Dealing with variety in supply chain
Henry Ford once remarked: “You can have any colour of car, as long as it is black”. The quote highlighted what Ford had to offer, as a choice set, to its customers in the early 20th century. Model T was produced in black colour only.
Today no marketer can dare make a statement like Henry Ford. They know that the customers of 21st century are already spoilt for choice. As more and more categories and products are moving away from being just a utility item to a differentiated, preferred brand, the marketers of competing brands are trying to beat each other, while trying to overwhelm the customer with choice.
Our work with fashion companies showed us that even the color black is not one shade! There are at least 10 different types of black shades. On the face of it, it looks crazy but once you place all the shades (of a product category like fabric) in front of customer, who prefers a black color; he will understand the nuances and select the one he likes and at same time will also be able to identify the shades, which are not his preference. This means that if he had not seen the 10 different shades of black, he would not have known his own preferences. Similarly there will be other customers preferring black color, who will zero down on a different version of black. This means that exposure to variety creates new niche need based segments in the market. But at the same time the demand for black, which would have aggregated under one shade, now gets dis-aggregated across 10 different shades. The explosion of variety is causing what can be seen as “Disaggregation of demand”. Such demand dis-aggregation is seen in almost every product category, bitten by the variety bug. With competition reacting with more variety, the consumers are now seeing an explosion variety in many product categories like garments, shoes, furniture, consumer durables, automobile, electronics, books, music, and even some industrial products.
Demand disaggregation is leading to formation of what Chris Anderson calls the “Long Tail” of demand. Chris Anderson, in his best seller, The Long Tail, identified the problem across product categories. It is no longer the rule of Pareto where 20% of SKUs are supposed to cover 80% of sales. Now 20% of SKUs cover just about 30%-40% sales and rest 60% of sales comes from addition of small volume of sale, each from many SKUs, thus creating the long tail of demand.
Desegregation of demand offers a huge challenge for both production as well as distribution. With large variety, increasing over the years, production facilities are being loaded with less and less volume per SKU over the years. At the same time with more variety, the demand for individual SKU also becomes erratic. The tussle between marketing and production is growing over the years, as production would want to meet their efficiency targets and prefer bigger batches per SKU but marketing would want less volume per SKU to meet the requirement of channel partners since the total volume sold by the company does not increase dramatically.
In distribution, the problem of erratic demand multiplies, as the disaggregated variety is further “pushed” to the downstream side to channel partners like distributers and retailers. With increasing variety, and regular push of SKUs to the downstream links, demand becomes even more erratic (multiplier effect of disaggregation of demand). This results in deterioration of inventory turns of the channel partners. With time, and increasing variety, the turns are bound to deteriorate. With reducing inventory turns, the purchasing power (OTB or Open To Buy budget) will deteriorate over a period of time. It has already started happening for some fashion manufacturing and distribution companies in India. The volume of business sold through their distribution channel is deteriorating over the time, while the companies are looking to create new channels to get the volume growth. The retail chains, particularly those selling products in fashion domain are bearing this brunt of reducing inventory turns and more markdowns.
Statistically speaking, the best place to hold and sell a wide variety is a central warehouse, which can serve not just as a warehouse location but also a selling location. This means that if we cannot stop the demand disaggregation due to variety, we can prevent the deterioration (the multiplier effect) by holding it at one place. If by a magic wand, we could get all customers of a country to visit the warehouse and buy from it, they would not only get the best variety but the inventory turns of the manufacturer will also be the best, due to aggregation benefits. Amazon has got the magic wand by using the Internet to get the demand on to its central warehouse. Amazon can hold a huge variety and enjoys inventory turns much larger than any physical bookstore, which has to move the variety to meet need of a local catchment area. No doubt, we see many physical bookstores struggling to keep the business alive. Amazon was able to do this because in case of books, DVDs and music CDs, the purchase decisions can happen in the virtual world.
However there are many product categories where buying decisions involves actual “look”, “feel” or “trials” with the physical product. Most customers would prefer to buy many products after experiencing it in terms of “look”, “feel” and “trials” like garments, consumer electronics etc. So producers of many such categories will have to depend on channel partners to reach out to the retail counters. So these industries have to send the variety to the downstream links of the supply chain. However, can we prevent the multiplier effect of demand disaggregation with this inventory movement? Is there a way to move the inventory downstream to the point of sale without suffering from all problems of deterioration of inventory turns?
The way to deal with variety at point of sales has to be a combination of a 3-pronged strategy depending on the type of product and industry.
1. Supply Chain Strategy: from push to pull
If one has to have variety and suffer from desegregated demand, the least one can do is stop taking actions which deteriorate the situation further. So if inventory has to move downstream – we should move as less as possible and hold more at the central location. This means that companies have to shift from a push mode to pull mode. The lead-time of supply has to come down and we should develop capabilities to replenish frequently and small quantities per SKU. This is the only pragmatic approach, as with increasing variety, most forecasting tool will fail miserably, trying to forecast desegregated and erratic demand patterns. So, instead of forecast and push inventory based on forecast. We should move inventory only based on consumption, while ensuring that the replenishment lead-time is shortened dramatically from current levels.
