Network Design: the 5 good questions to ask

A lot of our clients reflect upon the relevance of their network design and wish to challenge their flow organization in order to address the stakes of reactivity and efficiency imposed by the market. This reflection is even more necessary that most retailers (more than 80% according to recent polls) and an increasing number of industrials tackle the omnichannel battle. Being closest to the consumer, providing quick and customized service at the lowest cost… many challenges that put high pressure on logistics networks and force players to rethink their organization. How many warehouses and where to place them? What storage strategy to associate? What delivery method to choose?

Oftentimes, the evolutions of logistics networks are done under constraint, for instance to compensate deficits of storage surface. In France, more than 8 million square meters of additional storage surface have been put out on the market in 2018 by 40 logistics players. This approach is pushed by a global trend of regionalization of just-in-time production.

Yet, to be efficient, the flow master plan cannot be apprehended opportunistically, link by link, but rather in a holistic way. The network design is part of a mid/long term logic and aims at supporting the company’s business plan implementation. This requires the top management’s implication in order to have a transversal reading of the challenges and opportunities faced by the company. The evolution of supply and demand will have to be integrated upstream in the different scenario modeling, implying a reflection in the break.

In the most mature organizations, the network design will even be used to draw change management in other corporate functions because it moves the set of corporate functions.

benchmark 2018In a recent benchmark among players of the retail sector (brand with a network of at least 100 stores in France), the declared costs of their Supply Chain fluctuate between 8 and 15%. These cost gaps are partly due to specific services offered by the different brands. For instance, the fact of offering to customers and/or stores, a per unit preparation requiring an unpacking operation, increases structurally the picking costs by 2 to 3%. Without mentioning the costs implied by a customer promise to D+1, allowing a craftsman to place his/her orders at 5pm in order to be delivered the day after on his site before 7am…This creates additional costs that need to be precisely estimated on the field to let the decision maker pick the right choices in terms of service offerings.

Mind the popular misconception that outsourcing logistics is THE solution to optimize the couple cost/service. Indeed, it will increase the business expertise and bring flexibility but it can also be disappointing. The markup operated by providers represents on average between 5 to 15% of increase compared to an internal production. Furthermore, if the handled products present strong specificities for instance in terms of congestion or fragility, it is more than likely that your current logistics team have an expertise that will be very difficult or expensive to find amongst the market players.


Once you’ve had this operational understanding of your costs, you will have all elements in hand to define the right level of service that you want to offer. For instance, the linear improvement of delivery times can be very expensive and sometimes completely unnecessary compared to your customers’ or prospects’ real expectations.

In addition to an interpretation per product type, the reflection should also relate to the customer segmentation according to their attractiveness. This attractiveness can be assessed according to the margin or market growth level or to the customer’s turnover potential increase.

Example of Matrix

Product types-costing

The definition of customer promise requires a very good knowledge of the market. For example, we have been recently asked by a cosmetics subcontractor to reduce its lead time from 14 to 10 weeks on the market. Collaborative workshops allowed us to find out that the current lead time was perfectly acceptable by the contractors. However, the latter is completely uncompromising on the respect of this commitment. We have reworked the flow organization in order to secure this lead time, without applying additional constraints on the flow scheme.

In the various modelized scenarios, a robust sensitivity analysis must be realized in order to check the impact of hypotheses on the cost, quality and lead time results. A wrong anticipation of risks leads to important additional costs, whether they are indirect through dispute processing in 60% of cases or direct with the application of penalties in more than half cases. They can even result in customer loss in 30% of cases. Some risks are relatively well apprehended by industrials, like supplier risk which appears now almost systematically in the model’s evaluation grids, or country risks which lead to a world known and shared ranking, created by FM Global, an American damage insurance company.

However, other parameters are much more difficult to model because they’re often unknown.

In the first place, there is the regulatory risk. For example, the customs charge evolutions represent a confusing factor for Supply Chains in our global economic paradigm. Closer and more concrete, the taxation threat which constrains refrigerated freight is like a sword of Damocles above logistics flow schemes with per kilometer costs which could triple. The multicriteria analyses and the workshops conducted within the network design study allow to choose the scenario that will optimize benefits while limiting risks. The most resilient organizations use “what if” scenarios in order to build a continuous activity plan and secure quick backup plans’ implementation.

