How Much Does Yesterday Cost?

Typical sales compensation schemes have a lower limit, a minimum sell price (or margin percentage) below which the sales representative will not earn a commission. This makes obvious good sense: if the company is not going to earn a profit from the sale, the sales agent shouldn’t either.

It’s also common that businesses sometimes need to waive this common-sense requirement. When a sale represents a first break at a coveted major buyer, when a sale is expected to lead to additional purchases or service work or other additional revenue, it can make better sense to take the initial sale at lower than usual profit – sometimes even at a loss – in view of the longer-term payback.

Both of these expectations ought to incorporate delivery schedules. No company is in the business of delivering products before they are ordered (but nice try, Amazon); most business-to-business (B2B) companies with a sales force that receives commission cannot offer instant delivery. Usually if a sales rep makes a commission in a B2B context, the product must either be configured or selected from a complex range of options that require specialized knowledge to make a good choice.

In other words, the sales rep needs to know the product.

Expanding on that, the sales rep needs to know the capabilities of the product and of the company – how they support the product in terms of training, warranty service, spare parts availability, and whatever else goes along with that particular product’s complexity that justifies paying a sales rep.

Barring exceptional and pre-approved circumstances, a sales rep who sells a product at no profit has done a disservice to the company, neglecting his duty, and does not deserve to be paid.

A sales rep who sells a product requiring faster-than-usual delivery has also done a disservice to the company and does not deserve to be paid—again, except in pre-approved situations. I would argue that it is actually worse than selling below cost.

Selling below normal margins has an obvious, easily countable cost; fewer dollars come in. The number is right there in black and white.

Rushing an orders sets off a chain reaction of costs, most of which are not measureable or are not clearly attributable directly to that one order. Rush orders usually first affect the salaried staff, whose pay is rolled into the overall operating overhead of a company. Depending on the business and the size of the rushed order, responding to a rush order can take up time from the highest-paid people in the company. Again: sometimes this is the right course of action. But the endorphin rush of responding to an “emergency” makes it all too easy for business managers to forget that the best use of their time is planning the company’s future. Any time an upper-level manager gets involved in the present, it is a signal that the day-to-day processes of the company are not meeting the needs of the business; it is the managerial equivalent of a machine break-down.

Managers rushing around saving the day inevitably impacts the workers, too, who must halt their usual practice (ideally following a reliably excellent process) to accommodate the boss’ special request. This might immediately lead to overtime and measurable, attributable costs incurred. But more of the cost is likely to hide in the shadows: skipping inspection or maintenance or documentation, something where the cost of the sacrifice won’t be felt immediately, and won’t even matter if the skipped step is skipped just this once.

It is never just this once.

Once the sales rep has learned that he can delight his customer with the exceptional performance, he’ll ask for it again – he’ll need it again. Once managers learn that they feel like heroes when they orchestrate the impossible, they’ll want to do it again. And once production workers learn they can skip a step without noticeable repercussions, they’ll do it whenever they need to.

A rush order is not really like a house on fire, despite the frequent reference to “putting out fires.” A house fire is catastrophic; nobody just “gets used to it.” A rush order is more like a small stone hitting your windshield. Nobody dies. Maybe it makes a small chip or crack. Maybe it doesn’t even make a mark! But if it happens all the time, you will lose that entire windshield. So how often is too often?

Generally a business won’t improve until its customers demand it. Prices don’t fall unless sales are slipping. Quality doesn’t improve unless orders are lost. The wake-up call that faster delivery times are needed will generally come in the form of these rush orders, these expedites that shake the tree from bottom to top.

Managers must recognize that an expedite is the operating failure of the machinery they are responsible to keep running.

Sales reps must recognize that an expedite is never free.

