In the second episode, Dr. Rakesh Singh hosts Lora Cecere, Founder of Supply Chain Insights, as they discuss the complex domain of supply chain and it is future. She is the most followed and popular Supply chain Guru on LinkedIn. She works with supply chain leaders to take teams to higher levels of excellence. Her research firm SupplyChainInsights is paving new directions in building thought-leading supply chain research.
Laura, why do some supply chains do better than others?
The companies that do the best have a clear definition of what is supply chain excellence. That sounds trivial, but it is not. This is actually something that I have been studying for the last few years. A lot of times people will say, they want to drive supply chain excellence. But when I ask them the question of what does excellence mean and how excellence drives strategy, they cannot answer the question. Unless there is clarity, the organization is not aligned. I find that the greater the gap between commercial teams, sales, marketing, corporate communications, and operations teams, distribution, customer service, procurement, and transportation the greater the gap in alignment; the tougher it is for those organizations to work together in disruption. Number two – is that traditional supply chain is inside out. It is not outside in. And by outside in I am talking about a signal that moves market to market, senses market changes, consumption, distribution, and does that with almost zero latency. The sensing time from consumption to the organization is small.
Now, the interesting thing is, only nine percent of companies actively design their supply chains.
The design of the supply chain is such that supply chain can operate market to market. Now, the interesting thing is, only nine percent of companies actively design their supply chains. I am a chemical engineer. I would not have gotten out of school if I had not been able to design distillation columns and heat exchangers. I am just shocked that so many companies are not actively designing their supply chains for things like inventory buffer strategies sourcing strategies postponement strategies outsourcing strategies. Instead, we have got used to planning for tax efficiency transactional efficiency. We are not really managing the supply chain as a complex non-linear system, where we design it. We have buffers, we have push-pull decoupling points, and we have a clear definition of strategy.
What do you mean by transactional? And when does the supply chain become strategic or start contributing to competitive advantage?
The supply chain becomes strategic when it is no longer a functional discussion. A functional discussion is about best in class transportation, manufacturing, procurement. When we are able to look at selling, supplying, manufacturing, and logistics together – how they tie to the business model and how they tie to a strategy, then it becomes more strategic. One of the issues is that most of our associations and a lot of our writing is about functional supply chain, procurement, and manufacturing. But this concept of source make and deliver working together to drive excellence in business models is something that takes it to a higher level and really looks at alignment.
When we are able to look at selling, supplying, manufacturing, and logistics together – how they tie to the business model and how they tie to a strategy, then it becomes more strategic.
One of the issues is the total cost approach. Often manufacturing or procurement or logistics functions will try to reduce functional costs and throw the supply chain out of balance. This will actually increase cost because you to look at the interaction of source, make, and deliver together to be able to manage total cost. Unfortunately, even well-intending companies will throw the supply chain out of balance.
What according to you would be end-to-end supply chain?
To me the best supply chains starts with the customer and the customer’s customer. And go all the way to the supplier’s supplier including the logistics and the second and third tier suppliers. To me end to end is bi-directional orchestration from market to suppliers and back. This means asking questions like – do I have a feasible plan, do I have the best cost options in terms of sourcing strategies, am I able to sense and translate demand with near latency to reduce the bullwhip effect etc. Many times when people talk to me about end to end, it is –order to cash or procure to pay – a very transactional narrow definition that does not drive value.
Many times when people talk to me about end to end, it is – order to cash or procure to pay – a very transactional narrow definition that does not drive value
If I am only focused on efficiency within the four walls of my supply chain I will throw the supply chain out of balance. A lot of transaction oriented technology companies like ERP and consultants have perpetuated this. This is not serving us well in the pandemic. If Icannotsense, cannot translate, and cannot orchestrate, then I probably will make a lot of mistakes. I will have empty shelves, I will send wrong signal to suppliers, and I will have the wrong inventory in my warehouse. One of the issues in the United States right now is, those inventories are high, and it is the wrong inventory. The doors are really blowing off the warehouses because we got the wrong stuff.