The transition plan generally falls within a period of 2 to 3 years to let all players of the value chain adjust their job to this new scheme. It is essential that the identified projects involve the entire organization, as the project’s success isn’t only the Supply Chain’s responsibility. The crucial role of IS in these projects is obvious and requires the right level of technical competencies (internal or external). One of the key success factors through the network design evolution lies on the good evaluation of necessary resources and an agile driving of potential threatening bottlenecks. The savings and benefits planning will systematically be linked to a resource plan in order to have a realistic understanding of P&L impacts on a short and mid-term.

The entire organization and its partners must be aligned on the target flow scheme which will let either the organization follow new dynamics, or accelerate its decline, in case of failure. Priorities and projects phasing are an integral part of the network design. All industrials agree on the fact that one of the key success factors of such a large project is to make all the impacted employees share responsibility, be it operational people in Supply Chain, marketing, procurement or finance positions. All must be convinced by the approach in order to support the project both internally and among clients. Indeed, with the digitization of the purchasing act, there is no doubt that the network design, is more than ever, a differentiating strategic element for the company.


In a nutshell…

All of these questions are necessary if one wants to define an optimal flow scheme. Numerous network design solutions exist to support this strategic reflection but beware of the shortcut that implies that the tool will answer the questions. Its utilization is, on one hand, often limited by the level and the size distribution of the available data and will not be able to replace collaborative workshops among jobs. Once defined, the network design should then be regularly challenged, as many internal and external parameters influence this equilibrium.

Supply Chain in retail: prioritizing the shelf

For fifteen years, information systems have taken a leading role in the management of flows: ERP communication and collaborative approaches between manufacturers and retailers (SSM and now GPFR) have become the industry standard. Even better, stocks are now controlled on the basis of actual consumption (VMI), completely relieving the sales areas of responsibility of supplies. Yet despite all these developments and associated investments, the Out Of Stock (OOS) rate recorded on shelf stock barely drop below 8%. Are we facing a structural OOS rate? Can we go further to improve the product availability? Traditional retailers are asking the question more than ever at a time when, to deal with increased competition from online distribution, the answers are to provide a wide range of products with immediate availability.

In 2011, a study by the ECR concluded that the average rate of actual product availability in large retailers was less than 90%. This figure surprised many France no exception. Studies conducted in the United States have come to similar conclusions, averaging 8% OOS on shelf (Corsten and Gruen 2003). Even more astonishing, this figure varies very little when you factor is investments for tools and technology. In 2008, a report commissioned by the Coca-Cola Company reported an OOS of 8 to 9% … the same rate that had been recorded 12 years earlier.

Remobilizing Store Teams on Supply Chain Issues

We are very far from the service rate known in the industry, in the warehouses of the same retailers that show an average 99% of product availability. How do we explain and reduce this gap?

The first challenge is to align Supply Chain and Sales with this finding. Indeed, when we ask store managers about their product availability level, they typically announce figures that are far removed from the reality on the shelves. The OOS rate is usually underestimated, and even more the loss of associated revenue because store managers are assuming sales will almost systematically be captured by other products. A survey conducted by Danone showed that in nearly half of cases, the client preferred to do without the product or buy it via a competing brand. In defense of store managers, few Supply Chain reporting guidelines incorporate this concept of “On-Shelf Availability”. Indicators often only cover the platform availability and delivery lead time in store. But what about the actual customer service? If one agrees that the role of Supply Chain is to deliver the right product, at the right time, in the right place, the performance indicators should focus on the one place where the goods are really needed — the shelf. Yet OSA (“On-Shelf Availability”) is an indicator rarely included in the Supply Chain vocabulary for retail.

