the supply chain digest letter the supply chain...

www.scdigest.com/letter Your Monthly Digest for Supply Chain Information November 2009

THE SUPPLY CHAIN DIGEST LETTER

1

continued on page 4

continued on page 13

inside

Demand Planning 2009 - Progress, Opportunity .................. 1

The Drive to POS ........ 1

Demand Planning Thought Leaders Discussion ................... 2

Solution Profile: JDA Software ............... 3

Solution Profile: Logility ........................ 7

Solution Profile: TrueDemand .............. 12

What Demand Planning Leaders are Doing ................... 15

Demand Planning 2009 – Progress, Opportunity

Forecasting… Demand Planning… Demand Management… Demand “Sensing”…

These and many more terms are focused on the very tough job of developing plans for how much, and of what, a business is going to sell.

The stakes are high – even for medium sized companies, every 10% improvement in forecast accuracy can deliver millions of dollars to the bottom line through inventory reduction and fewer sales lost due to out-of-stocks. That’s why AMR Research has placed Forecast Accuracy at the top of its “hierarchy of supply chain metrics,” noting that “Demand Forecast Accuracy has predictive power: the extent of a company’s demand visibility can predict the responsiveness of its supply chain.”

The Drive to POS

Truly leveraging Point-of-Sale data has been something of a holy grail for companies in demand planning processes and supply chain operations.

While the term “POS” has retail connotations, the reality is that the term applies to end customer demand in whatever industry a company operates.

For many years, the challenge was that good POS data was simply not available. It is interesting to note, for example, that in the consumer goods to retail sector, actual consumption data was never the driver of such programs as Efficient Consumer Response (ECR), Vendor Managed Inventory (VMI) or Collaborative Planning, Forecasting and Replenishment (CPFR).

For most consumer goods manufacturers, the data situation is much different today, as mostly accurate POS data from retailers or third party services is largely available – though the cost, quality, timeliness, access mechanisms and other attributes still vary widely among sources.

TheSupply Chain Digest

Letter


™LetterThis Month: Demand Planning

November 2009 Your Monthly Digest for Supply Chain Information www.scdigest.com/letter


2

Demand Planning Thought Leadership Discussion: Getting Closer to Actual Customer Demand

Supply Chain Digest’s Dan Gilmore recently spoke with David Johnston, Sr. Vice President of

Manufacturing and Wholesale for JDA Software, on several themes related to Demand Planning.

The Q&A below is excerpted from a video discussion on this topic, which can be viewed at

SCDigest’s Demand Planning resources page: www.scdigest.com/demand_planning.php.

continued - page 3

Gilmore: Let’s take a couple of key areas of demand planning, starting with the consumer goods industry and data. While POS data has been available for a number of years, relatively few companies have really leveraged that data. Why is that, and is that changing?

Johnston: That’s a great question. I think there was a rush to just consume the data when it was made available without a good plan around what kind of value that data could provide. There were a lot of failed projects because there was no plan in place.

The other barrier was that there just wasn’t the technology available that could scale from a manufacturer’s perspective to take that data and truly integrate it back into demand planning systems and use it to make a better forecast.

That has clearly changed. Now you have the ability to take POS sales data and manage it at the right level of detail to get an early read on the trends in the market and then compare that against the shipment plan that you have in place to drive your supply chain.

Gilmore: What other changes are you seeing in demand planning?

Johnston: The other part of the process that has changed is this focus on consensus demand planning. What that means is that you now have to get traditional supply chain planners synchronized with account teams. They are the closest to what’s happening in the field and the programs that are being run for trade funds spending - the programs that are going to be the drivers of demand.

With the integration of POS sales data and making that data available, the demand planning process becomes less of a forecast based on what happened in the past and more of a focused prediction of what’s going to change moving forward. That POS sales data provides critical insights the account and sales teams didn’t have before.

Gilmore: Is that what we really mean with this term that’s been out there quite a bit lately called “demand sensing?”

Johnston: That’s part of it, but not all of it. Demand sensing is certainly about providing more of an automated statistical view of how you can leverage the most current understanding of demand. So, for instance, if I know this week what my ordering trend looks like next week given my lead times, I can use that visibility to change a short term forecast based on the understanding of that portion of the demand.

Gilmore: OK, that’s the view from the consumer goods side. What’s happening in retail with regards to demand planning?



3

Demand Planning

SOLUTION PROFILE

SCD

(Continued from Page 2): Thought Leaders Discussion... Demand Planning

SOLUTION PROFILEJohnston: Retailers are increasingly focused on what is selling at the shelf, and as such have been focused on what I would call “optimization of replenishment execution.” They are looking to put in place the most cost-effective replenishment strategies.

They are also trying to answer questions such as how do they best leverage favorable pricing brackets, and how order plans can optimize logistics costs.

Retailers are also looking to ensure that order forecasts account for promotions, new product introductions into specific markets, resets, assortments, and other factors in a more comprehensive way than they have in the past.

Retailers are making investments now into the technology solutions required to get to a time-phased view of orders, very similar to what manufacturers have been doing for years. Again, a barrier has been software scalability. There have been very few products that can scale to the volumes required in retail to provide a time-phased view of order projections that can take into account all the different variables that drive those forecasts - promotions, new product reductions, resets, store openings, stores closings, etc.

