I read a lot of books.    I typically have a few books “in the queue” but occasionally I’ll run out of reading material and have to wait for my next trip to the city to buy more.   When I travel, I’ll bring the book I’m currently reading and one or two more just in case I run out during the trip.  I’ll often re-read books I’ve enjoyed so I tend not to sell or give away books I’ve already read. As a result, my book shelves are stuffed way beyond their capacity.

So, when the Kindle came out a few years ago, I watched with interest.  There were still a few bugs to work out, so I waited.  The Kindle was also kind of expensive and…oh yeah…it wasn’t available to us cave-dwelling Canadians.  The Kindle 3 was announced in July.  It had some nice new features and a great price and was available north of the border, so I decided that I would treat myself for my birthday and ordered one on August 17th.

When I placed my order, Amazon couldn’t tell me when my new Kindle would ship.  Hmmm.   Yesterday, a couple of days into September, Amazon STILL couldn’t tell me when my Kindle will ship…not online anyway.  What’s going on?  Do they honestly not know when they can ship my product?  Is the date so bad that they are afraid I’ll go elsewhere?   Those among you that have been involved in supply chain, customer service or sales, know that the only thing worse than not shipping a product on the date a customer wants it is not being able to tell the customer when they can expect it.  

So what is the cause of the delay?  Bad Forecast?  Supply issues?  We just don’t know and likely won’t know until Amazon tells us (if they ever do). Regardless of the cause, what is stopping Amazon with providing me with a delivery date?  We all know that there are supply chain analytics available that can be used to identify an expected ship date.  Perhaps, being new to the brand owner /  manufacturing side of things, Amazon might not yet be aware of what is possible with good supply chain software.

Desperate for Amazon to tell me where the heck my Kindle is, I broke down and called customer service and finally got an answer that it should ship likely on September 10th.  Why that date wasn’t or couldn’t be provided online when I checked the status, I don’t know.

As someone who understands the complexity of the supply chain, I know that sometimes we can’t ship when a customer wants a product.  However, as a customer, I want to know when you CAN ship my product.  Amazon, you gotta give your customers any insight you can….and don’t make them work for it!

Ah, enough of that….There is a great article by David M. Katz in CFO magazine about demand planning leading up to the holiday season. With all the talk of a double-dip recession, CFO’s are asking themselves do they ramp up for growth or hunker down and wait out another drought? 

Ultimately, the question is – how and what do you forecast?  There are risks on both sides if you get it wrong – excess inventory or missed sales opportunities.  As such, at times like these where there is little to no predictability, experts are now saying that the ability to respond and react to demand is more critical than the ability to forecast/plan it.  Can I hear a hallelujah!

As the article points out…

In these volatile times, it’s hard for companies to get a reliable read on what to expect from their customers and how to deal with rapid shifts in demand. Here are four steps that supply-chain experts say are essential to coping with a fast-changing economic landscape:

1.Ditch the long-range forecasts.

2. Avoid gut feeling.

3. Go granular.

4. Launch an S.O.P.  

Any other steps that should be included?

First you have to decide where the concept can be applied.  This suggests that you examine your existing processes to determine where you can address a significant portion of the normal variation with a minimum of organizational effort.  That alone can be a major stumbling block and too often I’ve heard the comment “everything is an exception around here”.  So the challenge is identifying what constitutes a meaningful exception, and in my book, a sign of a good process is where something is an exception less than 10% of the time.  The next step is to identify who needs to act, how they will be notified, and what tools they will need to address the exception.  Not a trivial job, but well worth the effort.   

Let’s use a common administrative process as an example, committing to a sales order delivery date.  For this example, your business uses either a traditional ATP process, or perhaps product lead time to automatically establish a proposed commit date.  If in 90% of the cases this results in a date that is in alignment with the customer need date, then you have the basis for implementing an exception based process.  In this case, only those orders that do not meet the customer requested date would be identified as an exception and flagged for special consideration.  All other orders would be automatically committed and confirmed with the customer.  This might be refined further to establish tolerances where the exception is only in cases where the delivery date is more than 3 days later than the customer request.