Cheap Justice

Writing about the opening of an Amazon fulfillment center, The Economist says:

Alas, the influx of jobs has not boosted wages for the region’s forklift drivers and order-fillers. In the years since Amazon opened its doors in Lexington County, annual earnings for warehouse workers in the area have fallen from $47,000 to $32,000, a decline of over 30%

In a story from its January 2018 print edition, The Economist takes Amazon to task for underpaying its employees, probably. What “underpaying” means is unclear, coming from a publication founded, one assumes, on economic principles. In the online edition the article carries the suggestive sub-head “Is the world’s largest online retailer underpaying its employees?” Then, after five paragraphs of implying that Amazon somehow systematically causes people to lose income, The Economist finally allows that:

Another possible explanation for Amazon’s pay is its reliance on unskilled workers with minimal qualifications. David Neumark of the University of California, Irvine, who has written about the impact of Walmart’s growth on retail wages, says Amazon’s highly automated warehouses may not require as many workers who can, say, operate a pallet jack.

Hmmm, maybe.

A forklift carrying a load can total about eight tons (16,000 lbs or about 7,200 kg). The actual weight can vary quite a bit depending on what’s on the pallet and how high the pallet will need to be lifted, among other things, but just to get a sense of the weight, that would be around four times the average car in the USA. Heavier objects are harder to stop, and forklifts often move in spaces where there are literally only inches (two or three) to spare between the vehicle and something that isn’t meant to be hit. Forklifts do move much more slowly than cars on public roads, but when it comes to damaging things, 16,000 lbs will crush most things even if it is moving slowly.

Forklifts are mainly used to move pallet loads, and pallets are usually stacked with cases or cartons.

Amazon is in the business of shipping usually small quantities of diverse things going to diverse places. While Amazon might consolidate parcels going to the same distribution point on a shared pallet as the last stage in the shipping process before loading onto a truck, most of the handling of objects in an Amazon warehouse involves people handling the items directly, one at a time or one case at a time. Rather than sending a forklift to go fetch a pallet of corn chips to be sent to a corner store, Amazon will dispatch a person to pick one fitness tracking band and one pair of earbuds.

In short, Amazon’s warehouses will be designed mainly for human traffic. E-commerce and the mail order catalogs of yesteryear will general have this human-traffic based warehouse design.

But most of our warehouses have been built for traditional supply chains, which mostly move pallet loads of goods: pallets upon pallets of frozen chickens from the slaughterhouse to the grocery distribution warehouse, and then from there one pallet for this store, one pallet for that store, etc, etc.

So it is quite likely, not just possible, that on average, Amazon employs fewer forklift drivers than the pre-Amazon average warehouse.

It is also quite likely that a person required to safely operate a 16,000 lb vehicle in tight space would be paid more than a person required to walk, read labels, and pick things up.

The complain that Amazon lowers wages for “warehouse workers” is like complaining that McDonald’s lowers wages for chefs, because after a McDonald’s opened on the same block as Le Snootier’s, average wages for restaurant workers on that block went down. Nobody at Le Snootier’s got paid any less; but people who were not qualified to work at Le Snootier’s got paid more than they were getting before.

This sort of insinuation that Amazon is somehow taking away from people something that they already had is familiar work for rabble-rousing, any time someone needs a rabble roused, but it is particularly deplorable for a publication calling itself The Economist. That highly relevant possibility that perhaps apples are being compared with oranges was admitted somewhere in the article does not absolve them of headlining the question and leading off with all kinds of spurious facts.

The “average” wage for warehouse workers may have been lowered because of the increase in the number of workers who are doing jobs that have always been priced at the lower end of the range. If you are trying to examine the effect that Amazon has on “warehouse workers,” why not look at the total wages paid to all warehouse workers after Amazon opened a warehouse in a given area? Or if you are trying to examine the effect on the community, why not look at total count of people employed, or average income per capita? Those measures may very well have gone up. But we don’t know, because The Economist didn’t consider them worth examining.

Is Amazon virtuous in all of its dealings? I hardly think so. But let it be found guilty of crimes it has committed, rather than being found guilty because it has done well for itself.

Imagine that Bob’s truck has gone missing. “I think Frank stole Bob’s truck,” I say.