Failing to be measured by the Central Supply Chain, OOS products in stores are often relegated to (anecdotal) field surveys performed by the sales teams. This process, which has the merit to exist, however, has two major limitations:


It is not possible to calculate the real OOS rate of the store, for the following reasons:

  • These relatively time-consuming surveys cannot be conducted on 100% of the store’s inventory.
  • Made at a predefined frequency, it gives only a static view of OSS but does not allow you to understand the duration (or circumstances) between each review.
  • Finally, usually performed in the morning, at off-peak time rush and often just after replenishment, they cannot be considered reliable. Client traffic usually follows a “camel hump” curve (see illustration below), with increased traffic in the late afternoon. Thus, OOS products in the afternoon potentially represent a higher loss of turnover figures for distributors than those identified (and corrected) during the morning.

The second major limitation we see is that these OOS surveys are undermanaged and underutilized.

During this operation, the store employee will strive to optimize the presentation of the shelves, and potentially to verify the presence of the product in the storage area. With our clients, we work with operational teams on the development of a simple diagnosis & resolution tool to help them utilize the information system to localize the goods. From there, we can easily understand and deal with the root cause of OOS products. This analyses often concludes that the central supply-chain is responsible for many of the failures.

Build a Field Action Plan

Empirically, it is found that 50 to 60% of OOS on-shelf depend directly on store operations, the remaining portion being connected to the upstream supply chain (suppliers / platform failures).

Among the errors attributable to the stores, listed are those which occur most frequently and which must be better managed:

  • The theoretical stock is wrong, thereby preventing replenishment triggers
  • The quantity on shelf is not properly aligned / set-up (facing and depth)
  • And finally, a well-worn subject but that still represents a potential improvement for retailers, the product is in the storage area but it did not make it to the shelf.

This year we worked with a chain of hypermarkets where stores complained of a malfunction in their automatic ordering system. The conclusions of the on the ground analysis confirmed this rule (see below the summary of the causes of observed failures):

It’s a fact: products often travel thousands of kilometers without any problem, to then get lost in the last meters in-store. The goods and information flows are under control until they enter the storage area – which is where store organization needs to take over.

  • What Operational Solutions can be Undertook to Avoid these Problems in Stores?

Some retailers, with the support of manufacturers, develop tools to encompass the shelf in their global vision of supply flows. RFID is the most acclaimed. A survey conducted in October 2015 with professional retailers confirmed that 63% of interviewed companies were investigating RFID as a solution or already deploying it as a solution. However today, the solution is deployed in only 6% of organizations … So from here to have an RFID chip on every product, what we do?

It would be unrealistic to base the improvement of customer service only on the technology. For example, electronic tags that incorporate logistical information (available stock, work in progress…) have not had a significant impact on product availability. While access to information is faster, it’s utilization is still necessary in order to know how to use it more effectively.

It is therefore necessary to work with operational teams to develop an efficient information processing system.

Two main areas need to be analyzed:

  • Review the Process
  • Strengthen the Managerial Control

Goods receiving, shelves replenishment and more generally operations planning in stores

The storage area often resembles a ‘black box’, which generates either overstock or out of stock products. It is essential to redesign how this area is organized into a transit area instead of a storage area. The reconfiguration of the storage area is often a prerequisite to optimize the replenishment process and the activity management. Whether the store has dedicated replenishment teams or mobilize their sales staff during off-peak hours, the challenge remains the same — ensure that the information and goods efficiently circulate between the storage and the sales area.

Change Competencies in the Retail Sector

In terms of optimizing the supply chain, we can only welcome the development of technologies that allow a transversal and centralized management of supply flows. However, these tools should not limit the responsibilities of the section manager but instead, allow him to position himself on higher value added activities. The release time on order placement (what, how, when?) must be reallocated to stock calibration and the global flow management during the last meters of flow.

Any technological change, as it enters the supply chain of retail companies, must be accompanied by a change of roles & competencies. In the retail sector, the explosion of the offer (range width multiplied by an average of 2 of the last 15 years) and the pressure on costs create many challenges for an organization. The shelf is not only a place to showcase products for sales, but is also a storage space. The roles and competencies of sales teams must evolve accordingly and retailers must adapt their organization to this transformation.

5 Best Practices to Better Control Your Distribution Costs

What if your margins were incorrect…?