A lot of retailers are making strides toward having a time-phased view of order projections. They are now going to be in a position to provide not just POS sales data, but to actually provide that forward looking view into the orders that can truly complement and supplement the shipment forecasts that manufacturers have today down at the consumer level.

Gilmore: Yes, and I think part of it is the technology as you said, and part of it is that you are seeing a lot of retailers that are evolving from being almost primarily merchants to also being supply chain focused companies.

Johnston: Yes. If you look at the retail channel masters, without using any names, those retailers have grown to the size and scale to have a competitive advantage in supply chain. Now it’s those same companies that are moving to the next level.

JDA’s Demand Management solution set is designed to enable retailers, manufacturers and wholesaler-distributors to more profitably anticipate, create and satisfy customer demand. This comprehensive suite integrates JDA’s proven planning, analysis, optimization and execution capabilities with collaborative workflow for seamless integration and interoperation spanning a wide range of processes across the consumer-driven supply chain. With Demand Management, your decision makers can work together to reduce operational costs and increase top- line revenues across the enterprise and down to the store level. JDA’s Demand Management solution is comprised of JDA Demand, JDA Seasonal Profiling, JDA Demand Decomposition and JDA Demand Classification.

Key Customers:

Avon Products, Inc., IKEA, Defense Logistics Agency, Church & Dwight, Macy’s, K-Swiss, Kraft Foods,

Lowes, OfficeMax, Tyson Foods, Sara Lee, Black & Decker, Unilever

Contact Info:

www.jda.com or [email protected] 1 800 479 7382 or 1 480 308 3000.

Featured Collateral:

• White Paper: Managing Demand Volatility, Exceeding Customer Satisfaction Levels: A Look at What Progressive Retailers are Doing

• Critical Strategies for Building an Agile, Responsive Supply Chain to Drive One Synchronized View of Demand

• Case Study: Tyson Foods Increased Forecast Accuracy by 15%

Available at the JDA Software web site or the Demand Planning Resources page: www.scdigest.com/demand_planning.php



4

continued - page 5

Should forecast accuracy really be at the top of the metric hierarchy? That is perhaps a subject for debate, but clearly the level of forecast accuracy has a dramatic impact on a company’s supply chain and bottom line performance.

The importance of demand planning has been crystallized in this period of economic downturn, as many of the rules companies thought they knew about forecasting were turned on their heads, as companies struggled to cope with an amazingly volatile business environment.

The importance of more accurate forecasts is certainly well understood – but getting there is never easy.

“Forecasting has never been an easy discipline to apply and communicate,” says Mark Lawless, a Senior Consultant for the Institute of Business Forecasting & Planning. “It may be one of the least understood (though most pervasive) among the business decision making tools in use by managers at all levels in the organization.”

Demand Planning – the Basics

“Demand Planning” is an extension of “forecasting,” and implies a more comprehensive approach to predicting what customers are going to buy.

Demand planning is all at once a concept, a process, and a category of supply chain software – many solution providers use “Demand Planning” or some derivative of that term in the names of their forecasting modules.

From the outset, supply chain management has moved towards the vision of being “demand-driven” – in which the entire supply chain is pulled by actual consumer demand. But most companies are not there yet. In fact, few have actually reached high levels of maturity in being able to leverage “Point of Sale” data, whether that is in a retail store or wherever final customer demand may be.

Even for companies in a “build-to-order” supply chain model, where production is driven directly from customer orders, forecasting remains critical to drive

(Continued from Page 1): Demand Planning 2009...

The Hierarchy of Supply Chain Metrics

WHAT IT TELLS YOU

• Perfect Order Predictor

• Performance Tradeoffs

• Cashflow Health

• Customer versusSupplier Balance

• Root Cause Analysis

• Surgical intervention

WHAT IT IS

• Demand Visibility

• Responsiveness• Cost/Margins

• Cash-to-Cash

• Operational Effectiveness

TO

P T

IER

MID

LE

VE

LG

RO

UN

D L

EV

EL

CostDetail

ProductionScheduleVariable

PlantUtilization

WIP & FGInventory

Order CycleTime

PerfectOrder Detail

DirectMaterialCosts

Purchasing Costs

RawMaterial

Inventory

SupplierOn Time

SupplierQuality

DPO InventoryTotal

DSO

PerfectOrder

SCM Cost

DemandForecast

Source: AMR Research 2008



5


supply chain strategy and operational and tactical supply plans.

Of course, demand planning is done at multiple levels – long-term, mid-term and short term. The granularity and precision needed at each level can vary substantially. For example, the longer-term the forecast, the less the “mix” of the forecast is important; mix is usually critically important in shorter term forecasts, however, impacting not only the ability to meet demand but also a company’s profitability.

One thing seems very clear: being “adequate” versus “excellent” in demand planning processes is a recipe for mediocrity in overall business performance as well.

How Mature are Demand Planning Processes Today?

The answer to that question depends on whom you ask.