“Why do you think Frank did it?” you ask.

“Well, Frank buys cheap boots. A guy that buys cheap boots seems like the kind of guy that would steal a truck,” I answer.

Maybe Frank did steal the truck! But the evidence offered is utterly irrelevant to the charge.

The Economist should be heartily ashamed of printing such insinuation-mongering.

UPS Nearing a Strategic Dead-End

UPS is at risk of being caught in a strategic dead-end. Low-margin parcel business is on the rise, putting pressure on operating costs. Leaving aside fuel costs, reducing labor costs is the urgent priority. There are some obvious possibilities for reduction in labor costs, but these methods are already being led by competitors. Outsourced “independent contractor” labor is essential to FedEx, and being taken to the next level by “gig economy” companies like Uber. Automating manual tasks—both driving vehicles and handling packages—is a major focus for companies ranging from Uber to Amazon to Google.

UPS so far has done a creditable job keeping tabs on the developing drone technology. The approach of multiplying the productivity of a human in the loop is almost surely the right first commercial deployment. In the long run, though, it is difficult to imagine UPS sustaining an advantage in technology, against competitors like Amazon and Google (and others). If delivery is dominated by self-driving vehicles serving as a mothership for a swarm of last-yard delivery drones, which company is most likely to have proprietary, competitive technology?

Optimizing markets served might offer some respite. A delivery company may choose to serve residences only in densely-populated areas; or, contrarily, might build a competitive advantage out of maximizing every-door service. But here UPS is bracketed; the cherry picking the easy neighborhoods is the obvious starting point of every new delivery service, from boutique couriers to segment-jumping giants like Google and Amazon. And, like it or not, every-door delivery is the domain of the United States Postal Service. USPS could be less efficient or effective than UPS, but it is unlikely to go away; and it’s very hard to win a competition when your competitor cannot lose.

When considering the quandary facing UPS for an MBA course, I thought the way forward would be through adding additional on-site services to delivery: pet walking, dry cleaning pickup and drop off, etc. Whether voicing a generic management platitude or a genuine conviction, UPS Vice President of Engineering, John Dodero, hints at this strategic advantage: “UPS drivers are the face of the organization. Customers depend on them.” There’s a reasonable chance that people might trust the uniformed, institutionalized UPS crew more than the ad-hoc opportunistic crews employed by Uber and other crowd-sourced delivery upstarts. But Amazon took direct aim at this last remaining advantage with their Amazon Key service. Now, the first mover is not always the ultimate winner; UPS could leverage the concept better than Amazon. Ask Palm Pilot or Blackberry how that might work out. But it’s a bad sign for UPS that Amazon believes they don’t need UPS’ long history of knocking on doors to be trusted to enter them.


So what to do? Should the executive team at UPS declare the whole thing useless and sell the company before the music stops? Is UPS now approaching the peak point that Yahoo and AOL and so many other now-forgotten companies crossed over?

The possibility is terribly real. But it is not so certain that UPS should declare defeat without trying. In particular, UPS should not abandon the in-home services front without a fight. UPS has more ubiquity, more longevity of customer-end provider interaction, and more accountability than any other provider, except possibly the USPS. To avoid an unwinnable struggle against its own union and its Uber-like competitors, UPS should steer hard away from low-paid door-knockers, and instead immediately begin rolling out a higher paying job with additional duties. Call it ParcelPlus: Parcels plus pet-walker; parcels plus flowers; parcels plus dry cleaning. Adding duties to the drivers will only make UPS’s staffing logistics more complicated; but competency in tackling such logistics is arguably one of the few UPS can confidently claim as an advantage.

