Large fashion retail companies have been driving their sales growth in recent years by expanding their store network. Analysis of data released by the largest publicly traded retailers in the country reveals great ambitions in terms of expanding the number of stores over the next five years. Apart from its objectives of increasing same-store sales (Same Store Sales), network expansion seems to be imperative to enable long-term growth in fashion retail. In large urban centers, it is necessary to increase the presence with the consumer, following the dynamics of evolution of cities. Income and consumption in the country are deconcentrated towards the North, Northeast and Midwest, and move inland, creating new opportunities for store positioning.
Figure 1 – Projected growth in number of stores – fashion retail
Source: IR websites of each company
Regional, cultural and climate differences, difficulties in accessing and transporting to regions farther from the big cities, social mobility and the growth of the new middle class, bringing a new contingent of consumers with tastes and desires that have not yet been mapped, all of this increases the unpredictability of the demand and puts great pressure on the planning, supply and supply areas of companies.
To ensure economically healthy growth, maintaining or expanding margins, companies must be aware of two main indicators: the amount and time that “holes” remain in the range of products at points of sale, lack of sizes, colors or models ( Break); or the need for price reductions or sales to rotate the stock of overstocked products (Markdown).
While the first impacts the potential result due to non-sales, due to the loss of opportunity (and the margin) of the product that could be sold, the second erodes the planned margin for products that are in stock, reducing the effective result. Given the complexity of the product lines, styles and brands of the typical assortment of a large fashion retail chain, it is common for companies to present stockouts and markdowns at the same time in their network, and even in each store.
Even considering that part of the difficulty in carrying out the supply lies in the uncertainties regarding the acceptance or identification of consumers with style elements present in the products, and that the greater the fashion content in the products, the greater the risk of non-identification, supply planning it can be a decisive factor in mitigating or expanding the impact of disruption or markdown.
This article aims to address product supply strategies, relevant factors for defining these strategies, and show how the alignment of different strategies with specific characteristics of product clusters can bring benefits to fashion retail.
SUPPLY CHAIN FOR FASHION RETAIL
The textile and clothing sector in Brazil has gone through different phases over the last 30 years: from a protected market until the early 1990s, when a process of deregulation, trade opening and productivity gains in Brazilian ports began, to the open to international competition. After opening, factors such as the competitiveness of major Asian producers (China, South Korea, India and Taiwan) and exchange rate fluctuations began to impact the segment's trade balance.
With relevant competitiveness in personnel costs, in an industry still intensive in labor, the migration of a good part of textile and clothing production to Asia was not only a Brazilian movement, but a global one. The impact of this was an increase in the time from design to the store, the necessary increase in batches for economies of scale in transportation, and the growing impact of errors in sales forecasting for large fashion retail chains. Typical lead-times for this operation ranged from eight to 14 months between the conception and design of the collection, making and approval of prototypes, purchase of fabrics, production, quality inspection and transport from the producer to the retail distribution center, where the products reach the stores.
Figure 2 – Example of a collection lifecycle
In the mid-1990s, the Spanish Zara (Inditex), one of the largest fashion retail companies in the world, innovated its supply chain by opting for local production, vertical integration and a sales mapping system in stores that allowed correcting the mix or enter new collections in approximately two weeks. Named “Fast Fashion”, the model had a great impact on the sector, leading competition to efforts to shorten the lead-time for introducing new collections.
Typically, fashion products are supplied in waves (collections) that have thematic unity. As collections succeed each other, each product has a limited exposure time. The entire collection must arrive in stores in a synchronized way, as there is a harmonization between the various pieces, which refers to an idealization of style, in turn linked to aesthetic characteristics, climatic seasons, etc. Thus, synchrony in the supply chain is crucial for the success of the collection, and the greater the fashion content, the greater the degree of obsolescence of the product.
On the other hand, the existing fashion content in the products helps to reinforce the differentiation and exclusivity not only of the items, but also of the retail chains themselves. On a scale that goes from the basics (a classic white knitted shirt, for example) to “fashion” (a product with a specific style, which seeks differentiation), the pieces have very different demand behaviors and characteristics.
In his article in the Harvard Business Review, Fisher (1997) observes that the most successful companies are those that best adapt their supply chain to the characteristics of their products. The author proposes two main categories of products – Functional and Innovative – according to the predictability of their demand behavior. From the correct characterization of these product categories, the adequate structure of the supply chain and a specific set of supply strategies for each company could be inferred, which can be summarized as follows: companies that operate in the market through functional products should adopt process and cost efficiency strategies, while innovative categories would require quick market response strategies.
Figure 3 – Characteristics of product clusters
When analyzing a fashion retail chain, we can find items that present characteristics of both Functional and Innovative products. In fact, if it were possible to compose a scale that at one end had a typically Functional product and at the other a typically Innovative one, we could position the store assortment along the entire range of that scale. However, a set of factors has led companies to erroneously adopt a single strategy for supplying their products. Wanke (2001) registers the pressure for low inventories, due to the obsolescence cost of maintaining high inventories, and, on the other hand, the pressure for increasing the availability of an increasingly diversified mix of products to enable mass customization , as the main factors for the design of the companies' supply chain.