While forecasting has probably been around since business began, and technology-enabled forecasting for some 30 years, the evidence is that many companies still have a long way to go. In fact, SCDigest has found that in an effort to go “lean,” many companies have actually reduced their staffing levels in the demand planning area over the past few years, and in some cases taken something of a step backward in their demand planning skill levels and results.

That was a luxury that could perhaps be modestly afforded when times were good, but when the financial crisis and Great Recession hit, it was “back to basics”

for many companies in terms of their demand planning processes. Companies suddenly worked very hard to get a lot more granular in their forecasts and closer to end consumer demand points.

Our research shows that companies tend to go through distinct phases when implementing new demand planning strategies, which often look something like this:

Phase 1: Process Development – Companies develop more standard and rigorous processes for creation of forecasts/demand plans, often using the “work flow” tools of today’s demand planning software to help electronically enforce those processes. The goals are usually to simply get better at demand planning and to shorten the forecasting cycle.

Phase 2: Single Number Forecast – In phase 1, companies may or may not have evolved to where a single number forecast drives demand and supply plans. But after getting the basics down, this is generally the next phase. Usually to get there requires integration of the demand planning activity into a (new) Sales and Operations Planning process, where “bottom up” and “top down” plans are reconciled.

Phase 3: Greater Statistical Sophistication - Companies begin to get more mature in their use of demand planning software, using some of the more advanced algorithms and other techniques available to increase forecast accuracy of specific categories of SKUs or to solve specific problems. Most companies in fact only

continued - page 6

“SCDigest has found that in an effort to go “lean,” many companies have actually reduced their

staffing levels in the demand planning area over the past few years...”



6


scratch the surface of what their demand planning tools can offer.

Phase 4: Trading Partner Integration – Collaborative demand planning with customers is widely embraced and those trading partner forecasts are more explicitly brought into the demand planning and S&OP processes; there is also more collaborative sharing of demand plans with suppliers.

Phase 5: Demand Management – Companies become skilled at not only predicting and responding to demand, but more actively shaping it.

What is this concept of Demand Management really all about?

Companies increasingly recognize that forecasting – especially statistical forecasting – is really only one part of the demand management process. In a nutshell, forecasting is about predicting demand; demand management is about influencing demand, and ensuring that the key functional areas of the company are tightly integrated with executing against that demand plan.

Said another way, demand management is all about coupling true market demand with the potential for

Demand Planning Maturity

Ventana Research recently described a “maturity model” for demand planning, and surveyed more

than 200 companies to assess where they stood in terms of demand planning practices.

A summary of the interesting results are below:

Tactical – The company does not have an integrated sales forecasting and demand planning process.

The information that the processes produce is not very accurate or timely, and the responsibility for

managing the process and providing the plans often is left to people not directly in touch with buyers

and customers. 38% of respondents fell into this category.

Advanced - The company has taken steps to make the process more integrated and incorporates

a wider range of data in developing its forecasts and plans. More of the right people are involved in

producing the plans, and the company produces them faster and in shorter cycles. Yet accuracy still

lags, and the plans and forecasts are not timely enough. 30% of respondents fell into this category.

Strategic - The company has integrated sales forecasting and demand planning. It brings the people

closest to the customers into a highly accurate process. The company

emphasizes the importance of accuracy by measuring and rewarding it.

It collects detailed information to use in the forecasting and planning

efforts so that it can identify more of the root causes for exceptions

that occur. 24% of respondents fell into this category.

Innovative - The company has a collaborative sales forecasting

and demand planning process that incorporates the people

who are most knowledgeable and best equipped to manage the

process. It reforecasts and replans frequently, and because it has

developed this skill, it is able to do it within the shortest possible cycle

time. Only 9% of respondents fell into this category.

continued - page 8



7

Demand Planning

SOLUTION PROFILEDemand Planning

SOLUTION PROFILE

With more than 1,250 customers worldwide, Logility is a leading provider of collaborative, best-of-breed supply chain solutions that assist any-sized company realize substantial bottom-line results in record time. Logility Voyager

Solutions is a complete supply chain management suite that provides supply chain visibility; demand, inventory and replenishment planning; Sales and Operations Planning (S&OP); inventory and supply optimization; manufacturing planning and scheduling; transportation planning and management; and warehouse management. Voyager Solutions automates the process of reliably predicting market demand, new product introductions, promotions and inventory policies. As a result, you can build plans that are totally attuned to the market.

Key Customers:

Brown Shoe Company • Electrolux Home Care Products • Intertape Polymer Group • Lance, Inc.

Pernod Ricard • Verizon Wireless

Website and Contact Info:

www.logility.com • [email protected] 800-762-5207


• Demand Planning Case Study: Lance, Inc. • Demand Planning Case Study: Intertape Polymer

Group

Available at the Logility website or the Demand Planning Resources page:

www.scdigest.com/demand_planning.php

Time to Enhance Demand Planning is NowBy Karin Bursa, Vice President, Marketing, Logility, Inc.

