certain cities for certain items. This comes on the heels of Amazon announcing same-day delivery. Is this something food mail order companies should be worried about?

I don't think so. This is set up for self-buyers (people buying for themselves), not gift givers, and food mail order is largely a gift business. (Imagine the message it sends if a person wearing a one-hour delivery shirt shows up at your door with your birthday present.) Another reason is perishability. The way same-hour and same-day works is they need to park inventory in every city to have it ready to ship on a moment's notice. That's a real pain to do with food — at least the food people want to give as gifts — since it tends to expire at a rapid rate.

Should any business be afraid? In general it pays to be afraid of eBay and Amazon so yes, someone should be scared. If you sell commodity electronics or home goods — i.e. if you're K Mart or Sears — you might want to be extra fearful. Well, if you're one of those two you're probably already scared to death since you're so far behind in retail period. WalMart — they might get a little nervous too. They've announced same-day delivery as well, using their network of stores, but they've yet to make their online business connect like Amazon's does.

Then again maybe there's not much to worry about. Others have tried this and failed plenty of times. Most notoriously, I remember New York's Urban Fetch and Kozmo in the late 1990s. You could order a CD, a box of donuts and a pint of Ben and Jerry's (my first order) and a bike messenger would deliver them to you in an hour, often with free warm cookies! It was amazing, so amazing that it lost mountains of money and both were gone within a year.

10 Awesome Effects Single Face Racks Will Have

Last week we got our first single facing rack for picking. Where we used to have whole cases now we have single bottles. On its face this seems like one of those operational improvements that only an operational nerd would like. Well, that's probably true. It also seems rather incremental. We used to put whole cases in a rack, now we put a single bottle, big whoop. It doesn't seem like that much. That's not the case. This is huge.

This is one of those rare incremental lean improvements that will have tremendous and far-reaching effects. You don't get big leaps like this often in operations. When you kick off a lean transformation you get major improvements by going from batch to flow and implementing pull systems like marketplaces with kanbans. Those usually have huge space-savings and labor-cost improvements. Jaw-dropping numbers, like our first year labor savings of over 30%. From then on the changes have smaller bottom line effects. Sometimes they feel nonexistent. (We've had years after our initial lean transformation where our labor cost as a percent of sales went up, not down.) Sometimes it can feel like two steps forward one step back. This, on the other hand, is like twenty steps forward.
 

We can't single face every product—yet. Maybe 150 of our 800 items to start with. That number will go up over time, though, as we figure out how to make more jars flow and not break in the rack, unprotected by their case.

10 Effects Single Face Racks Will Have

1. Cases are usually 3 products wide. That was the width of a pick slot. Now a pick slot is one product wide. We just cut our floor space for this section by two thirds.

2. Since pickers only need to pick the first product and never have to reach back into a case we don't need to size the racks to have reach-in space. The vertical space between shelves can shrink. We can fit more shelves in, perhaps up to 25% more. We just cut our floor space by another 25%.

3. Combine the above effects and this section's pickers will walk 75% less.

4. We can add hundreds of new foods to the pick line without moving to a new building (we still have back stock space to work out first).

5. Pickers won't have to sort in the box (visually) for the next product.

6. Pickers can grab a product out of this rack more easily and quickly than out of a case.

7. Pickers can get visual cues about address slots since the products have color and shape variation, unlike cardboard boxes.
 
8. Pickers can spot problems like dents and leaks more easily.

9. Pickers won't have to manage the recycling of cardboard boxes.

10. The rack is faster to adjust so we can add items, delete items, and move items more cheaply. Now a fast mover can move to an easier pick location quickly. We can re-locate a super fast mover to the front of the pick line so it's picked and done and doesn't pass through all the pick stations. Or we can spread fast movers in the holiday so we even the workload at pick.

Randomness Helps

I never stop being amazed by small, counter-intuitive operations innovations. From a New Yorker article on Yoox, the online fashion site:

When clothes arrive at the warehouse...folded items are placed in black plastic storage bins that look like large milk crates. The crates are packed randomly—pants, shirts, and sweaters are mixed together willy-nilly. When an order comes through...it is placed on a conveyor belt that...delivers it to the correct wrapping station, which are manned by humans. Were items sorted with their like, the humans would have to search all of them to find the one matching the order, but since items are sorted randomly, it's easy to spot the right one. "Chaos is our friend," Guillot said.

I don't know how to use randomness like this at ZMO but maybe there's a way?