Let me go to your definition of supply chain and make it even simpler. Even technology companies look at factory to consumer, or factory to supplier, and they break the supply chain into different parts. This is the way people start looking at supply chain. You rightly said that during the pandemic this might not work. Let me put the larger question of pandemic and the geopolitical risk – do you think that the sourcing strategy in future needs to change – one from a perspective of a firm or a company and two from the perspective of a foreign company but in terms of geography.
Well you asked me a number of questions. Let me see if I can get into some of this. First off, if we are actively designing the supply chain, we do not generalize. We look at what makes the most sense across source, make, and deliver together, and how to minimize risk, how to design buffer strategies and how to account for the changing friction on borders. The global supply chains are based on frictionless borders. With the world now on a regional focused government, whether it is the UK, US, or China, we have more friction on borders. That is problematic for supply chains because things take longer to cross borders, paperwork is higher, and taxation is higher.
Only one-third of companies know where their second and third tier suppliers are.
I would account for that in the design of supply chains – which will be either global because materials are sourced from all regions of the world or regional. I think we will move to regional supply chains that have more control around buffer strategies and we will look at sourcing strategies more holistically. I will tell you a funny story I did an article for LinkedIn where I interviewed supply chain leaders and one of their surprises was that they felt that they had minimized risk and sourcing by diversifying suppliers. But they found that their suppliers used the same contract manufacturers and were in the same regions. They really had not diversified. Only one-third of companies know where their second and third tier suppliers are. I think we are going to get a lot more serious around where our suppliers are, who is depended on whom, and design the supply chain for that. But there are a lot of votaries of local or near shoring – these are geopolitical dynamics that may impact locational strategy for your supply basis. But what happens when you go nearby locations, what should guide the nearby investment by firms – whether they should vertically integrateor go to another country which is close by and diversify from own country so that they are secured during the pandemics or any natural disasters or any geopolitical threat. It is hard to generalize. It really gets down to what is the strategy, what is happening in the geopolitical world, and shifts in country leaders. The design of the supply chain needs to be looking holistically at what is happening. And we need to do a better job of managing what I am going to call planning master data. Planning master data allows us to look at how much friction there is on the border. It is looking at what are the real lead times what is the real time to get a container off a vessel, what is the delay at the port, what is the feasibility of transportation and what are we seeing in terms of container availability. We are not good at getting data and designing the supply chain for what is really happening, the feasibility of frictionless borders and the movement of logistics. We generalize, and generalizations are dangerous in supply chain.
Do we conduct stress tests in supply chains or some kind of stress test for the supply base, because it is very important for us to understand the time to recover and time to survival as Professor David Lipkesays. If the time to survival is less than the time to recover, businesses may need to look at local supply chains. They need to because such pandemics may not happen. But then, political tensions and friction on borders may disturb the supply base during important time of recovery. For example, India is facing a great threat from china. There is a border tension; there is also a lot of political rhetoric from both the sides. If you look at our main industry pharmaceutical, we see 70 percent of API’s from China. How does one really look at such situations where dependence is so high?
We cannot generalize. The design for source, make, and deliver for pharmaceuticals is very different from high tech or for apparel. And we need to not only look at the transactional inputs like labor and sourcing but also, we need to look at the feasibility of moving goods and receiving goods across borders. And that is one of the things we are not accounting for.
When we make the choice to cross borders, it should be for a good reason, it should be for unique sourcing, it should befor growing markets. It should not just be to chase the lowest transactional cost.
I was working with a toy company that designed a toy that had parts sourced from different countries. These countries were difficult to get goods in and out of. And they did not really look at the feasibility of movement of these goods or sourcing these goods. They were only looking at the lowest cost of manufacture. Because they were not looking at it holistically – of make, source, and deliver together – and the feasibility of movement of these goods, they had an infeasible plan. We have so many companies chasing an infeasible plan. When we make the choice to cross borders, it should be for a good reason, it should be for unique sourcing, it should before growing markets. It should not just be to chase the lowest transactional cost. Because we do not design the supply chain, and we do not simulate the shocks to the supply chain. Discrete event simulation allows me to test the feasibility of the plan. People make a lot of bad decisions.