In a dynamic economy, the need for a synchronized demand-driven supply chain becomes even more critical. Although it is a natural reaction to pull back to just the basics for survival in the current global economy, smart companies will be more strategic by investing in their supply chains to improve business processes and leverage technology. Especially in a recession, you must be able to respond quickly to dynamic demand and provide consistently high in-stock service and on-time deliveries. By focusing on SCM initiatives that increase demand visibility, companies can more accurately forecast demand, efficiently deliver finished goods, and consistently meet customer service goals with less inventory and higher profits.

Boost Visibility with the Right Demand Planning Solution

The correlation between profitability and the ability to effectively manage supply and demand is clear. Research indicates that one week’s inventory advantage translates into a 1% cost advantage—giving you a competitive advantage that can be measured by lower costs, better product availability and improved profitability. This means gaining demand visibility and creating a supply chain network that can quickly and flexibly meet rapidly changing market demand should be a top priority. To become more demand-driven, better demand visibility can be generated combining multiple demand signals, including forecasts, actual consumption data, syndicated industry data, and collaborative planning between supplier and customer.

Demand planning solutions such as Logility Voyager Solutions™ can give your entire supply chain access to critical and accurate demand visibility and provide the foundation for sales and operations planning (S&OP). Leveraging a portfolio of forecasting models to automatically generate forecasts at multiple business levels, from sales and marketing to logistics to financial, you can boost service levels, shorten cycle times, reduce inventory investment and minimize obsolescence.

Visibility and Accuracy Lead to Dramatic Improvements

Lance, Inc., one of the largest producers and distributers of well-known snack food brands, implemented Logility Voyager Solutions and put a demand planning process in place with three objectives in mind: improve forecast accuracy, reduce days-on-hand inventory and streamline the executive S&OP process. Lance knew that more accurate forecasts would lead to meeting the company’s goals of reducing working capital, improving operational efficiencies, increasing freshness at the shelf and optimizing inventory investments.

With Logility, Lance’s forecast accuracy increased from 50% to an impressive 70%, a figure which is considered best-in-class in the snack food industry. Better demand visibility has also led to a significant reduction in days-on-hand inventory (a critical measurement in the perishable food market), and a streamlined S&OP process has driven numerous additional benefits across the business.

The success of every company depends on its ability to profitably meet customer demand. SCD



8

(Continued from Page 6: Demand Planning 2009...

capturing a given share of that demand, and then agreeing on a plan (based on financial and strategic imperatives) that will be used as the basis for tactical execution activities across sales, marketing and the supply chain functions. The paradigm changes from responding to demand to influencing demand consistent with the company’s goals and objectives.

While this integrated process (captured largely but not completely in the concept of Sales and Operations Planning) will substantially improve the results,

forecast inaccuracy will still be present: things change, competitors respond, etc. Companies must therefore consistently tune those forecasts and execution plans as those changes are understood.

At a recent conference, Dr. Larry Lapide of MIT, a recognized expert on demand planning and S&OP, noted a fundamental problem: in many companies, the gulf between sales/marketing and supply chain operations is still very wide. Despite the strong adoption of S&OP, which has closed that gap from a little to a lot, depending on the company, the “demand side” and the “supply side” are still often viewed as distinct disciplines that touch but don’t fully intersect.

The notion of “demand management,” requires, however, coordinated decision-making among supply chain, marketing, sales and customer service functions.

It’s not an easy challenge. This need for balancing partly conflicting goals and achieving integrated decision-making must be realized across long-term, medium term, and short-term (sometimes near real-time) horizons. Full realization of the demand management concept implies not just chasing revenue and sales growth, but making optimal financial and market decisions considering all factors – a level of discipline difficult for most companies to pursue.

Demand Planning Process and Technology Trends

In researching this issue of the Supply Chain Digest Letter, we’ve identified several important process and technology trends relative to demand planning:

Rise of “Demand Sensing”: Maybe it is surprising, but it’s only in the past couple of years that companies have recognized they needed to get closer to what is actually happening at the end customer, at the time it happens.

Hence, the birth of the concept of “demand sensing,” which AMR Research defines as follows: “Demand sensing is the translation of downstream data with minimal latency to understand what is being sold, who is buying the product, and the impact of demand-shaping programs. These three demand elements are then translated into requirements to craft a profitable demand response through internal processes for demand translation.”

Stated another way: demand sensing is about reducing the latency of the feedback loops from actual customer demand and using that intelligence to adjust production, replenishment and promotional plans in a much more responsive way than has traditionally been the case.

While that view of demand sensing has something of a consumer goods connotation to it, the reality is that companies in many sectors have an opportunity to recognize near-real-time demand activity and adjust execution appropriately.

A growing number of software vendors have started to provide “demand sensing” capabilities, which are often used to adjust their near-time forecasts based on actual customer pull. Such tools are touted as being capable of substantially reducing short term forecasting error, with the expected benefits on inventory levels and the ability to meet “true demand.”

continued - page 9

“...demand sensing is about reducing the latency of the feedback loops from actual customer demand and using

that intelligence to adjust production, replenishment and promotional plans in a much more responsive way than has traditionally been the case.”



