Retail assortment matrix: from SKU list to shelf display
Managing an assortment based purely on experience and intuition is becoming increasingly difficult. Category managers have access to data on sales, stock levels, turnover, deliveries, store characteristics, and shopper behaviour. However, this data alone does not answer the main question: which products should be stocked in each specific store.
An assortment matrix makes it possible to formalise these decisions. It defines the category structure for different stores and helps manage changes: introducing new items, delisting positions, and accounting for differences in demand, formats, and other operational nuances.
Approving the matrix is only part of the process. An SKU must arrive at the store, be included in the planogram, and ultimately end up on the shelf. It is therefore essential not only to construct the assortment correctly, but also to trace the entire product journey from the listing decision to the actual display.
Let us examine how an assortment matrix is structured, what forms the foundation of its creation, why stores require different sets of SKUs, and how to maintain their relevance.
What an assortment matrix is and why it is needed
An assortment matrix is a list of products that must be stocked in specific stores within a retail chain. It shows not the entire set of SKUs available to the company, but rather their distribution across retail outlets.
For example, the chain’s total assortment may number several thousand SKUs, but an individual store only requires a portion of them. The composition depends on the format and floor area of the retail outlet, the region, demand, price segment, and other factors. Therefore, the same category can vary significantly in the number and selection of items across different stores.
In its simplest form, the matrix can be represented as a table where the rows are SKUs, the columns are stores, and their intersection indicates the presence of the product. In practice, it may contain additional parameters: category and subcategory, brand, supplier, price segment, SKU status, as well as introduction and delisting dates.
These data are used across various processes: from managing category composition and purchasing to inventory replenishment and display planning. At the same time, the assortment matrix should not be confused with the product catalog. The latter may store all items that the company works with or has previously worked with, whereas the matrix determines which of them should form part of a specific store’s assortment during a given period.
As a retail chain grows, managing such links becomes increasingly complex. If stores differ from one another, a simple binary division of "in the chain / not in the chain" is no longer sufficient. It is necessary to determine which products to include in each retail outlet’s assortment and according to which principles.
How an assortment matrix is formed
Developing an assortment matrix begins with understanding what the overall category offer should look like. What customer needs does it address? What role does it play for the store? Which price segments need to be represented? The answers to these questions establish the framework for selecting specific SKUs.
One of the primary references is demand. Sales data helps identify popular positions and evaluate the contribution of individual products to overall revenue. However, retaining only top sellers would be a mistake. A category requires core products, offerings across various price points, new arrivals, and seasonal items. Commercial terms and supply stability are also of critical importance.
When forming the matrix, several groups of factors are taken into account:
- Category structure: subcategories, brands, price segments, as well as the breadth and depth of the offering.
- Customer demand: sales volume, purchase frequency, regional, and local preferences.
- Store characteristics: format, floor area, location, and shopper profile.
- Product operational parameters: supply conditions, minimum order quantities, shelf life, and seasonality.
- Business objectives: private label expansion, new product launches, promotional activity, and supplier agreements.
A portion of the assortment typically forms a stable core — items present across many or all stores of a given format. This core is then supplemented to account for local demand and specific store features.
New products require a distinct approach, as they lack an established sales history. Their potential is evaluated using metrics from comparable items, category performance data, price positioning, and demand forecasting. Once sufficient data has been gathered, the decision regarding the SKU’s presence can be reviewed.
The category manager’s objective is not to crowd the matrix with as many promising products as possible, but to craft a balanced proposition. An excessive selection intensifies internal competition between similar items, increases inventory levels, and complicates category management.
Furthermore, a single set of SKUs will not necessarily suit the entire retail chain. The next step is to determine which stores require a specific assortment and according to which criteria they should be grouped together.
One chain — different assortment matrices
Even stores operating within the same format face different conditions. Location, shopper profile, average basket size, and demand for specific product groups vary. Consequently, an identical assortment matrix does not always suit the entire retail chain.
Creating a separate set of SKUs for each retail outlet is feasible while the chain is small. Once there are dozens or hundreds of stores, managing such a multitude of variants becomes complex, which is why retail outlets with similar characteristics are grouped into clusters.
The simplest approach is grouping by format. For example, a hypermarket is assigned an expanded assortment, a supermarket a core assortment, and a convenience store a more compact one. Floor area, region, location, and shopper profile can also be taken into account.