However with implementation of perfect pull systems, we do not expect the problem to disappear completely. At a retail point, even inventory of 1 unit of an SKU can turn out to be inventory of many months! This means that we need to not only have fast replenishment but also rapid replacement to take back what is not selling at the desired frequency. Taking back slow moving inventory at one store provides an opportunity to make it available at another store where it can be fast selling. The extra freight costs will be more than compensated with increased sales and reduced inventory.
2. Bounded Creativity – Differential Supply Chains
Now comes, the million dollar question, do we actually need so much variety to excite the customer. On one hand we know that increase in variety creates new niche segments, which gets created only after the customer is actually exposed to the variety. This definitely helps in beating competition and diverting demand to one’s products. But the pressure to add more and more variety also leads to situations where more variety does not create more niche segments or, even worse, the customer becomes indifferent to the variety (and for him any one of the choice set meets his need) This means the variety beyond a point does not create new niche segments but just desegregates the demand.
Today most designers are under pressure to show variety. Hence they are also churning out designs in masses, just to meet the numbers. So there are designs, which creates disaggregation without creating new niche customer segments. At the same time, it is also difficult to predict which designs will be a hit and ones, which ones will be the flops. Nobody really knows the point where increase in variety becomes irrelevant to customers. The best way out of the conflict is to take the intuition of designers and integrate design strategy with the supply chain strategy
Most fashion products have plants, which are “V” types (many end items are created from few common raw material and as we progress in the manufacturing flow, each step in manufacturing creates more variety using the WIP sent from the previous department), or “T” plants (where variety is created at the assembly point in manufacturing as a mix and match of common components).
The way to integrate designs with supply chain strategy is to limit the boundary of creativity of designers by limiting the pallet of designs from the divergent point of “V” plant or the common component pallet of the “T” plant. Awareness of designers of the supply chain advantages will help them take better decisions on how to use the ingredients for a final product. However, they are also allowed to create new designs from the start of the manufacturing process, fully aware that this adds to the lead-time of supply. For example, in case of garments, it can mean restricting designs from a yarn pallet or limited fabric types. Our experience shows that sheer awareness of supply chain impact on decisions helps designer makes effective choices.
The same strategy of differential lead times can be implemented with decision of different stocking points of SKUs in the supply chain hierarchy. Some items can be stored and replenished only at central warehouse, while others are available not just in central warehouse but also at downstream points. The rule of consumption based “pull” replenishment (instead of forecast based push) is implemented for any stock movement between any two points in the supply chain.
In case of books/ retailing, the differential lead time can be implemented with some items are stored at central warehouse and replenished to the shops, which the customer can pick up instantly, while for others, which are only at the central warehouse and sold directly through the Internet, the customer has to wait for a period.
The differential lead-time should help in aggregation of demand into “meaningful” variety (the one which creates new niche segments), which always enjoys the best service. The other additional variety is available but only with a higher lead-time of delivery.
This approach also requires an ability to move items between the two categories based on which items sell more and an ability to exit from products based on sale rates.
3. Managing Merchandize: Selling without the inventory
Decision to offer differing service levels to different products offers a challenge in selling particularly in cases, where product is stored ONLY at the aggregated location. The problem gets aggravated particularly in cases where product has to be experienced before the actual sale.
The buying experience usually consists of 3 components for decision-making:
Buying a consumer product involves one or more combination of the above 3 factors. In many products, the actual inventory of the SKU is used to provide the experience. If we are able to provide the above three experiences without having the specific inventory, then we can afford to keep the inventory at central location, while having effective props to provide the experience to the customer at the time of the purchase decision. Amazon is able to provide the experience of “sample” for a book by allowing people to read through the few pages of the book before they place an order on the internet.
Our work with furniture retail companies has shown that such props can be easily developed. For example, in case of furniture the retail point has one furniture type but the variety of different upholstery options is shown to customers by having fairly large fabric pieces, which can be placed on the sofa, which the user can look and as well feel. This prop is different from a catalogue because catalogue may not be able to provide the “real” experiences.
Not only do we have to have props but the retail store sales people also have to be trained to sell such items. They have to be very confident of the availability of such items at the central location. Remember that a retail sales man is always wary about selling items whose availability is not reliable. He is always the first point of contact for any immediate bad experience of the customer. The sales team should be confident of availability of such items; else it is impossible to expect the sales of “virtual ” variety at the retail point. The supply chain has to be tuned to ensure very high availability all the time regardless of SKU’s designated storage location
Once the salesmen are trained to sell “virtual” variety, we will get to know the true demand of such items. If they pick up volumes, we can decide to move the items physically to the retail point.
The variety war between marketing and supply chain managers is chronic in many supply chains. The way out of this conflict is not a unilateral decision to trim SKUs. The only way to trim SKUs is to allow it complete opportunity to stand the test of the market. The opportunity is provided to the end customer to experience the product at point of sale. The above approaches help provide the experience while preventing the problem of high inventories.
Who said one cannot have the cake and eat it too?
Size does matter!
Find out how to prevent sale loss due to stock out of ‘sizes’ without inflating inventory
Think small, reap big
Textile companies can solve their problems with huge inventory and significant markdowns
Manufacturing strategy for fashion garments
Textile manufacturing practices in India are primarily geared towards leveraging economies of scale and an increasingly non-existent " low cost advantage". This makes it an enormous challenge for large Indian garment manufacturing firms to take up higher margin but lower volume fashion garment orders. Find out how this industry can change gears.