Once the exception condition is defined, an effective process for dealing with them requires timely notification (alerts) to the people who must collaborate to establish an acceptable outcome.  In today’s largely outsourced supply chain, that can be both technologically and logistically challenging.  Not only do they need to be notified of the business condition requiring their attention, but given access to the tools and information that can lead to a rapid and reliable decision.  Using the order commit process as a further example, the ability to meet the customer request date might take one of several paths;

1. Product substitution (if availability exists)
2. Production acceleration (if capacity and material availability exist)
3. Order split (if a partial order can be delivered when the customer needs it)
4. Order prioritization

A well defined exception management process would consider options in a logical sequence and within a time frame that meets customer expectations for responsiveness.

Applying the concepts of exception management to ERP action messages is an area ripe with opportunity.  I’ve known organizations that get 30,000 or more action messages following an ERP regeneration.  In those organizations it is readily acknowledged that planners will never get through the action list.  Therefore, the real question is, “Are they working on the right actions?”  In one organization, a second level analysis was performed on the action queue to evaluate the messages and prioritize them with regards to their importance and impact.  This had a huge impact on planner productivity and overall business performance.  A well designed exception management system should have that effect where ever it is applied. 

The bottom line is that I strongly recommend examining if your organization has the tools to effectively implement exception management processes.  This requires the ability to generate alerts, identify the right participants, provide the right views of information, and facilitate collaboration where needed.  The investment to put this in place will typically yield returns that are often 10X within the first year.  With the economy now on the rebound, the time is right to better leverage your organizations human capital through the implementation of effective exception management processes.

Let us look outside of supply chain for examples where the data quality is good enough to make sensible decisions.  We know that child mortality is higher in poor countries than in rich countries.  The UNICEF mission is “To reduce child mortality by two-thirds, from 93 children of every 1,000 dying before age five in 1990 to 31 of every 1,000 in 2015.”  I’m good with the first part of the UNICEF mission (reduce child mortality by two-thirds) and will continue to donate to them on this basis.  It’s the second part that confuses me.  Does is really matter that in poor countries child mortality is 93 per 1000 births or 100 per 1000 births?  I would just go with “more than 90 per 1000 births”.  I just don’t see how the precision of the statistics improves the quality of UNICEF’s decisions.  And I think too often we confuse the 2 issues of quality of decision and quality of data.

Before I am misunderstood, let me state quite clearly that data quality can always be improved.  There is no question in my mind that all enterprises should have “data police” that ensure that the data is of reasonably good quality.  But let’s all recognize that data quality is like a sales forecast.  We all need to get better at it, but we will NEVER get it absolutely right, as in complete, correct, and there when we need it.  What we want to avoid is getting it absolutely wrong.

In addition, I think that data latency is a key element of data quality. So I don’t agree with Tom Wailgum. The speed with which you receive data is a big part of its quality. I would much rather have partially correct data quickly than precise data slowly. One of the most important insights that can be gained from data is trend. Trend is often more important than the actual value, and trend is totally absent from Tom Wailgum’s discussion. In other words, data should have 3 major measures of quality:

• Completeness
• Correctness
• Timeliness

In case we forget, people are operating supply chains right now with the quality of the data they have right now. They are making multi-million dollar decisions in the long term based upon the current data. They are making 1000’s of decisions on a daily basis – expedite this PO, cancel that PO, promise this date to a customer, … – based upon the current data.  In many cases they are using paper and pencils and gut-instinct to make these decisions, and more often than not they are the right decisions. Maybe not precisely correct, but still correct. I am sure many of you have horror stories of when bad decisions were made using bad data. I am equally sure that you have horror stories of bad decisions have been made on good data.  And, by the way, I am sure you have many stories of good decisions having been made on bad data.  Above all, I am sure that many of your horror stories will revolve around having known about something too late, and many of your good stories will revolve around having known about something quickly.  The value of knowing sooner is the central lesson to be learned from the famous Beer Game that illustrates the Bull-Whip Effect.