Although you know it, you are not able to measure it precisely: all your customers are not profitable! Depending on the market, we can estimate that up to 20% of your customers lose you money. This is not so much characterized by what they are buying, but as through the way the product or the service is delivered. We are going to share factual cases encountered in the distribution sector and the FMCG industry. Those cases demonstrate how they had to adapt their approach regarding Supply Chain cost in order to make the right decision to drive their margins.

Piloter vous coûts de distribution

Mieux appréhender vos coûts opérationnels

# 1: Better understand your operational costs

This need is higher as the more “customer oriented” approaches developed in the past years have brought huge complexity to the operations of manufacturers and distributors. More and more constraints weigh on the Supply Chain, first and foremost the stake of ultra-reactivity.

For example, companies in the business of food-processing of dried products have seen the frequency of delivery multiplied by 4 in the 10 past years, as the distributors tend to maintain the lowest inventory level. Obviously orders splitting brings higher costs such as freight cost, but they can also generate a cost increase in order preparation and management. Those additional costs are rarely accurately calculated and therefore are not distributed to all the partners of the value chain.

However, it is important to calculate accurately the operational costs; that is, to identify the specific costs linked to a customer or a product range, with the aim at driving their profitability… Most of the time, the consolidated P&L approach hides critical operational facts.

We recommend a 2-step approach:

Step 1

Prepare a physical and information flows mapping as a VSM (Value Stream Mapping) to get the full picture of the operations, from customer needs to customer delivery, starting with the general framework and adding the specificities.

Step 2

Analyze and observe in detail each flow in order to characterize and quantify the distribution operations of each single customer. During on-floor measurements, it is pretty common to observe distribution costs ranging from 1 to 3, depending on customer requirements… Distorting significantly the margin calculated for a product range.

Two main types of cost could cause these profitability gaps:

  • Operations: packaging, check, picking, specific freight, mark down rate…
  • Organization and management process related to the customer: forecasts, after-sale services, range review and rebates…

The question then becomes to reallocate the operational and administrative costs in line with the enforcement of your organization around a customer or another.

Reallocation of Supply Chain Costs over Product Range

New Product Badge Blue

# 2: Challenge your referencing

Are you sure that your Marketing teams (Category Management, Market Manager,…) know the real distribution costs related to your customers?

It is more than likely that the step of cost mapping (BP #1) will bring a fresh eye to your analysis of product range strategy. What could you do for the clients with a lack of profitability? The first obvious solution, to increase the selling price, is very rarely sustainable from a commercial perspective.

Therefore all you can do is optimize your Supply Chain costs, consolidating the logistic operations on a reduced number of products. This implies proceeding with a comprehensive analysis of the products mix for each customer and to ensure that their products selection includes the minimum of low profitability references.

We recommend 2 ways to initiate this process:


Range bottom approach

to study the margin for a product and to compare it with the recalculated Supply Chain costs in order to validate whether the product should remain in your offer. This pragmatic and robust calculation enables the internal alignment of the teams and to objectify the range review.

This analysis, led by an FMCG leader, highlighted that 8% of the products from the range bottom have a higher distribution cost than the margin realized for the 12 past months.


Range top approach

to identify the penetration rate of your 20/80 with your class A, B and C clients.

This analysis underlines the customers with exclusive references or even exclusively developed ones, while others reference of the Pareto could meet their requirements. Final objective: develop the consolidation of the product range.

This can of course be triggered by a specific customer request with the purpose of setting yourself apart from competitors, but in fact this is far from being systematic. In the example hereunder of one of our client in the FMCG industry, we were able to suggest an alternative meeting their needs, for 20% of the references. Some specific or mono-customer products have been replaced with references of the 20/80 and thus removed from the product catalogue.

Penetration Rate of the Products Pareto

#3: Share the information

A customer cannot figure out the impact of its decisions on your activities, and the related loss of productivity… You must bring up this question.

The best “reflexes” to have:

Partager l'information

Share with your customers the operations linked to their specific requests.

Give them information about your costs, in particular the costs having threshold in opposition to linear ones. Keep in mind that your customer cannot figure out the threshold effect. You can work together on min/max parameters that your customers can easily understand and hence accept.


Define internal rules.