Fisher (1997) explains that it is common and expected that company managers seek an efficient and lean supply chain. Even when some product categories cross the boundary from Functional to Innovative, demanding differentiated strategies, few adapt their strategies to the needs of these products.
In the specific case of fashion retail, the supply chain is structured to provide a perfect service experience, in which the availability and variety of products are part of the communication of the design and fashion component of the collection, that is, a relevant part of this experience. service. For example, a Style manager determines not only the characteristics of the piece itself, but the way in which the different pieces of a collection (pants, skirts, blouses, accessories) are articulated to communicate the style and reach a certain customer profile.
Thus, as the pieces in the collection “go together”, there is communication from the collection on how to dress, and this increases the potential for purchases per customer (if you like the pants, you can buy the blouse and accessory that match). To ensure availability, variety and supply chain synchronization, companies choose to push collections to stores as quickly as possible, eliminating intermediate inventories and managing demand through Markdown and Stockout.
However, the impact of poorly planned sourcing of a product line can be significant in terms of a collection's bottom line. For Fisher, the effect of the inadequacy of supply strategies to the characteristics of the products will be the increase in logistical costs (larger inventories, unnecessary movements and Markdown), coexisting with a low level of service (Breakout).
PRODUCT-STRATEGY MATRIX
Wanke (2001 – 2002) proposes a typology for classifying product flows in three main dimensions: coordination, allocation in space and basis for activation in time. From them, it would be possible to derive the main decisions related to the flow of products: push/pull, centralize/decentralize the stock and produce to stock or to order.
Product flow coordination would be linked to the degree of visibility of real demand within the supply chain and the resupply cycle time. A pulled product flow would be a consequence of the high degree of demand visibility or a short resupply cycle, implying the feasibility of triggering resupply based on information closer to actual demand. On the other hand, low demand visibility or a long resupply cycle would imply a pushed flow of products, based on product sales projections as a triggering factor.
In the case of fashion retail, real demand information is widely available to the company, as the sale to the final consumer takes place in the chain of owned stores. However, as seen previously, the product resupply cycle has very long deadlines, implying a decision for a pushed product flow. The exception would be the Zara model which, through quick identification of fashion trends, visibility of real sales and vertical integration in the supply chain, allows for a faster response and a flow of products.
Likewise, a flow of pushed products would be linked to items whose demand behavior has a greater degree of unpredictability, such as, for example, products with greater bets on differentiated style content. This would lead to the adoption of more qualitative sales forecasting methods. Otherwise, more basic products would benefit from a stream of more robust statistical forecasting techniques.
Space allocation decisions are related to the positioning of stock in the supply chain, centralization or decentralization of stocks. Stock decentralization corresponds to an anticipation of the decision to allocate that stock in space, while centralization would be the postponement of this decision, keeping the stock in a limited number of points in the chain. In a simplified way, products with a high cost or value and a high degree of obsolescence (given by the duration of their life cycle) or degree of perishability (given by the expiration date), where the risk of an early decision to allocate stock to stores higher, would imply a decision to centralize inventories. However, with the decision for the flow of products pushed to stores by fashion retailers, the decision to position stocks has been taken, leaving instruments such as Markdown to resolve any variations in demand.
The basis for triggering in time correlates factors such as cost of goods sold (COGS), degree of obsolescence and degree of perishability, the fixed/variable cost structure of the production process and the degree of flexibility of the manufacturing process, to the production decision for stocks or upon request.
The diagram in Figure 4 registers macro-steps for the development of a Product-Strategy Matrix adapting the design of the supply chain to specific clusters of products within the assortment of a fashion retail company.
Figure 4 – Macro-steps for preparing the Strategy-Product Matrix
The design of supply strategies consists of specifying the tactical and operational measures to be taken together, aiming to maximize the performance of the products (impact on the company's strategic indicators) associated with it.
In operational terms, these decisions can be translated in several ways. The decision on product flow characteristics can vary from a fully pushed flow, a fully pulled flow, or a mixed flow, where an initial batch of supplies is sent to stores, and later batches wait at the supplier or in a distribution center. to be moved to stores according to demand.
The company may want to assume the costs of reallocating certain product inventories between stores according to the item's value and transportation costs. Product clusters can be stored on a CD, with stock security parameters for continuous supply of stores according to demand. In other cases, a premium-priced product could be held in a distribution center or a small set of stores with the highest demand, and rely on a quick-response (premium shipping) scheme to restock other stores.
Once the strategies have been designed, the next step is to assess any restrictions on their implementation and needs for adapting internal processes.
- Impacts on suppliers: new ways of working may be required of suppliers, such as, for example, changes in the frequency of placing orders, different sizes of purchase batches or reduction in lead-time for resupply. Some of these measures generate an increase in operational complexity for suppliers, which may have an impact on costs or the need to develop new suppliers.
- Impact on the need for area and layout of distribution centers: the adoption of strategies with flows of products pulled or mixed means that there is a need to adjust the size of the DCs to absorb a greater volume of centralized inventory.