9


continued - page 10

There are other aspects to the concept of demand sensing that are likely to gain traction. For example, estimating the price that potential customers would be willing to pay for an existing or new product, or understanding the “latent consideration sets” of different customer segments. This will get especially interesting as companies in developed countries increasingly look to tap into developed, developing, and emerging markets simultaneously.

“Decomposing Demand”: For many years, companies have been trying to better understand what actually causes and drives demand. These so-called “causal factors,” which can range from the weather to macroeconomics and everything else in-between, would greatly enhance a company’s ability to improve its forecast accuracy as well as potentially increase sales – if it could be done.

History says many companies have goals to get there, but somehow never quite make it.

That is starting to change.

We like the concept of “decomposing demand,” as shown in the graphic nearby. The goal with this approach is to not only predict “how much” as part of the demand planning process, but also to understand the “why” of different demand levels.

While some companies have tried to incorporate dozens of causal factors into their demand planning processes, experts say focusing on a smaller number (3-10) of the most powerful demand drivers may in the end produce better results.

Demand planning vendors are offering increasingly sophisticated tools to decompose those demand drivers, and you can expect to see more vendor focus on this area over the next few years.

Promotions&

Advertising

Halo

Seasonality

PAST FUTURE

Pull-forward+ Competition

Cannibalization

Sales HistoryBase Forecast

EVENTS

PRICING

EVENTS

PRICING

“Decomposing” Demand to Understand Drivers & Opportunities

Source: JDA Software



10

Focus on Qualitative Factors: Companies that have taken statistical forecasting about as far as they can go often come to a similar realization: what is missing are better insights into qualitative factors that can be used to enhance the statistical forecast.

These companies are constantly looking for insights that are not easily quantifiable or “in the data” that might impact actual demand, whether that is customer behavior, competitive activity, or other factors.

Sometimes, however, the focus on qualitative insight can be taken too far. In the consumer goods supply chain, for example, some companies that started to rely heavily on the forecasts from retail account teams to drive the final forecast numbers in fact found they were often better off using the statistical forecast from the demand planning system versus the overly optimistic forecasts from the sales teams.

The right answer is probably a balance – looking for opportunities to improve the forecast through qualitative input – but appropriately respecting the power of the system baseline forecast as well. Look for where the qualitative data can really add value.

New Category of “Super Users”: Some companies have started to carve out a new role of “super user” among demand planners.

What do these “super users” do? They are typically not responsible for day-to-day demand planning activities. Rather, they are focused on improving overall demand planning processes, looking at how the demand


planning software system can be “tuned” to deliver better statistical results, and maintaining overall demand planning system parameters.

This new position offers several advantages. First, it provides an additional career/path option within the forecasting group. Second, it should provide a platform for continuous improvement by focusing a small group or even a single individual on that task, while letting “regular” demand planners focus on getting the forecast right, rather than system issues.

Forecasting the Hard to Forecast: The so-called “Long Tail” phenomenon has led many companies to have an increasing number of low volume, very difficult to forecast products. The general trend towards product proliferation has led to similar challenges, as demand planners are faced with a growing array of new products for which there is no history.

Demand planning vendors, however, have come up with increasingly creative ways to deal with these challenges. So called “probalistic optimization,” is one approach, especially for slow moving items.

Software solutions also offer increasingly sophisticated ways to forecast new or otherwise “unforecastable” products, such as tools that allow planners to create a profile based on combinations of histories of several products thought to have similar characteristics. Once some sort of statistical baseline is established, it can be refined through additional planner knowledge and judgement.

“Demand planning vendors, however, have come up with increasingly creative ways to

deal with these challenges. So called “probalistic optimization,” is one approach,

especially for slow moving items.”

continued - page 11



11

You’ll find a wealth of resources on our web site: www.scdigest.com

Including: White papers • Video • Case studiesExpert columns • Supplier brochures

You’ll find the information and insight you need to better understand managing your Demand Planning!

Demand Planning

Resources

Resource web site: www.scdigest.com/demand_planning.php

SCD

Summing it Up

Demand Planning has never been more challenging – nor more important.

Given that fact, has forecast accuracy really improved? The data doesn’t completely support that notion, though clearly many companies have individually achieved outstanding improvements in forecast accuracy.

One might also say that given the challenges of SKU proliferation, more dynamic demand, increased reliance on new product introductions and promotions, and other factors, maintaining existing levels of forecast accuracy is in some ways something of an accomplishment – but at the same time, companies cannot be happy with the status quo, and must continually reach to improve forecast accuracy and process excellence.

The “leaning” of demand planning staffs is something companies should be concerned about, with planners now clearly managing many more SKUs than in the past – in some cases, many, many times more SKUs. Given the value of a good demand plan, we think companies really need to look at the value of adequately staffing demand planning teams. That said, if this is the trend,

then automation, management by exception, and other approaches that can be enabled by robust demand planning technologies will be key to being both lean and good.

We’ll end this section with a short story and a quote.

A couple of years ago, MIT’s Lapide wrote about this anecdote: The CEO of Anheuser-Busch looked primarily at one metric when it came to demand management – the level of interplant transfers. Though at first blush this appeared to indicate a relative lack of focus on the topic, upon further review it was brilliant in its simplicity: if transfers were high, it meant plants were producing too much or too little of a given product, or sales territories or regional promotional plans needed to be revised.