Also, let us not confuse the quality of the decision with the quality of the data. In other words, the decision might be directionally correct without being precise, and infinitely better than doing nothing. For example, it may be correct to split a purchase order (PO) for 1000 units and expedite part of the quantity in order to meet unexpected customer demand. We may chose to expedite 500 when it would have been better to expedite 600, but expediting 500 would be a lot better than not expediting any of the PO.  The decision to split the order and expedite part of it is 100% correct.  The quality of the data may mean the difference between expediting 500 and not 600.  I can accept that imprecision better than the inaction caused by waiting for “better” data before making a decision.

Naturally, we all want to avoid the situation where it would have been better not to split the order, but because of poor data quality a decision is made to split the order. In general I think the quality of supply chain data is a lot better than that.  The reason to have “data police” is that of course no-one knows which incorrect data will lead to disastrous decisions.

If the current data says “go North East” that is good enough for me. Leave it to the “accountants” to decide to “go 47 degrees 28 seconds”, especially if it takes 2 minutes to decide to “go North East” and 2 days to decide to “go 47 degrees 28 seconds”.  By the time the “accountants” have reached their conclusion, the entire demand and supply picture will have changed anyway.

In closing, I think we should all take a word of advice from Warren Buffet when he wrote in the Washington Post that “… it is better to be approximately right than precisely wrong.” I argue that most of the time waiting for precise data is precisely wrong, and that acting quickly based upon the existing data is approximately right.

Each week that I went to the store and saw a sign “Eggo Waffles are temporarily experiencing a shortage” on the freezer door my heart started pounding thinking about the drama I would have to endure at home.  This made me think about the whole issue of the supply chain breaking down for this product that is beloved by so many people.  I can’t remember in recent history (other than Elmo at Christmas) any time where a consumer product was missed so much by so many.  How could this happen??

Obviously, you can’t predict a natural disaster or even some technical repairs, but you could prepare for both of these unpredictable events if you had the ability to simulate these types of changes and understand the financial and operational impact in advance.  With a capability to simulate demand changes or supply changes, a company could put in place backup measures to ensure that if this event occurs they could minimize the disruption to their business.  Being able to respond to unpredictable events is clearly not unique to Kellogg’s; it is something all manufacturers should think about.  The world today is very unpredictable with many natural disasters, the state of the economy, political issues, etc.; I think it would be wise for everyone to plan for the unexpected and be able to respond quickly to change… as I would hate to go through the “Great Eggo Disaster” again with any other product.

Here are some points for consideration:

Perhaps a statistical forecasting package that could produce a more accurate forecast based on historical demand patterns is enough.  I would suspect otherwise though.  With things changing so fast, historical demand alone is an unreliable predictor of the future.

Would aggregation of multiple demand inputs take out some of the bumps?  If so, then a collaborative demand planning system could be considered.   This system would capture various inputs, (such as sales projections, marketing projections, management judgement, and possibly a statistical forecast) to develop a single collaborative forecast that would provide a more insightful view of the demand picture.  But here’s the catch, in today’s volatile marketplace, your plan will never be 100% accurate.  While you might be able to reduce forecasting error, better planning isn’t going to reduce demand volatility. 

So if you can’t accurately plan demand, then you need to respond to it.  That requires you to:

• Detect changes in real-time between the actual demand and your forecast,
• Change your supply plans in response to those changes, and
• Revise your forecast when appropriate.

But is this type of supply chain response agility enough?  Is improving customer service levels while reducing inventory risk in the face of volatility an urgent need?  If so, having the following capabilities can have significant impact on your performance:

• Determine target inventory levels and resulting replenishment plans in order to handle the “lumpy” demand.
• Detect when your inventory at various locations is in excess of current needs, especially if there is a shortage in that part in another location.
• Link replenishment to satisfy demand with your supply side planning to determine viability of potential changes.
• And when your supply simply can’t satisfy the demand fast enough, allocate the supply to particular orders or distribution channels, possibly based on a combination of fair share, equal share, and priority schemes.