Look for the tasks that used to be done by house servants or house wives; look for the tasks that people want people to do and won’t trust to robots. We are headed, with Titanic certainty, toward a civilization with a higher ratio of elderly than ever before. At the same time families have fractured into smaller and smaller pieces, and those elderly are less likely than ever before to be living with mid-life adults. Notwithstanding a lifestyle unfathomable to generations past, working-age adults increasingly feel economically straitened and unable to pay for full-time care. This implies a market for on-demand part-time elder care, with services ranging from a simple check in (make sure Mom didn’t fall), to paid socialization, to very carefully tailored forms of first-tier medical care (“assisted self-medication”). Although borrowing elements from existing business models, it would ultimately be a hybrid with no existing competitors, for which the involvement of humans is psychologically (even if not technologically) necessary, and which further benefits from uniformed, long-term providers.

On the back end, turn parasites into symbiotes. UPS has a vast infrastructure of trucks and hubs and aircraft. But competitors like Amazon are in an enviable, “unfair” position of being able to use UPS for segments which suit them, and skip out for segments which do not. Amazon can replace UPS’ services piece by piece, if they choose, and there is no obvious way to stop them. But the common enemy here is empty miles: empty UPS trucks, empty FedEx trucks, empty trucks in the vast ecosystem of Less-Than-Load (LTL) and trailer-load local, regional, and national carriers. By offering complete transparency of their available capacity, routes and timing in their vast back-end infrastructure, UPS could pick up utilization and cost absorption.

This is a bit like Google’s investments into internet infrastructure: make everyone’s data transfer better, increasing the appetite for search. Autonomous vehicles have the highest economic payoff with the lowest functional risk making predictable hub-to-hub routes on well-defined highways; the ability to add differentiated value on these routes will rapidly shrivel away. But connecting these major arteries to every point of need will remain a challenge for some time to come. Temporary closure of a main street in Small Town USA can result in diversion onto poorly-marked detours over roads not meant for trucks. Smaller businesses are likely to have more variation in shipping processes, more often requiring human interaction to sort out the quirks in today’s shipments.

By emphasizing high-margin value-add on the front line and low-margin, high-throughput on the back end, UPS can position itself to benefit from competitors in the same market space. Those competitors who are ahead of UPS on low-margin residential delivery can still make use of UPS’ robust, transparent, freely available and competitively priced major transportation infrastructure; but where simply getting a box to a doorstep is not enough, UPS can offer additional services through a trusted and accountable network of familiar faces.

In the end, UPS may find its best path forward in the counter-intuitive direction of adding more people to its last-mile crew, as long as it also finds more value-add services to perform – services which by their nature most people would be unwilling to relinquish to robots.

The only question that matters for UPS is: in what ways can UPS continue to outperform its emerging and established rivals? If that means reinterpreting UPS to mean Universal Personal Services, there’s no reason to hold back. Start small. Start now.

The Barnes and Noble Experience

The recent proposal to take Barnes and Noble private provides an occasion to consider what place a bookseller has in today’s marketplace. Any viable path forward must address the shift in book-buying from being the minimum possible effort to acquire knowledge and literature (essentially, a commodity) to a discretionary experience sought out when the urge for a particular aesthetic overcomes the expediency of digital media.

One of the case studies we looked at in my MBA program was Barnes and Noble, set at a point in time where Amazon was growing and traditional booksellers were in decline. The case study itself was from a time prior to Amazon opening physical stores, but we looked at the case shortly after news of the pilot Amazon stores broke.

While I do not recall the specific recommendations made by each classmate, I remember being struck by how many of them suggested that Barnes and Noble should get better at online sales. In essence, the consensus view was that Barnes and Noble should somehow out-compete Amazon in the area of Amazon’s greatest strength, after something like a decade of losing ground.

“Do what your competitors do, but do it better” strikes me as shortsighted advice than can only result in me-too, second place success. Customers, not competitors, should have our attention. It may at times be necessary to admit that a competitor made a good move; no company has a monopoly on good ideas. But imitation of a competitor should not be reflexive; it should be an outcome of re-examining our customer’s needs and our company’s strengths.

  • Is the competitor’s new offering a better way for your customers to achieve their goals?
  • What is the customer problem being solved?
  • Do you want to solve that problem?
  • Is there a different problem you are better able to solve?