Let me take you to the other side of the supply chain. Recently supply chain insights came out with a very good work on sales and operations planning. It is a great guide for us as a basis for audit for our own understanding and on our supply chain. What according to you are the common mistakes people make in terms of planning within the context of S&OP.
I am going to say five things. The first thing is – what is the role of the budget. Budgets are outdated the minute they are published. If I am going to anchor the S&OPplan to the budget, I will set S&OP into a death spiral that it will not recover from.S&OPshould be anchored in the market. It should be anchored on what is the potential to sell the product in the market, and when I do that, I sense what is happening in the market, I translate what is happening in the market, and I remove the functional barriers. Second thing is what the role of sales is. We started a policy of sales collaboration. The concept is that sales knows what the market is going to buy. It is not true. Sales is very politically driven by bonus and incentives. They will have negative or positive influence on bias based upon what those incentives are. Most organizations do not recognize this bias and they do not recognize the difference between the sales forecast and the market potential. A lot of times sales and marketing will have programs that meet their own incentives, but are misaligned with the market. As a result, the organization will go into dangerous spiral. Anchor the S&OP process in the market.
Third, make sure that your systems and your processes are delivering good outcomes. Many a times people will implement forecasting and they will not hold themselves accountable. Am I improving the forecast, am I improving the inventory signals? I recommend that companies type supply chains based upon coefficient of variation. And for every supply chain type, they measure forecast value add and they ask themselves the question – in demand planning and in the management of inventory am I improving the signal. Most companies are not. They have put in technology, they have got all these processes they have hired all these consultants and their forecast value add is actually negative toa low percent. They would do better without any of the technologies or any of the consultants because they are not holding themselves accountable. Is my work on planning giving me a better signal which is so essential?
What one number in this myriad of thousands of numbers do you want to anchor your forecast on? Not anybody that talks about one number forecast understands the basics of forecasting
Four measure improvement based upon a balanced scorecard. The functions should be aligned for reliability, and all functions should tie together for a balanced scorecard. If I measure functional metrics like purchase price variance or manufacturing or lowest transportation cost, I will throw the supply chain out of balance. Instead, in S&OP, I must align the organization with a balanced scorecard of total cost or operating margin, of growth of inventory terms not cash to cash. Because we have moved away from inventory management. We have chased cash to cash by elongating payables. Focus on inventory turns and focus on whether inventory is working for me and return on invested capital. If I have a balanced scorecard like that I can drive organizational alignment and get rid of a lot of the friction in the organization. If I focus and reward the functions on functional metrics, I will throw the supply chain out of balance and the plan will never be actualized.
The fifth point is to tie planning to execution. Our plan should be built into playbooks. I do not know whether the audience plays rugby, football, or soccer. But in our gaming systems, people will have playbooks so that when they go to the field they have got in their head the plays we are going to call. And when those plays happen, everybody knows what to do. The S&OPexecution should be similar to that. In the planning, you should build playbooks. Whether it is a new product launch outcome or a change of market or a changing competitor, you should change those playbooks based upon what is happening in the market. S&OPneeds to be tied to execution. People make the mistake like the belief in one number forecasting. How crazy is this? What one number in this myriad of thousands of numbers do you want to anchor your forecast on? Not anybody that talks about one number forecast understands the basics of forecasting. The second thing is I am going to tightly integrate the marketing forecasts, the sales forecast, and the supply chain forecast. How stupid is that? Each of those forecasting systems have a different data model, a different purpose. Instead what we have got to do is to synchronize and harmonize and then translate across that process. I cannot tightly integrate because the only thing in common with sales, operations, and financial forecasting is the word forecast. The third thing is, not designing the supply chain to look at inventory and buffer strategies, to be able to look at what should the supply chain be, and what is the potential. The fourth is not having a feasible plan underneath the meetings. There is a need to look at planned orders and if I have enough manufacturing capacity. As demand variability increases, I need more capacity. As border friction increases I need to change my logistics plan. So the plan needs to be feasible. The fifth is, people do not hold themselves accountable for action after review. They have a lot of S&OP processes but they do not ask themselves the question – am I improving what we are doing.
