Friday, October 28, 2011

Amazon Prime – Are There Lessons for Service-Parts? - by Brian Crounse

Amazon’s stock got hammered this week, following relatively weak third quarter earnings, and it’s no longer a $100 billion company. At the time of this writing, it’s “only” worth $98B. The Harvard Business Review has some insights on why the weak earnings for Q3 may not be so problematic.

To date, I’ve talked about Amazon’s online victory, and about its problems at a warehouse in PA. In this post, I want to look at one prominent piece of Amazon’s customer retention strategy, Amazon Prime.

I got to thinking about Prime after reading this Practical Ecommerce article, which sheds some light on Prime’s successes. For those of you not familiar with Prime, here’s how it works:
  • Sign up for $79 a year (79 is a prime number. Get it?)…
  • …or it’s free if you sign up for e.g. Amazon Mom
  • Get free standard or 2 day shipping on all orders (no minimum order size)
  • Or get $3.99 overnight shipping per unit
  • Access some streaming digital media (e.g. movies) bundled for free
Prime was first launched back in 2004 or so, but seems to have really gained traction recently (the article reports that there were 2 million Prime Customers in 2009, which increased to 5 million in 2011).

The genius of Prime is that it removes nearly all friction from Amazon purchases. Once Amazon gets a customer over the hurdle to commit to the $79, there really are no more hurdles for placing orders. Customers don’t have to do the math to check whether a purchase makes sense once shipping charges are added in. They don’t have to wait to pool enough purchases together to make shipping charges acceptable (or qualify for other free shipping programs that have minimum order sizes). They don’t have to wait a week for slow-boat free shipping. They don’t have to pay much more for next-day service. Amazon easily becomes the first option for making purchases.

The consumer part of me LOVES this. The part of me that likes supporting local businesses HATES this (because I buy so much more from Amazon now). The part of me that works on customer retention strategy LOVES this. The supply chain analyst in me HATES this.

Amazon is basically willing to risk significantly lower margins (due to less shipping revenue, and higher shipping costs) in exchange for growth opportunity.

For customers paying the $79, the margin risk to Amazon is actually quite low. The chart below shows Amazon’s net profit (gross profit minus fulfillment and transport expense, plus shipping charges1) for a few different cases, assuming no change in customer behavior after joining Prime. The baseline case is the blue curve: it starts at zero and slopes up based on Amazon’s operating margins (gross margin less fulfillment and net transport costs). The green line shows estimated profits for customers that always use Super Saver Shipping. Amazon makes less money on these customers per dollar purchased, because the savings in transport costs (slow boat vs. standard) are likely lower than the lost shipping charge revenue. But the point of Super Saver Shipping is to grow sales by converting ~$15 purchases into $25+ purchases.

The purple line is for Prime customers. Note that it gets a big head start at $0 purchases. The $79 membership fee is a significant profit source for any customer that buys less than $1,000 of goods a year. That’s because Amazon is in a low margin business2. By making Prime sufficiently awesome (and having the ability to make frictionless purchases is pretty awesome), Amazon can really pump its profits from smaller customers. In fact, if you buy less than about $800 a year from Amazon, and join Prime at $79, they don’t even need you to order anything the next year. You already gave them the profit they’d get from selling you stuff.

The key thing about Super Saver Shipping is that Amazon already gives up a lot of transport cost recovery, especially for larger customers, whose average order size is very likely to be greater than $25. While the gap between the blue curve (where Amazon recovers much of the transport costs by charging the customer) and the purple Prime curve gets large for large customers, the gap between the green Super Saver Shipping curve and the Prime curve is much smaller. What this means is that Amazon isn’t really risking much with Prime. Any incremental sales volume, for almost all customers, leads to greater profit, despite higher transport costs and lost shipping charges. That’s a pretty cool strategy.

So what’s the take-home for service-parts?

It depends on the customer. Our dealers are already committed to us; they are franchises. So the part we can focus on is making purchases from us frictionless, to maintain or increase dealer purchase loyalty. How do we do this? One proven way is Retail Inventory Management (RIM). At its best, RIM makes buying from the OEM very easy. Parts arrive, as needed, with relatively little attention and labor required from the parts manager. And the best OEMs design RIM policies so that dealers can get quickly the parts not stocked by RIM, at a reasonable cost, without hoops to jump through. Kind of like the no-minimum, $3.99 1-Day shipping option for Prime.

But not all RIM implementations are perfect. They don’t always do an obviously superior job of managing parts inventory, and we don’t always make it easy enough for dealers to obtain non-stocked parts. So, many of us have room for improvement. In recent years, we’ve hosted several successful roundtable and panel sessions about RIM at our benchmark conferences. The next session for NAPB participants is our Focus Day on RIM, coming this February. This will be a great opportunity to discuss how we can make our RIM systems act as our version of Amazon Prime.