However, even stores of the same format and comparable floor area can differ significantly in their demand structure. Therefore, clustering is also carried out at the level of individual categories. In this case, retail outlets are grouped based on similarities in sales for a specific product group. A single store may belong to one cluster for dairy products and to another for household chemicals or ready meals.
For instance, an analysis of coffee sales across a chain of 100 supermarkets revealed three demand patterns: in one group, the share of whole bean coffee is higher; in another, instant coffee; and in the third, coffee pods. Instead of an identical matrix for all stores, a core assortment can be formed and supplemented with items in demand within each cluster.
In this way, the chain adapts its offering to demand without having to manage the assortment of every retail outlet manually. At the same time, the composition of clusters needs to be periodically reviewed: consumer preferences shift, new stores open, and new sales data accumulates.
The composition of SKUs itself also changes. Some products are introduced into the chain, others have already accumulated sufficient statistics, and others are being prepared for delisting. Therefore, it is important to track not only where a product should be sold, but also its current status.
Matrix changes alongside the assortment: introducing and delisting SKUs
An assortment matrix reflects the state of the product range during a specific period. To manage the dynamics of these changes, each SKU is assigned a status: for example, "new", "active", "temporarily blocked", "marked for delisting", or "delisted". The specific set of statuses and transition rules depend on the business processes of the individual retail chain.
How to introduce new SKUs
When introducing a product, it is necessary to define what objective it serves and in which stores it should appear. A new arrival may broaden choice within a subcategory, cover a specific price segment, or replace an existing item.
A launch does not necessarily need to take place across the entire chain at once. An SKU can initially be included in the matrices of selected clusters, evaluated based on results, and then rolled out more widely. If a new product replaces another item, its introduction and the other’s delisting should ideally be synchronised to avoid temporary duplication of positions.
Active assortment also requires review
An "active" status does not guarantee a product’s permanent presence in the matrix. As data accumulates, sales performance, inventory turnover, margin profile, supply stability, and the SKU’s role within the category are continuously evaluated.
Such reviews are particularly crucial following a test launch. Based on the outcomes, a product can be retained in the selected stores, expanded to other clusters, or slated for delisting.
How to delist a product from the matrix
Delisting an SKU is not merely a matter of deleting an entry from a spreadsheet. First, a decision is made to discontinue sales, after which ordering is stopped or restricted while remaining stock is taken into account. Only once this process is complete is the product fully removed from the active assortment.
The pace of delisting can vary across stores: in some locations, stock will run out within days, while in others, sell-through will take longer. Statuses and key dates help manage this transition and distinguish the active assortment from lines that are already exiting the category.
However, changing a status in the matrix does not mean the update immediately translates to the store level. Discrepancies can easily arise on the path from the category manager’s decision to the physical shelf.
Why the assortment matrix diverges from reality
A product may be listed as active and assigned to a specific store, but this does not mean a customer will actually find it on the shop floor. Between the decision to list an SKU and its physical display lies a chain of events: delivery, ordering, replenishment, planogram updates, and store staff execution. A breakdown at any of these stages leads to a mismatch between actual stock availability and system data.
In practice, several typical scenarios arise:
- The SKU is in the matrix, but absent from the store: this can be caused by delivery delays, ordering errors, or insufficient stock allocations.
- The product has arrived, but is not included in the current planogram: for example, the matrix has already been expanded, but the shelf layout scheme has not yet been updated.
- The SKU has been delisted from the matrix, but remains on the shelf: this is acceptable during a clearance period to sell off remaining inventory, but requires ongoing monitoring.
- The assortment does not match store capacity: certain items may not fit onto the existing fixtures or may be allocated insufficient space.
- The actual shelf display differs from the approved version: the planogram has been updated, but changes on the shop floor have been executed only partially or with discrepancies.
The root causes of these situations vary; therefore, it is first necessary to identify where the misalignment occurred: during data management, delivery, display planning, or directly within the store.
For a category manager, this means that simply managing SKU store assignments is not enough. It is essential to ensure that an assortment decision has actually reached the store and can be effectively implemented within the existing shelf space. At this point, the assortment matrix connects directly to the planogram.
From assortment matrix to planogram
An assortment matrix defines which SKUs must be stocked in a store. A planogram translates this decision into a physical display, showing where each product will be located and how much space it will be allocated. At this stage, it becomes clear whether the chosen assortment aligns with the capacity of a specific retail outlet.