That leaves just one more question:  What types of users do you expect, what actions and decisions do you expect them to take, and how do you expect people to interact with the demand management system?

In answering some of these questions, one might find that what they thought they needed would only cover some of the total demand management picture.  So, it comes down “what are you really looking for?”

• Customer side equations will take prominence over rest of value chain
• Supply chains will get more integrated with marketing and service chains
• Speed and responsiveness will be key drivers for spend on new initiatives
• Cost will continue to play critical role in decision making
• Asset Management will gain more prominence and will help in accelerating “green” initiatives

I responded to Atul in the following manner:

We too are experiencing that prospects and customers are focusing a lot more attention on customer satisfaction as it pertains to on-time delivery of orders, but also to the enquiry-to quote and quote-to-order processes.

I couldn’t agree more with your third point about speed and responsiveness. Overall the trend we are observing is that consumer behaviour is pervading B2B transactions with ever shorter lead times. Coupled with the adoption of Lean and postponement strategies, companies have to be very responsive to changing demand, blurring the lines between planning and execution. These are the business drivers for your third point about agility and responsiveness.

Cost will always be a driver in supply chain management. If we adopt any of the Lean concepts it should be the elimination of waste. All too often I come across situations where the information and decision lead time exceeds the physical lead time to manufacture and/or deliver the order.

But this got me thinking about several other reports and observations that have come across my desk over the past 12 months.

First and foremost must be the article by Dan Gilmore at Supply Chain Digest highlighting the work done by Supply Chain Digest’s research arm CSCO (Chief Supply Chain Officer) Insights.  There is an excellent report titled ”Next Generation Supply Chain Management: Integrating Planning and Execution” available from this link. (Subscription required).  In the article, Dan Gilmore observes that “For many years, analysts and others have offered separate models of ‘supply chain planning’ and ‘supply chain execution’ processes, and the technology vendors were generally organized in that sense as well. You can find many diagrams that show hierarchical planning processes with no connection at all to execution, for example. The report argues, and the research supports, that this gap must be closed from a process perspective to meet the challenges of today’s supply chains.” I added the bolding because this is the key to being able to provide the speed and responsiveness to which Atul at Infosys refers.  Not only that, but also managing to contain if not reduce supply chain costs will depend on being able to reduce this gap between planning and execution.

Traditionally we have used inventory to buffer against what we would like to happen (the plan) and what actually happens (execution).  But this is no longer possible.  As the graphic below illustrates, as long ago as 2004 postpone strategies had pushed much of the inventory up the supply chain to the suppliers.  They too have adopted Lean and postponement strategies, leading to even lower inventories.  And then there is the effect of the recent recession.  Nearly all the OEM’s I speak to are struggling to secure supply of components, clearly indicating reduced inventory levels in the suppliers. I wish I had equivalent inventory figures for 2009.  Anyone willing to provide these figures?

Then there is the excellent blog written by Lora Cecere recently titled “Tackling the Black Hole in the Center of Your Supply Chain” in which she states “We now know that fixed data integration, one-dimensional rules mapping, and traditional master data techniques from ERP to Supply Chain Optimization are insufficient.  As a result, plans are created and consumed in isolation, and transactional systems hum along with little– to no — guided intelligence.”  So as the speed of business has increased – some would describe this as volatility – the supply chain systems have not kept up.

And most of the information is now external to your organization.  Companies have being trying desperately to get point-of-sale information to get early trend analysis of sales.  At the same time, many brand owners have largely outsourced manufacturing, not only lengthening the physical supply of goods, but also the time and effort it takes to make a decision.  All of these factors are only making the gaps between planning and execution even wider.  But the business need is to close this gap; to respond to demand changes quickly and effectively.  As Lora Cecere, states, the solutions from the 1990’s have not kept pace with the business needs.  Throwing more ERP at the problem isn’t the solution.  At their heart, all ERP systems are essentially accounting packages.  They deal with your data – financial and operational – but provide very little help in dealing with the majority of the information, which now exists outside of your organization.