This is a critical misunderstanding amongst supply chain professionals. And you as you rightly said, sales, finance, and supply chain plans have different database and different context. Integrating them may not be the right way. But making them harmonious and synchronizing them together would bring a very strong insight into your own organizational supply chain which can define your competitive advantage. I would like you to go a little deeper into it because this is something very interesting in terms of interoperability and market to market delivery. If you could bring that particular context here it would help.
When I was a Gartner analyst, they started the research on what would integration look like? What is the CRM data model? What is the financial data model? What is the supply chain data model? I had a group of analysts and we worked two years to answer this question and at the end, we really came back with each of these data models is different serving a different group of users with a different business purpose. A financial data model is to really guide the financial organization for corporate reporting for the treasury. The supply chain forecast should be input into the financial forecast but never be tightly integrated. And the supply chain forecast should never be constrained for the financial forecast because if it is tightly constrained you’ll miss market opportunities because the budget is not really synchronized to market. Likewise, the sales forecast is very valuable for pipeline management for looking at how many sales people do I need, or perhaps how many trucks I need or planning for sales. But really can not be tightly integrated to supply chain forecast because the supply chain forecast needs to be accurate at a mix level which is the number of products that I have the number of product locations I have. And you really have got to look at mean absolute percent error to be able to look at error. And to be able to look at error based upon the number based upon coefficient of variation is a very different model than the sales pipeline forecast.
Let each of those organizations develop their own forecast for their purposes based upon the strategy. Over all of these forecasts, put an analytics layer
My belief after studying this for years is let those forecasts be. Let each of those organizations develop their own forecast for their purposes based upon the strategy. Overall of these forecasts, put an analytics layer. You can pick your analytics layer of choice – it could be descriptive analytics like tableau or Power BI or it could be one of the lightweight data models likeArkieva technology or Anaplan and allow users to query and look at the data models and to be able to bring the data together based upon their needs. Do not do this on schema. Do this as you need it because those forecasts are always changing. Hold the supply chain accountable for forecast value add, error, and bias, look at the differences, and drive communication. And any consultant that comes and talks to you about one number forecast or tightly integration of forecasts show them street because they just do not understand forecasting.
That sets up the context for the next question. You have been talking about an S&OP maturity model and you rightly said that large number of organizations are somewhere caught up in the middle. While reading and understanding integrated business planning one gets an idea that it is over the line process of bringing change in the organization. Where does it lie according to you in terms of S&OP maturity? Some people say it is advancedS&OPI read your report it says that itis one of the pillars – one of the five stages, and is the third stage of S&OP maturity. While there may be disagreement on that, can you tell us why you have classified that as the third stage and not the last stage where you bring sales finance and SCM together for reconciliation.
Well I have had many disagreements with OliverWight about this. You know I am not really that loved by OliverWight nor am I loved by the DDMRP group. But my research shows that if I am focused on the enterprise and making the enterprise more efficient, the enterprise will serve itself but not its customers. And companies that grow and companies that are profitable serve their customers. Market to market is from the customer’s customer to the supplier’s supplier with bi-directional orchestration.