But even RIM has limited upside. Our dealers are already fairly loyal customers. The customers that aren’t so loyal are the independent repairers. What might an Amazon Prime program look like for them?

To make buying from us as easy as buying from Amazon via Prime, think of all the hurdles involved with buying parts from our dealers.
  • Repairer has to be motivated to call a dealer, rather than their usual all-brands sources
  • Repairer has to wait on hold for 15 minutes
  • Repairer has to talk with a distracted parts manager to be sure they have the right part
  • Repairer has to wait several hours for the parts to be delivered
  • Repairer has to be prepared to bicker over a return if there’s something wrong with the parts (incorrect part, damage, etc.)
I’m just dreaming, but maybe we could follow Prime’s lead and do this: For a fixed price (that could be waived under various favorable conditions), independent repairers could:
  • Access an online multi-brand, detailed catalog of genuine parts;
  • Place an order online, which is then routed to the closest participating wholesaling dealer and confirmed, with a stated same-day ETA;
  • Offer guaranteed order response times - the repairer gets a discount if the parts are late, which both the dealer and OEM eat.
Actually, I’m not completely dreaming. Some OEMs have their own commercial parts sites that have some of these features. Others are looking closely at patching into aftermarket ordering systems. OEConnection’s RepairLink with MORE is in the right ballpark, at least on paper. But the key is in the execution, so that ordering parts from our wholesaling dealers is truly frictionless.

1We’re leaving marketing, S&GA, and other non-supply chain costs out of the equation here, which means we’re assuming they are fixed costs. The parameters for fulfillment, transport costs, and shipping charges are estimates based on Amazon’s annual reports. It’s difficult to precisely estimate key items like the incremental cost of 2-day vs. Super Saver Shipping, so these curves do have some uncertainty.
2It’s non-digital retail business, at least. Its digital businesses may be another story.

Friday, October 21, 2011

Running a Warehouse—What We Should Not Learn From Amazon - by Brian Crounse

A couple weeks ago, I wrote about Amazon’s impressive growth over the past decade, and what it means for the motor vehicle service parts industry (quick summary: “Even if we can’t see the money yet with digital service customers, maybe we need to take the long view and be confident in our efforts”).

There are a number of other aspects of Amazon’s business that I think provide insights for service parts. But for this post, I want to talk about how not to run a warehouse.

You may have seen news in the past month (originally in the Lehigh Valley Morning Call) about some problems Amazon has been having at its warehouses in Breinigsville, PA. Among other things, it seems that the heat index inside the Amazon warehouses often rose above 100 degrees this past summer, which resulted in situations where workers reported seeing “paramedics bring people out of the warehouse in wheelchairs and on stretchers.” In particular, on June 3, 2011, Amazon acknowledged in a letter to OSHA that “15 out of 1,600 employees experienced heat-related symptoms.” And on June 10th, a doctor at the Lehigh Valley Hospital called OSHA to report that “several patients have come in the last couple days with heat related injuries” from Amazon.

Let’s put a few things in context. In our database of North American service parts warehouses, 65 of 225 warehouses that provided us with CY2010 safety data had zero recordable incidents over the entire year. Having one percent of your workforce suffering from overheating on a single day is simply unacceptable.

If you have ambulances on-call outside your workplace, you’ve got problems.

Such problems are unusual, even in warehouses in hotter climates. I asked a couple clients, folks who have spent considerable time in warehouse environments, what they thought. Here’s what one said:
I had seen some of the original reports on that incident and found it odd … our PA warehouse is fairly cool even in warm weather, so its not really an issue there.