Sometimes, there is merit in looking for some simple answers.

Finally the quote, from Albert Einstein, which we think relates directly to the fact that forecasts are almost by definition likely to be wrong – but crucial nonetheless.

“All models are wrong,” Einstein said. “Some are useful.”




12

SCD

Demand Planning

SOLUTION PROFILEDemand Planning

SOLUTION PROFILE

TrueDemand empowers retail account teams of consumer goods companies to achieve a dramatic and sustainable lift in sales. The solutions and services identify the root causes of lost sales at the shelf and enable swift action to align product availability with true customer demand.

Using daily point-of-sale (POS) data from retailers, TrueDemand helps you coordinate your supply chain team, your account team, and your store operations team to meet the consumer demand at the shelf and the service level requirement of the retailer at the lowest inventory investment.

Key Customers:

A partial list of customers includes: Conagra Foods and Kimberly Clark

Website and Contact Info:

www.tdemand.comUmur Cubukcu: 408-399-1924, x244

[email protected]


• White Paper: Can You Market a Hole on the Shelf?

• On-Demand Webinar: Solving the $100 Billion Retail Sales Execution Challenge (with Conagra Foods): http://resources.tdemand.com/2008/10/10/webinar-the-execution-advantage/

Available at the TrueDemand web site or the Demand Planning Resources page:

www.scdigest.com/demand_planning.php

Executing to True Demand

Eric Peters, CEO, TrueDemand

Thirty percent of the time the shopper does not know what brand they are going to buy until they are standing at the shelf. An empty shelf is a lost sale. But that is more than one lost sale. Retail ordering systems will notice that sales are dropping, but not a reason why. Because future retail orders are influenced by recent sales patterns, the retail orders get reduced. And then, you enter into a sales death spiral that is difficult to reverse. Smaller orders equate to lower service levels at the shelf, lower service levels result in more lost sales and so the cycle goes. It is more critical than ever that the product is on the shelf, in the right condition, at the right place and priced to meet that consumer demand. Most demand planning systems and processes in the market today evolved before the widespread availability of shelf level information. Even though many of the systems allow forecasting at the major customer level, the forecasts are based on historical information together with high level human input. They do not have the ability to utilize the real time visibility provided by retailer data and incorporate that data to make adjustments in the short term forecast. They are slow to react, causing lower service levels and higher inventory. They do not comprehend which levers to pull to respond to the retailer and consumer decisions – the ability to influence the retailer’s forecast, pricing, and replenishment parameters, the ability to utilize store presence to make adjustments at the store and the ability to utilize VMI relationships to react faster to shelf level changes.

To take full advantage of this new information and these new opportunities, companies must be able to understand “what is the true demand versus the sales history of an item”. How do you intelligently manage inventory in the tightest band, while maintaining the highest levels of service? What are the root causes of problems and how do you quickly react to resolve these problems? And of course, how do you create the most accurate daily forecast of item demand at the store, at the retailer distribution center and at the manufacturing distribution center? By taking shelf and retailer demand information and driving it back through the supply chain, you truly can improve your chances that the shopper will buy your product when they are standing at the shelf.



13continued - page 14

(Continued from Page 1): The Drive to POS

That data availability isn’t necessarily common in other industries. In preparing for this report, we spoke with a VP of Supply Chain for a high tech components company who said a major customer, one of the largest companies in the industry, still simply refused to share actual POS data with her company. It would only share its own forecasts. Many other trading partner relationships fall into the same category.

In other cases, POS data is a challenge for an entire industry. It is virtually impossible, for example, for medical device and supplies manufacturers to see demand beyond the distributors through which they sell. Actual consumption at healthcare facilities simply isn’t available.

As POS data has become available at many companies, the question then becomes: “What do we do with it?”

The reality has been that for manufacturers selling through large retail outlets, the technology wasn’t really available for well-leveraging this mass of data by trying to forecast each SKU at each store. The problem was just too big.

That has changed just recently, as several supply chain planning vendors, using a new approach to the technology, seemed to have cracked the processing nut.

However, knowing how to fully leverage POS insight, and changing business process and relationships to realize the potential results, is another matter.

This discussion leads to two important principles worth considering:

1. Forecasting should almost always be initially about future end consumer/customer demand, not future shipments. History-based forecasting techniques have usually focused at the shipment level, since that was the best data available. That may or may not accurately reflect true customer demand patterns.

Some argue that the tendency to focus on shipments rather than true demand is often exacerbated by having the demand planning function housed within the supply chain group, which tends to think inherently in terms of orders and shipments.

“In our experience,” wrote consultants Colleen Crum and George Palmatier in the classic book Demand Management Best Practices “it is not unusual for the demand forecast to be biased toward the supply organization’s ability to produce and deliver.”

Of course, the difference between orders and shipments from channel partners and what is actual customer pull is at the root of the “Bullwhip Effect.” First noted some nearly 20 years ago, industry-wide progress against that supply chain phenomenon has been modest at best ever since.