In the case of Barnes and Noble, by the time of the case study Amazon’s e-commerce was well established. The possibility of Barnes and Noble surpassing Amazon’s breadth of selection or expeditious delivery was already remote. Competing on convenience or selection could only doom Barnes and Noble to falling further and further behind.

But a visit to Barnes and Noble has some pleasures that browsing Amazon can’t match. The library-like quiet and serenity, the physical experience of walking among rows of books, even the bookish smell; visiting a physical bookstore reminds you of everything an online store inherently lacks.

My last visit in person to a Barnes and Noble put me in mind of another store, a furniture store. It might have been Ethan Allen or Williams Sonoma – I don’t remember which. I remember the aesthetic sense of everything being clean, organized, and coordinated. I remember the implausible but alluring feeling that I could take that serenity home if I bought the furniture. And I remember Barnes and Noble feeling the same way.

Imagine a clean and well-lit living room where you can read for a few hours in peace. What if there were three or four rooms, in different styles for different moods? What if there were daycare on site, pay by the hour? What if you could meet there, once a month or once a quarter, for a white-tablecloth dinner, with some old friends and new acquaintances, knowing in advance some of the books you had in common?

What can you get, in person, that you will never get online?

Google Can’t Handle Reality – Yet

How hard can it be to make a fully-automated warehouse?

Well, so far it has stumped Amazon and Google.

To be fair, it depends on what exactly you’ll be doing in the warehouse. There are lights-out warehouses in use today. But widespread adoption remains elusive, always just about to take over the industry thanks to the latest technology somebody is selling. Despite what Business Insider called an “army of robots” in Walmart warehouses, there is still an army of people, too. Amazon is in a similar situation, with a growing human workforce despite acquiring Kiva Systems.

Google is also trying to solve automation of material handling. Just look at the effort they have put into it so far:

  1. At the end of 2013, Google had Andy Rubin in charge of a collection of seven robotics companies. Rubin was credited with the success of the Android mobile operating system. For robotics, he saw “clear opportunities” in “both manufacturing and logistics,” intending to “sell products sooner rather than later” according to the NY Times. But Rubin left a little less than a year later, with Google brushing off questions about competing with Amazon.
  2. Nevertheless, in 2015 Google filed for patent, granted about a year later, pertaining to automated vehicles working in a warehouse.
  3. Another patent granted in 2016 deals with coordination between flying and wheeled delivery robots.
  4. Google had a keynote presentation at the 2014 annual conference of the Material Handling Industry trade association.
  5. In March 2016, Google unveiled efforts to train a robotic picking arm. Not coincidentally, Amazon has for three years running been sponsoring a robotics challenge for picking.
  6. Speaking of training a learning algorithm to pick parts, Google Glass is now used in factories and warehouses.
  7. Google also has a long-running and slowly growing delivery service, Google Express. Partnering with retailers like Target and Walmart helps the traditional retailers compete with Amazon.


While the evidence suggests Google is gearing up to enter warehousing, material handling and logistics, as yet no specific products or services have been announced. If the business vision is to provide an open platform logistics service, a low-key approach makes sense; the last thing existing retailers need in their battle against Amazon is another Amazon under a Google brand.

Also, much as is the case with their autonomous car division Waymo, Google has nothing distinctive to offer until their warehouse robots are really autonomous. There are a plethora of warehouse automation products in the market today, and the only thing they all have in common is a failure to fully replace human workers.

Google’s failure to bring any warehouse solutions to market could be taken by incumbents as sign that they are safe – the challenge is just too great for Google to handle. Or it could just be a matter of time before Google Search is returning real, tangible results in the warehouses of retailers around the world, competing with Amazon’s “walled garden” with an “open platform” that resembles the competitive position Google took to compete with Apple’s iPhone. Incumbent players in the warehousing solutions industry should look at the long list of phone manufacturers who did not survive the clash of the titans, and adjust their strategies accordingly.