What I found odd is that I have worked in warehouses in locations prone to extreme temperatures. … but processes are in place to handle it. Also, when I was at [a] warehouse in … California, [we] often saw temps over 100 F for 10+ days in a row. The key to ensuring employee safety and reasonable comfort were always the same… The issues appear to all be with how Amazon handled it, and not warehousing in general.
So, what exactly happened here?
  1. Amazon’s Breinigsville warehouses are “activity-dense.” It’s possible to estimate Amazon’s warehouse space productivity from their annual reports; the amount of dollars shipped per square foot per year are fairly high relative to the service parts warehouses we know best here at Carlisle & Company.
  2. These warehouses are also “people-dense.” 1,600 workers (more during peak seasons) in 1.6 million square feet of warehouse space is a lot. I suspect the prior tenants of these warehouses (the buildings date back to 2000, but Amazon’s only been in them since 2010) didn’t have this many people packed in them. By itself, space efficiency is a good thing, but when you have a lot of people, with a lot of machines, moving a lot of product, it going to generate a lot of heat.
  3. These buildings may not have very good thermal performance. It’s hard to say much without getting in the building and doing an energy audit, but the one thing that’s apparent from the aerial view is that these warehouses have black roofs. That’s not unusual for warehouses in the northeast, but it sure doesn’t help keep the building cool in the summer. For a contrast, if you look north a few hundred yards, you’ll find a very bright, white roof on the NestlĂ©® Pure Life® water bottling facility at 305 Nestle Way. This building was certified as LEED Gold in 2009. Or, go take a look at the warehouses near the Ontario, CA airport: all light-colored roofs.
  4. Amazon doesn’t seem to care very much about its warehouse workers. Is that provocative enough? I am basing that statement on 3 pieces of evidence: The Morning Call story, a recent WSJ article by Richard Brandt, and a recent accidental public Google+ post (key quote: “Amazon does everything wrong, and Google does everything right”). The following is an excerpt from the Brandt article.

    During the first few weeks, everyone at the company was working until two or three in the morning to get the books packed, addressed and shipped. Mr. Bezos had neglected to order packing tables, so people ended up on their knees on the concrete floor to package the books. He later recalled in a speech that, after hours of doing this, he commented to one of the employees that they had to get knee pads. The employee, Nicholas Lovejoy, "looked at me like I was a Martian," Mr. Bezos said. Mr. Lovejoy suggested the obvious: Buy some tables. "I thought that was the most brilliant idea I had ever heard in my life," he said.
It appears to me that Bezos and Amazon have a laser focus on their core strategic priorities, but don’t pay as much attention to the details as they should. Yes, I know that Bezos spent a week working at an Amazon warehouse in 2009. And that he applied the five whys in response to a safety incident in 2004. But, again, if you have ambulances parked outside your warehouse in 2011, something’s amiss.

In summary: Amazon:
  1. Pushes a lot of volume through its warehouses
  2. Does so with a lot of people
  3. In warehouses with black roofs
  4. Doesn’t always place a premium on workplace safety
Maybe these problems, at least #3 and #4, are isolated to these particular warehouses. I hope so, because fixing a couple warehouses is easier than fixing the whole company. But if these issues are Amazon-wide, Amazon needs to make some changes. Here are two suggestions:
  1. Amazon should focus more on sustainability, making it a core part of corporate culture. Why is this relevant? Because high-performance, sustainable buildings promote safety and productivity.

    Amazon is not completely remiss in this area; here’s a page that lists of a number of energy-saving kaizen projects, describes Amazon’s LEED Gold HQ buildings, and notes that four Amazon warehouses were LEED-CI (Commercial Interior) certified in 2009. These are all good things. But Amazon should and could do more—clearly, not all warehouses are high performing. In contrast, Google focuses enormous amounts of attention and resources on energy efficiency and renewable energy.
  2. Take care of its people, for real. Amazon clearly has morale and rentention problems, at least at its Breinigsville warehouses. Amazon can probably get away with this in the current labor market, but 1) it’s not a sustainable long-term strategy and 2) it’s not the right way to treat people. I am reminded of Paul O'Neill’s strategy when he arrived at Alcoa as an outsider CEO in 1987. He choose safety as his number one key performance indicator because 1) it won him support of the unions, 2) trained the organization to focus on, and achieve, specific targets, and 3) it was the right thing to do. O'Neill was able to then extend his success with improving workplace safety to improve all aspects of the business, resulting in a very successful tenure at Alcoa through 2000.
These are things that all of us can be doing, if we’re not already.

In the next post in this series, I’ll get back to something Amazon does well – retain its customers, and what it means for service parts.

Thursday, October 13, 2011

Quick & Easy Lessons from the Consumer Call Center Data Book - by Jay Cremins & Sarah Outslay

Recently, one of our benchmark participants was asked: “Why were you interested in participating in this benchmark?” Their reply was superb: “So that we can move toward more common definitions” … and answer “where we are weak or best.”

The reply was superb because improving continuously, as required to stay relevant in today’s business environment, is impossible if you fail to discover best in class practices, ideas, and processes to implement. In this particular case, the benchmark was consumer call center oriented.

Using examples from the consumer call center benchmark as a model, the remainder of this document will dissect two key facets of the reply: “common definitions” and “weak or best.”

Common Definitions

Industries are different … companies are different … business units are different … call centers are definitely different. Complexities include managing consumer-facing call centers, dealer-facing call centers, mixed-purpose call centers, a combination of two, or a combination of three. There is only one way to appropriately compare practices. Collaborating among peers to develop and agree upon common definitions.