2. It is important to understand what actual customer demand is, not just what was sold or shipped. A recent NRF study suggested manufacturers lose up to 15% of potential sales from a variety of supply chain woes, ranging from out-of-stocks to poor promotional execution to damaged product on the shelf. Actual demand is really the combination of realized sales plus lost sales.

In additional to solving the challenges that cause lost sales, an important point is that these lost sales can become self-fulfilling prophesies. As stock outs or other issues cause lost sales, the system or planners may view this as a drop in customer demand, which causes replenishment changes that further depress sales, until real demand itself has artificially been significantly cut.

Leveraging the Data

Black and Decker’s Hardware and Home Improvement division is one company that has started to really leverage POS as part of its demand planning processes – and achieved substantial benefits.

“...lost sales can become self-fulfilling prophesies. As stock outs or other issues cause lost sales, the system or planners may view this as a drop in customer demand, which causes replenishment changes that further depress sales, until real demand itself has artificially been significantly cut.”



14

SCD

(Continued from ): CO2 Emissions...

“We can compare forecasts, shipment history as well as POS and order history for any of our SKUs at any given time,” says Scott Strickland, vice president of information systems for the division. “At the end of 2007, this resulted in a 10.4% improvement in forecast accuracy.”

So what are the keys to leveraging POS data effectively? Here are a number of principles:

Get the Data if it is Available: You can’t use what you don’t receive. Leading consumer goods manufacturers get as much of 90% of their sales volume reported through a combination of retail and third-party POS data feeds. While there is some debate as to whether the cost of certain POS feeds may be worth it, we believe the clear trend is to capture as much POS data as is available.

Use the Data Collaboratively: POS data is generally the basis for a retailer or other channel partner’s own forecast. Using this as the basis of discussion and collaboration with partners can pay big dividends by itself in synchronizing plans and inventory levels. In retailers, POS forecasts serve as the “language” used internally; manufacturers should want to be reading from the same source book when working collaboratively.

Focus on Data Quality: There can be issues with POS data quality, and having someone keep tabs on that is critical to ensuring the effort to use POS is not impacted significantly by bad data. Ditto with failure to receive POS data transmissions – it still happens, and companies must be able to recognize that failure and be able to automatically handle and integrate any retransmissions. Data sent as spreadsheets – still common – is often especially prone to data quality issues.

Develop Standards for Data Normalization: Channel partners may report the same basic data in myriad ways. A common problem is different definitions of “week,” for example (Sunday to Saturday, Monday to Sunday, etc.). Since there is no standard for this data, companies must develop their own approaches to normalizing the data so that it will “line up” correctly. Consistency in representing the data is key, even if some compromises have to be made.

Focus on Identifying Inconsistencies between Shipment and POS Data: A first place to start in historical or future analysis is where shipment plans for

a product seem out of sync versus POS forecasts. Most demand planning systems should be able to automate this process, setting parameters for exception alerts.

Drive POS Data to the Top Place in Forecasting Demand: Will POS be used to simply augment traditional forecasts, or will it be the predominant driver of the final forecast? Demand planning leaders are tending towards the latter state.

Leverage the Data in Multiple Ways: POS data can provide a rich set of information that can be leveraged in multiple ways to gain additional consumer and market insight beyond the base forecast. There are obvious areas, such as the ability to better understand the impact of promotional activities and the behavior of new product introductions. It is critical that this insight not be “lost” for future promotions or products. Other opportunities include spotting leading market indicators sooner, and using POS data to drive internal cross functional alignment by focusing on actual consumer behavior as the focal point of decisions.

In the end, whatever industry you are in, a change of mindset is often required, where the goal of the organization becomes “shipping to consumption” - and then working internally and with channel partners to make that happen.

One approach to that is to change accounting standards, either formally or informally. That is the tactic Panasonic Consumer Electronics used a few years ago, when it changed its sales compensation to trigger commissions and quota coverage only when its products moved from a retailer’s DC, not when they were shipped to the retailer. ON Semiconductor took an even more radical step, changing its own revenue recognition standards in a similar way, booking sales only when its distributors sold the company’s products (“sell through” model) not when the product was shipped to those channels (“sell to” model).

For virtually everyone, the following principles are true: (1) Effective use of POS data regardless of industry will result in better forecasts; and (2) By comparing supplier shipments to retail/customer “takeaway,” both parties can identify opportunities to improve order patterns that at once reduce the Bullwhip Effect, inventories and out-of-stocks for both the manufacturer and channel partners.



15

What Demand Planning Leaders are Doing

Demand planning is a difficult and complex process. Below you will find a series of short snapshots of steps four companies have taken to improve forecasting and demand planning.

Heinz Insists on 100% Dedication to Accuracy

As food giant Heinz was moving to improve and professionalize its forecasting process a few years ago, a focus on “100% accuracy” began to permeate the process – a change of culture driven in part by the CEO.

Heinz’s CEO stated frequently that forecast accuracy was now of paramount importance not just to the supply chain but for the entire Heinz business. Previously, “under promising and over delivering” was often the accepted norm. That approach could no longer be tolerated, the CEO made it clear; also out was “hiding bad news” as long as possible, hoping a “Hail Mary” pass at the last minute would enable someone to meet their forecast numbers.