What are your thoughts?  Do you experience this gap?  Are your systems able to cope.  Will your next breakthrough in performance come from learning to plan better, or learning to respond to plan variance?  In other words, closing this gap between planning and execution.  Robust debate encouraged.

In my first post to this blog, I want to point out three table stake capabilities that ERP vendors have failed to provide. I’ve been with Kinaxis for four weeks and have been immersing myself in the world of supply chain and S&OP trying to figure out what really drives organizations to improve their supply chain. In talking to my colleagues, which include some of the world’s most knowledgeable supply chain experts who work side by side with some of the world’s best run manufacturing companies, I’ve learned that customer service is at the top of the list.

In these volatile, post-recessionary times, manufacturers need to know when a customer order is at risk, and then they need to figure out fast how to course correct. Unfortunately the solutions in place – at one extreme, error prone manual spreadsheets, or at the other extreme, multi-million dollar ERP modules – can’t offer the following capabilities:

1. SPEED.  When a customer calls you with a potential order, how long does it take you to get back to them with a promise date?  Increasingly, customers will want feedback in a matter of minutes – not hours or days. If you’re saying to yourself “That’s impossible.” Well, it is possible.

2. RESPONSIBILITY ASSIGNMENT.   When an unexpected event puts orders at risk, do you know who to call? How many steps are required to understand and rectify the situation? Being responsive to keep a customer order ship date on time requires an immediate response whether that be to an internal resource or external supplier.

3. ONE VIEW. As you rely more heavily on outsourced manufacturing and suppliers, it becomes increasingly difficult to see and coordinate your supply chain from beginning to end. A ‘twenty-ten’ supply chain solution must be able to provide visibility across sites and suppliers.

The last capability might not stand up to the argument that it cannot be found elsewhere because given enough time and money, anything is possible. However, most manufacturers don’t have enough time and money to continue down the ERP path. Business leaders have realized that now it’s time for something different. It’s going to be an interesting decade.

This month’s challenge discusses a maker of wireless and radio frequency components that is starting to see signs of an economic recovery but is very concerned with the risks of overestimating or underestimating the resurgence of business.  They are struggling with what is needed from a process and technology standpoint to get an accurate picture of demand going forward.

We were fortunate to have Dr. Larry Lapide, a research affiliate with MIT’s Center for Transportation & Logistics,  join us in providing his take on a solution to the challenge as described below:

Hinnts Wireless has been fortunate to weather a bleak two years. As CEO, I watched annual revenues fall 20% to approximately $225 million. A maker of wireless and radio frequency components for OEMs and system integrators, Hinnts is highly dependent upon commercial and industrial markets. As they slowed, we slowed. Now we are beginning to see our sales trickle back, and our major customers indicate that within six months they expect their orders to get back to prerecession levels. In addition, we are launching a number of new products over the next six months that promise to hit new markets and bring in new customers.

While the severity of the market slowdown came as a surprise, I do not want us to be surprised when our markets rebound. Our management has always believed — perhaps, erroneously — that Hinnts has fairly level month-to-month sales volumes for all product lines, and so we relied on simple methods for scheduling production and triggering our supply chains. But as our markets get more complex, our scheduling will as well, and we cannot afford to overestimate or underestimate — as we’ve done in the past. I see now that every dollar counts, and the money we’ve lost on obsolete or discounted inventories, overstaffing, or missed sales would have made the last two years more bearable.

I would like to have a more accurate picture of our customers and their plans and forecasts, and use that information more proactively to integrate production with our suppliers and to staff our plants. With a recovery likely, is it time to invest in processes and technologies to help Hinnts and our suppliers sense and respond to demand quickly and cost effectively? Can I afford the investment, and what can Hinnts expect to gain?

What would be your recommendation?  Here are ours….