In our recent consumer call center study, the OEM participants had highly variable consumer-facing Average Handling Time (AHT). This data was misleading.

It was not the AHT that was widely variable; rather, it was each firm’s definition of AHT that was widely variable. Collectively agreeing on a common definition allowed valid comparisons and meaningful analysis, such as the correlation between overall satisfaction and phone AHT:

Based on the data reported by participants, the correlation between reducing AHT and increasing satisfaction is tenuous. Lesson learned: AHT matters; but, it is not a primary driver of consumer-facing satisfaction.

Weak or Best

Prior to this benchmark, OEM B believed that they managed attrition at a best in class rate:

Because they chose to be subjected to outside comparison, they are now aware of opportunities to improve – and have a peer from whom they may learn how.

OEM B also discovered opportunities to improve a high level corporate goal: consumer-facing satisfaction.

Based on the study data, consumer-facing satisfaction does not correlate with:
  • AHT (see above)
  • ASA – Average Speed to Answer
However, analysis of the benchmark data indicates that consumer-facing satisfaction does correlate with:
  • Case load per agent
  • “Quality Performance” metrics
Lesson learned: AHT and ASA matter; but, case load per agent and quality performance metrics correlate with consumer-facing satisfaction.

These examples, all from the call center benchmark data book, highlight how we all use benchmarking as a fundamental strategic process “so that we can move toward more common definitions” … and answer “where we are weak or best.”

Friday, October 7, 2011

Quality Focus Day - The Power of Benchmarking - by Charlotte Williamson

Think back to preschool. What were some of the key skills you learned?
Did you include counting on that list? Maybe you remember teaching your children how to count to ten before they even began school. This is one of the most basic skills we learn as children, and it is used on a daily basis throughout life.

At last week’s NAPB Focus Day, we spent 8 hours discussing warehouse quality: damages, counting errors, wrong part errors, etc. We’ve seen companies add voice technology and scanners (and remove these tools as well), all in an attempt to be more productive and perform with higher quality. But the quality tool that came up over and over again is something you learned in pre-school: how to count. It happens to everyone, but when you lose track in the middle of counting, it can cause errors in our dealers’ orders and impact our quality and satisfaction scores. The anecdotal evidence shows that shortages tend to occur when picking quantities from 2 to about 10. The problems seem to go away when you get into the hundreds of units and can be helped by having quiet zones in the warehouse, allowing our workers to focus on the task at hand.

The NAPB Focus Days allow experts from the major heavy equipment and automotive manufacturers to spend the day sharing and learning from each other. Each expert has time to present their struggles, tools, and best practices, followed by questions from the group. The attendees came away with lists of ideas for improving quality to take back to their teams. Some ideas are easy, can be implemented quickly, and are low cost. Others will take more planning and budget, but have been tested and proven to be effective by our peers.

The solutions presented by the participants range the gamut from working with people and refining processes to implementing new technology.

One company has found great success in reducing picking errors by working with its people. All employees go through a 35-minute training to learn tips and tricks for counting quickly and accurately. They have a 3 -minute narrated version of the training for on-the-floor use at the time of counting errors. This reinforces the basic skills needed to prevent shortages and overages, resulting in almost a 50% reduction in picking errors year-over-year.

Another company found that they can improve their dealers’ receiving processes by intentionally sending them an over-shipment, under-shipment, and mis-pick and then monitoring the resulting claims. For any dealer who does not accurately report the incorrect shipments, the company will give them time to develop a plan to fix their receiving processes. In the past, dealers have tended to report shortages but not overages. Now that the dealers know they are being monitored, overage claims have become more balanced with shortages.

Sometimes technology changes can have unintended consequences. One company found that switching to voice picking not only let them improve quality, but also enabled them to monitor employees in detail and view productivity over specific time periods. This allows them to see gaps in the day where little or no work is being performed. They can use this information for discussions with employees about their work performance and have seen an immediate improvement after the employees were made aware of this functionality and their own performance.

These are just three of the many ideas and success stories shared by the 20 participating companies during 8 hours of presentations and discussions. The participants were very interested to hear the challenges that others are facing and the initiatives they are using to overcome them. This is a unique forum at which companies can interact at many different levels (strategic and tactical) and benefit from each others’ knowledge and experience. Even those who are doing well can learn from the others during the Focus Days.

We look forward to seeing subject matter experts at the 2012 NAPB Focus Days, which are being scheduled to deep dive on the topics of Supply Chain Collaboration, RIM, Pick/Pack/Sort/Ship, as well as two sessions on eMarketing.