Market to market is from the customer’s customer to the supplier’s supplier with bi-directional orchestration
IBP has the concept of making the enterprise more efficient more integrated and self-serving. If I as an organization am self-serving, I am not going to be resilient; I am not going to be able to understand market activities. Instead what I need is an architecture that allows me to take consumption data and move consumption data with almost no latency across the functions and align the functions to serve the organization outside in. Outside in processes are not well understood and not really developed enough.VMI was an example of an outside-in process. In my research, I interviewed a hundred companies. No one uses VMI as a sensing signal. They bring VMI to order management because there is no place to put it in enterprise architectures. We have more consumption data, more market data than we have ever had before. Yet nobody uses it. In the downturn, General Mills sat on more point of sale data in the United States than anybody else in the sales organization. They did not read the market -that people walked away from cereals during the recession and did not return. They built factories, and as a result, they had to lay off because they did not read the market. They had the potential with the point of sale data to have read the market and to drive better behavior. But they did not. They stayed tethered to the order signal and the order signal is not a good representation of demand. When we are talking about market to market, we start with consumption, we start with the consumer, and we start with the channel. We translate the channel with near latency across the organization and we get the right information to the suppliers market to market. It is for that reason that IBP is the third stage in the S&OP maturity. And people that are market to market are fourth and fifth stage.
There is another confusion. In your maturity stages, you say demand-driven is the fourth stage. Given the proliferation of DDMRP literature, is there any correlation between your fourth stage called demand-driven and DDMRP?
DDMRP is a translation of order patterns to material requirement planning. DDMRP is very tethered to the order. The order is not a good representation of demand
No. I think demand-driven is anything but DDMRP. I think DDMRP co-opted the term demand driven has really been dangerous in the market. DDMRP is a translation of order patterns to material requirement planning. DDMRP is very tethered to the order. The order is not a good representation of demand and DDMRP is only helping me with the translation of an order pattern to material buffers. I am not saying that that is not valuable. But it is not demand driven and it is away of translating an order pattern to materials. And we need to recognize it as that. Demand driven is looking at how do I get market signals into the organization and make them usable market to market. Or market driven is how do I get demand signals from the market through the organization to suppliers and logistics providers and orchestrate the signal to be able to work at price volume relationships. So for example, as suppliers change pricing – which is happening a lot in the pandemic – or sourcing is more constrained, or logistics is more constrained, how do I work this market to market that is the fifth stage?DDMRPdoes not even hit my grid and I think it is done some real damage in the market.
When we talk about market to market and about pandemic, three or four things have happened. One is the way the consumer behaviors have changed in terms of composition of basket of goods, frequency of buying, volumes of buying, channels of buying, and so on. And second, if you see the kind of outside that needs to be brought in, there is a challenge like the crisis of declining macro economy across the globe. We had a figure where it says that the pandemic damaged markets in terms of their GDP growth. How do you incorporate these macro variables as an outside in, into your demand planning, supply chain management, and market to market alignment of the supply chain?
Only four percent of companies are driving performance improvement faster in their peer group.
Let us start with the market and the consumption. I think what is happening is the phenomena of what I call the hammer and the dance. So as rates of COVID increase, people shut down. They go to their homes, and they change their behavior. They are not at restaurants, they are not at bars, they are not at sporting events, they are not at hotels, and they are not on airplanes. All of that affects demand. In the United States, I could not get my favorite toilet paper for weeks. That is a big company with great supply chain processes. They did not read the market. They really did not design the supply chain. The retailer became a bottleneck. The number of retail doors to get the shifts in demand to the market became a constraint. They accepted it and they tried to stuff more product through those retail doors and they could not get the product to market. They did not redesign. They did not go through the front door of the retailer. The big change is 40 percent of food in the United States go through restaurants. When people go to their home and they are no longer eating at restaurants, people are buying more flour, cocoa more baking goods. Shelves were bare. I have never seen a shelf bear for flour. All over the world, those shifts are happening with the hammer and the dance. What we have got to do is -we have got to sense locally what those shifts are and translate the demand to the supply chain with almost no latency. If I am not good at sensing, and I am not good at translation, then the supply chain will be on the back foot. If Ido not design for the increases in volume, you are going to be like this large CPG company that could not get toilet paper and towels to the retail shelf during the pandemic. This particular time requires us to think differently and requires us to use data in different ways. It is my hope that out of the pandemic we will build better, and we will challenge our systems and our thinking to be outside in. Let me just give you an example I teach classes where I ask companies and students to draw the supply chain. The first time I asked people to do this they, look at me like in disbelief. Only one student has started the drawing with the customer. 99 percent of students will draw a supplier or factory and that is their mental model of supply chain. It is about supply. It is not about the market and the customer, and about outside-in processes. We must change that to build better and to drive higher value for companies in the pandemic.