It is this re-emergence of e-commerce that has given the consumer the buying power.  Of course, the recession has had a role to play by making price a key buying criterion, possibly the key criterion.  But e-commerce allows the consumer to compare several alternatives, in terms of both brand and price, in a fraction of the time and with greatly reduced effort.  While mall shopping isn’t going away any time soon, there is no doubt that e-commerce has shifted the buying power to the consumer through ease of use and ease of choice.  For example, my daughter went to the mall with a friend and came home excited about a skirt she had seen in a store but couldn’t afford it.  My wife called her friend to get a few more details and then spent about 20 minutes on the internet finding the best deal, which was 30% less than the store price, including shipping.  It would have taken her at least 20 minutes to drive to the mall and the 30% reduction does not include the cost of driving to the mall.  And it arrived in time for Christmas.  What’s not to like about this story?

So how does this relate to our market, which is much more in the low volume, high mix and build-to-order category, rather than the high volume, low mix and make-to-stock environment typical of CPG?  As has been commented by many people before me, retail-like behaviour is being adopted in more industrial environments.  I visited a fab-less semiconductor manufacturer in late December that is wrestling with increasing demands for a much wider choice of product capabilities coupled with expectations of greatly reduced order to delivery lead times.  The lead time expectation is a lot less than the manufacturing lead time so the semiconductor manufacturer is looking at postponement strategies including, very importantly, die reservations in the foundry, which of course they do not own.  Because of a greater product portfolio they cannot afford to keep the same levels of inventory because of the associated risks of price reduction and obsolescence.  As I stated in the opening paragraph, the adoption of consumer behaviour in a business-to-business environment has been increasing over the past few years, and will only accelerate.

Perhaps it is Lora’s point about the effect Wal-Mart is having on the supply chain that best captures the impact consumer and retail behaviour is having on the larger manufacturing sector.  Because Wal-Mart is such a dominant player, initiatives enforced by Wal-Mart soon trickle down the supply chain through multiple tiers of supply and affect other industries too.  Lora selects 3 initiatives: “sustainability scorecards, rethinking inventory strategies, and the initiation of the Supplier Alliance Program.”

Let’s start with inventory.  For centuries, inventory has been used as a buffer between demand and supply, starting with grain silo’s and other food stores.  The fab-less semiconductor manufacturer I mentioned above is, like many other manufacturers, adopting postponement strategies including reducing inventories to preserve cash, while at the same time being faced by the need for shorter order to delivery times.  Obviously there is a lot that can be done to improve manufacturing flexibility and shorten change-overs, but the biggest gains are to be had in reducing the order processing times, especially the time it takes to determine if the order can be delivered on time and in full.  Given the reduction in inventory, the issue has gone from available-to-promise (promising from finished goods inventories) to capable-to-promise (determining if the products can be manufactured in time), blurring the distinction between execution and planning.

The sustainability scorecard will perhaps have the biggest and widest long term effect on the supply chain.  As stated in a New York Times article, “In the future they may also have information about the product’s carbon footprint, the gallons of water used to create it, and the air pollution left in its wake.” With the impact of environmental legislation also trickling through the supply chain, particularly the electronics supply chain, it will only be a short time before a full product sustainability scorecard will be required including “carbon” accounting.  I think it is only some time before we will have a “carbon cost of goods sold.” And, as commented on in the NYT article by Tim Marrin, associate director of external relations for Procter & Gamble, “The last thing a supplier really wants is when you’re doing a separate index for every retailer.”  Wal-Mart has the market “muscle” to see this through and to ensure a standard is adopted across the industry.  For assembled products, such as consumer electronics, this means that the suppliers to the brand owners will also have to conform.  Which is how we will see the “trickle down” effect influence the adoption of a sustainability index on labels permeate other industries.  To be fair to high tech, particularly computing, they have had a start rating in effect for some years.  But the Wal-Mart initiative will take this to a whole new level of detail and accountability.

Am I just still too full of Christmas “cheer”?