For example, are you certain to guarantee a service rate “whatever it costs”? What is the marginal logistic costs of last-minute orders? Re-prioritization, freight offset, reconciliation meeting, we often observe that the process of definition of the ATP (Available To Promise) is not clearly defined…or, if it is formalized, it is not be followed in daily operations.


Finally, make decisions based on factual figures!

The principal of differentiated Supply Chain too often consists in providing the best services to the largest customers. What if you shift the cursor from the size of the customers to their profitability? It makes sense to deliver a higher level of logistic services to your most profitable clients, rather than your larger clients for which you have a limited leverage.

#4: Harmonize and standardize your logistics operations

Do you control the variability of your operations and of the fluctuation of your productivity?

This question can quickly become a complicated one if you consider the processes of picking and order preparation with a large diversity of clients. The main task is to develop a guidelines framework and to harmonize the practices as per the standards. It means to apply Lean Manufacturing methods to your logistics operations. The elimination of the MUDAS enables the achievement of significant productivity gains in the logistic centers as well as a quick achievement of the first improvements.

In non-automatized warehouses where the picking is manual, we observe, by experience, that on average 30% of tasks performed by the operators and the foremen are pure wastes. The implementation of an improvement plan enables waste reduction of 10 to 15%, implying as much for you on your operational margin.

Logistcs Operation Harmonization and Standardization

Three main levers can be used to implement such an improvement plan:


Tools & processes

implementation of standards defined with the teams, revision of the overall logistics plan and its associated equipment.


The organization

review of the daily timeline, de-partitioning of the process and working expertise…


Management systems

performance management, objectives cascading, feedback on previous experiences and improvement loop…

#5: Change the negotiation paradigm

It is still very rare to find customer-supplier Supply Chains talking directly together. Most of the time, negotiation goes through the Commercial and Purchasing teams, who do not have the full picture needed to create a transversal Supply Chain.

Fortunately, collaborative initiatives between manufacturers and distributors are more and more frequent. Improvement projects can be summed up in 5 majors themes: forecasts, in-shelf availability, data exchange, supply management, logistics pooling. All those working groups, focusing on customer service, forget to discuss about the global and transversal costs of the supply chain.Source: ECR (Effective Consumer Response).

We regularly support our clients, manufacturers and distributors, in their negotiation process and we often observe that the notion of transversal Supply Chain remains an abstract concept, as each party battles to optimize its own logistic costs against sometimes their global performance. This implies to develop a strategic alliance model and to modify the traditional approach of the buyer-supplier Supply Chain.

Negotation Approaches

The most common optimization axes of customer-supplier logistics:


Incoterms optimization:

FOB, enabling work on batch optimization and to get economies of scale; EXW, giving the option to the distributor to optimize the cost – if he has good conditions – without the consideration of MOQ (Minimum Order Quantity)


Revision of the supply chains’ organization

stocked flow, cross-dock, short / local distribution network,…

Beyond these two axes, a complete understanding of the Supply Chain costs “beyond the wall” enables to open many other improvement paths. This requires to get the expert persons together to redefine the adequate needs regarding the impact on the distribution costs.

Impact of Supply Chain Organisation on Manufacturer Supply Costs

Example of supplier costs impacted by the different types of supply chain organization

In a difficult economic context, where the negotiation between the manufacturer and distributors are more and more strained (as shown in the situation of February 29th of this year, when 20% of the agreement between suppliers and retailers were not signed off on), it is urgent to change the paradigm and to open the horizon of negotiation. It is clearly in the collaborative Supply Chain approaches that the source of cost optimization lays down.


In this quest of improving customer service, it is essential to survive, to drive and control the distribution costs.

Cost modeling is a compulsory tool to make the right decision in an agile way. The data management, which becomes more and more accurate, enables us to get the logistics costs at the level “offer/client”.

Our clients opt for this approach, Cost-to-Serve, with 3 main impacts for their companies:

  • Understand the real profits made by each product range and for each customer
  • Identify opportunities to optimize costs for their transversal Supply chain
  • Simulate the impacts of potential evolutions of their business models and adjust their Supply Chain accordingly