“We insisted that forecasting volumes be based on solid facts and data with a 100% dedication to accuracy,” a Heinz manager of forecasting and demand planning recently wrote.

Heinz also uses consumption at retail as the foundation of its forecast, later translating that POS consumption data into a shipment forecast.

Brown Shoe Synchronizes Offshore Factory Schedules with Consumer Demand

Brown Shoe Company sources most of its shoes offshore. While the company had a good process for forecasting shipment needs from a mid-term perspective, it did not have the capability to adjust orders to suppliers based on what was actually happening at retail.

Today, Brown Shoe closely collaborates with its Famous Footwear retail store chain on inventory planning by obtaining POS data on weekly sales in order to get retail replenishment as close to true market demand as possible.

The system rolls forward weekly and recalculates the forecast for each item using the most recent week of POS data. New demand planning software generates forecasts for all of the retail items. Weekly POS data is also fed into Brown Shoe’s “In-Season Replenishment” system, which calculates inventory needs and adjusts the receipt flow based on how the item is performing in stores.

Brown Shoes is now able to make in-season adjustments to orders from its suppliers. Forecast accuracy is around 80% 13 weeks into the future, a critical time frame because that is when the factory needs to cut material. Overall, this approach has reduced the lead times by about 50 days for core Brown Shoe items selling at Famous Footwear because orders went from being monthly to every other week, resulting in smaller quantities.

Enterasys Tracks “Value” of Forecasts Adjustments

Enterasys Networks, part of industrial giant Siemens, is a leading provider of computer network solutions.

As part of an ambitious initiative to improve both demand planning and Sales & Operations Planning processes, the company decided to focus hard on the “value add” of adjustments to the statistical baseline by marketing and sales managers.

The company tracks the original baseline forecast, the changes to that made by managers, and then actual sales results. A “score” is developed that analyzes by individual whether those adjustments improved or detracted from the accuracy of the baseline versus actual results.

Not surprisingly, managers have become a lot more focused on the quality and fact-based support for their recommended adjustments to the baseline numbers.

Goodyear Drives to Near Total POS Visibility

Three years ago, Goodyear Tire & Rubber launched a demand-signal initiative to get complete, standardized point-of-sale data from retailers. It had a head start, since it has 600 company-owned stores and 1,200 Goodyear-affiliated independent dealerships. But it had to work out POS data feeds with hundreds of distributors and national retailers, including Sears and Discount Tire.

At the beginning of 2009, Goodyear was already acquiring POS demand data for about 70% of its retail sales resulting from the initiative. The goal was to reach 90% of total retail sales by the end of the year.

Goodyear uses the data to improve its forecasts and store replenishment, as well as to gain insight into the success of new product launches and the effectiveness of its advertising campaigns.

What Demand Planning Leaders are Doing

SCD

THE SUPPLY CHAIN DIGEST LETTERP.O. Box 714Springboro, OH 45066

PRSRT STDU.S. POSTAGE

P A I DSPRINGBORO, OH

PERMIT 714

• B2BMediaGroup provides the best in supply chain and logistics information. Our publications and informative Websites provide the latest news, insight, and opinions to help you keep informed, improve performance, and find the solutions and services you need. There is nothing else like them in the industry - see for yourself. Subscribe or visit today - it’s free.

• SupplyChainDigest:Weekly on-line newsletter and valuable Website. The very best in the industry, www.scdigest.com.

• TheSupplyChainTelevisionChannel:On-line, broadband video channel devoted to supply chain and logistics. News, case studies, thought leaders, education and training. Coming Soon!

• SCDigestOn-Target: Weekly e-magazine with insightful news and articles by key focus areas: supply chain trends, distribution and material handling, transportation management, global supply chain and logistics, RFID and AIDC, procurement and sourcing, manufacturing.

• TheSupplyChainDigestLetter: Hard copy newsletter published monthly focused on a specific topic, with concise, in-depth information and insight. It’s a great read! www.scdigest.com/letter

Editor: CONNIE VENEMA Publisher: DAN GILMORE

The SUPPLY CHAIN DIGEST LETTER is published monthly by B2B Media Group, PO Box 714, Springboro, OH 45066, Copy-right 2009. B2B Media Group LLC, All rights reserved. A one-year subscription includes 11 monthly issues plus email transmis-sions of news and updates. Subscribers may make occasional photocopies of articles within THE SUPPLY CHAIN DIGEST LETTER, but may not reproduce the publication in its entirety. Routine or systematic photocopying or distribution of the news-letter or articles from the newsletter is a copyright violation. To ensure compliance with all copyright regulations or to acquire a license for distribution or republishing, please contact B2B Media at 937-350-7915, or by email at [email protected]. Periodi-cal postage paid at SPRINGBORO, OH and additional mailing offices. Postmaster: Send address changes to: SUPPLY CHAIN DIGEST, Box 714, Springboro, Ohio 45066, 937-350-7915, email: [email protected].

SUBSCRIBE TODAY

This Month: Demand Planning


Letter


™Letter

the supply chain digest letter the supply chain...

Documents