- Impact on systems: need to adapt transactional systems to the new operational requirements and automate the calculation of parameters in the Product-Strategy Matrix.
- Demand and inventory planning: new methods of sales forecasting and planning can lead to gaps in systems and training of the team responsible for this planning, as well as in parameterization of systems for the proper management of the product cycle.
- New operational processes: specific sourcing strategies and new product flows (such as, for example, reallocating inventory between stores or express delivery with premium transport) can lead to the need for new processes, with fiscal impact, in systems and providers of logistics services .The starting point for collecting and processing data should be the company's product hierarchy. Usually, this is a marketing decision that helps the Purchasing/Commercial area in its internal structuring, according to the consumer profile. However, even if there is a previous qualitative classification of the products (fashion, basic, etc.), this should not be considered for purposes of allocation to the strategies, since the quantitative measurement of the characteristics of the products may indicate that they behave in a similar way. demand different from that considered by the company.
Another point of attention is the level of data aggregation (see, for example, a product hierarchy and its different levels of aggregation in Figure 5): fashion retail companies tend to accumulate a very high and detailed number of sales data of its products, and analyzing them at the maximum level of detail would lead to an impractical number of clusters for evaluation, with non-perennial classifications. It is desirable that the products are grouped at a level that, on the one hand, facilitates the operational task of analysis (which will have to be replicated from time to time, as the portfolio of products is changing), and at the same time produces classification and differentiation relevant across product clusters.
Figure 5 – Product hierarchy of a retail chain
With the data in hand, the next step is to analyze and measure the characteristics of the products. The definition of calculation formulas for each characteristic is the object of discussion in this stage, identifying database fields that should be used.
Among these, the parameters described in Figure 6 stand out.
Figure 6 - Matrix of parameters for decisions
Once the parameters in each characteristic have been calculated, the products must be segmented into two sets based on the definition of a border point among the values found, generating the qualitative classification that the items must assume in those characteristics. For example, class A or B volume, low or high amplitude, high or low margin, short or long life cycle.
Figure 7 – Example of Product-Strategy Matrix
Other characteristics can also be incorporated into the analysis, according to the particularities of each company. However, it is recommended to keep the number of analysis drivers limited, to facilitate the allocation of products to the strategies and, subsequently, the implementation and understanding of the strategies by the operational teams. Each analysis driver must be tested to assess whether it generates significant differentiation in relation to those already analyzed.
The elaboration of a Product-Strategy Matrix should seek simplification, based on an effort to condense clusters that are not relevant in terms of quantity of products, sales or margin into a few strategies. A matrix that presents a high number of alternative strategies can significantly increase the cost of planning and operational control.
In addition to the relevance of the cluster, operational restrictions can contribute to the impracticability of a given strategy. For example, the determination of centralizing the stock of a cluster of products in suppliers may come up against its lack of storage capacity. Likewise, the lack of a premium transport service that delivers a significant cut in transport lead-time for a supply flow may imply a review of a certain strategy.
Figure 8 - Description of strategies
The grouping of strategies into macro-strategies can also be used to reduce risks in the implementation process. A strategy for a small cluster can be implemented in a second moment, with a higher level of process maturity, in order to reduce risks inherent to the learning curve.
Validation of the Product-Strategy Matrix also includes the elaboration of Business Cases that detail the impacts of the main strategies on the main indicators of the company, the potential for increasing sales by reducing Stockouts in stores, or reducing sales in liquidation (Markdown) . These results should also be analyzed as a measure of relevance for the adoption of differentiated strategies for small clusters of products.
After validating the Product-Strategy Matrix, processes and operational parameters must be redesigned and evaluated, focusing on the analysis of potential restrictions and risks. The final product should be a Roadmap for the Implementation of Supply Strategies.
CONCLUSION
Starting from the premise that the large assortment of products present in fashion retail chains can be grouped into clusters that present different characteristics and demand behavior, the article suggests a methodology to develop and implement a set of different supply strategies, more aligned with the characteristics of each cluster of products. Strategies basically answer questions about characteristics of product flows, sales forecasting techniques, inventory position in the supply chain, and production to stock or to order.
The process supports companies in expanding the level of knowledge about their products and establishes a Framework for strategic decisions in the supply chain, at a time of expansion of the store network, which tends to increase the complexity of the supply chain.
BIBLIOGRAPHIC REFERENCES
FISHER, ML What is the Right Supply Chain for Your Product? Harvard Business Review. March-April 1997.
WANKE, P. The Impact of Product Characteristics, Operation and Demand on the Type of Organization of the Product Flow: Field Research in Six Sectors of the Ranking Exame Melhores e Maiores. ILOS Institute, 2002. Available at:
https://ilos.com.br/web/index.php?option=com_content&task=view&id=1164&Itemid=74&lang=pt
WANKE, P. Logistics Positioning Strategy: Concepts, Implications and Analysis of the Brazilian Reality. ILOS Institute, 2001. Available at:
https://ilos.com.br/web/index.php?option=com_content&task=view&id=992&Itemid=74&lang=pt