If we have to avoid technology built by consulting, what are your recommendations for people who are looking at technology solutions for aligning supply chain post pandemic?
I would suggest that they take an inventory of the technologies they have. Many times people will have technologies they do not even know about. First, ask themselves the question of how can they use technologies better. For example a lot of companies will have network design tools and have not really used them, do not really understand how to use them, do not really understand the basics of form and function of inventory do not know how to drive a simulation. Look for help to use the tools you have is number one. Number two is spend some time thinking about how you get market data. Get in a big room and draw all the outages that you’ve had you know, all the shorts you know. Look at your order short data and ask yourself what data if you had could have helped you to minimize that short. Where is the data, what kind of data? Then look at the tools you have for analytics and ask yourself – how do I use my current analytics tools to sense market data and to drive better alignment in the organization and the market. That is the second thing. A lot of times, companies have data on marketing or sales or R&D that they do not know they have. And they are not using it and they are not using their current tools. The third thing is do not put an RFP into the market for consultants or a technology. First, use what we have got. MostRFPs send us into dismal spiral where we are having lots of discussions that lead to very expensive projects that do not drive values. Let us use what we have and let us measure how effectively we are using our technologies with forecast value add coefficient of variation, the ability for sensing what is the demand latency from your channel how do you decrease that. Ground yourself in measurement and align the organization to the strategy that is my thought.
What kind of capabilities do you think are important to build an organization and supply chain professionals to be relevant for future?
I am currently working on a report on talent. I think talent is the missing link in the supply chain. I was talking to a consultant yesterday, who had worked for very large consulting companies, and was teaching now at a university in the United States – Cornell. He was working on developing a program for supply chain leaders.I said to him the three most important things that you need to teach young students is – let us be honest with ourselves –supply chain management as a practice of source, make, and deliver started in 1982. We are on the fourth generation of supply chain professionals. We think we have bestpractices. I do not think so. Only four percent of companies are driving performance improvement faster in their peer group. I think we have historic practices and that we are not really holding ourselves accountable for balance sheet results. That is number one. The second thing is teach supply chain leaders how to drive businessvalue – how to talk the language of the balance sheet, how to drive influence management – most supply chain leaders talk acronyms that nobody would understand. Let us teach them influence management. Let us teach them the ability to listen. Let us teach them the ability to problemsolve and to speak the language of business. The third is, I find that a lot of students are taken into the world of supply chain and they get a little naïve. They do not understand that organizations are political. They do not understand that sales and marketing are not aligned, or theydo not understand that people in transportation do not necessarily talk to people in sourcing. What I would recommend is that people become change agents to drive alignment. That they bring people together, that they drive facilitated workshops on how groups can worktogether to drive change. The fourthis to question the status quo but to do it with respect and humility and to train yourself on the art of the possible, what are these new techniques, how do I use unstructured data, what is cognitive computing, what is a rules-based ontology, how do I change what is the art of the possible. Educate yourself. But then influence the organization in ways that are meaningful. You know youdo not have the answers. You only have the spark. Widen your sphere of influence so that you can be a change agent in the right way and you can enlist a guiding coalition against a problem that your organization has driving the art of the possible. Fifth, and important, is get to know your customer. Get to know your customer in meaningful ways. Stay grounded in how you serve the customer because most organizations do not. I am going to just close with an example. I recently moved and I bought a new washer. The washer got dropped in transit. Does not work. New washer is expensive.I called the manufacturer no response. Called the distributor, no response. The new washer is going to come in three weeks. I am not happy.No one in that organization cares. Am I going to buy that brand again? No. Butmy buying cycle for that brand will bea long time.But I am talking to my friends and my rating and reviews of that particular product will not be good. Make your organization easy to do business with. Care about your customer deliver against the promise of your customer and do not lose sight of that. ♦♦♦♦♦