Check whether the assortment fits
Even two stores operating under the same format can vary in fixture configuration and dimensions. Therefore, an identical SKU count does not guarantee that products can be arranged using the same template.
When building a planogram, factors such as packaging dimensions, shelf sizes, number of facings, adjacent grouping rules, and other merchandising requirements are taken into account. As a result, it may turn out that there is insufficient space for certain lines. This situation often arises after expanding a category or introducing several new products without revising the existing assortment.
The reverse can also occur: the fixtures may accommodate a broader category offer than the current matrix provides for. This can be factored into the next assortment review.
Allocate space between SKUs
Simply being included in the planogram is not enough — securing an appropriate amount of shelf space is crucial. Products within the same matrix can differ significantly in demand, packaging dimensions, and sales velocity; therefore, assigning an identical number of facings across all items is rarely justified.
If a high-demand SKU is allocated too little space, the product may run out on the shelf before the next replenishment cycle. Conversely, excessive space given to a slow-moving item restricts opportunities for other products within the category. Space allocation must therefore balance sales data, stock levels, and required shelf capacity.
Identify discrepancies before updating display schemes
Aligning the matrix with the planogram makes it possible to detect discrepancies before the updated layout is communicated to the store. For instance, it helps identify an SKU that is included in the assortment but missing from the planogram, or an item that has been delisted from the matrix yet remains on the shelf layout.
When managing a large store network, conducting such checks manually is difficult. In Greenshelf, teams can work with assortments directly during planogram creation and analysis, compare scenarios, and flag products missing from the layout. This ensures category updates are immediately evaluated against their implementation on specific store fixtures.
Ultimately, the planogram illustrates how an assortment decision will be executed on specific equipment and whether sufficient space is available. The next objective is to maintain this connection through subsequent assortment revisions.
How to keep an assortment matrix up to date
An assortment is in constant flux: new SKUs are introduced, supplier terms are renegotiated, seasonal campaigns draw to a close, certain lines are delisted, and others are rolled out to additional stores. The larger the retail chain, the more of these processes take place simultaneously. Consequently, the matrix must be updated regularly, with new data transmitted to connected systems in a timely manner.
Synchronise data
Information used for assortment management is typically distributed across multiple systems. The product catalog contains SKU specifications, the ERP holds procurement and inventory data, and the assortment management system stores store-level listing assignments. Sales figures, planograms, and automated replenishment parameters are maintained separately.
If data exchange is incomplete, information must be transferred manually. An SKU may already be marked as active in one source while remaining absent in another; a delisted item may linger on an active planogram; or updated packaging dimensions may fail to be reflected during display planning.
It is therefore critical to establish single sources of truth and clear synchronisation protocols. Whenever an SKU status changes, that update should flow seamlessly through all dependent processes.
Track changes, not just final lists
The matrix should be audited continuously, rather than solely during major category reviews. Introductions, substitutions, and delistings occur between main review cycles, with each decision potentially affecting dozens of retail outlets.
Tracking which SKUs have been added or removed, which store listings have changed, and when a new status comes into effect is essential. This makes it far easier to spot instances where a decision has been approved internally but remains unexecuted at subsequent operational stages.
The frequency of full category reviews depends on the specific category. Seasonal ranges require frequent adjustments, whereas stable product groups call for less regular intervention. Significant shifts in demand, supply chain disruptions, new store format launches, or major category revamps serve as key triggers for additional reviews.
Connect the matrix with downstream processes
The work does not end once a matrix entry is updated. Adding an SKU to a store must trigger corresponding actions in purchasing and planogram development. Delisting an item requires aligning order freezes, clearance of residual stock, and shelf layout adjustments.
To achieve this, systems for managing assortments, master data, planograms, inventory, and analytics must be integrated. Minimising manual handoffs between steps significantly reduces the risk of different departments operating with conflicting versions of the assortment.
Within this chain, Greenshelf is responsible for translating assortment decisions into physical shelf displays. Integration with corporate systems enables the platform to ingest necessary product master data and pass operational outputs downstream. When creating planograms, teams can verify SKU compositions, evaluate changes, and generate up-to-date display schemes for retail outlets.
In this way, managing the assortment matrix becomes a continuous process: decisions are distributed to the designated stores, transmitted to connected systems, and executed on the shop floor. This ensures sustained alignment between the approved assortment and what is actually made available to shoppers in-store.
Conclusion
An assortment matrix does not merely define a master list of products for a retail chain; it shapes category compositions for specific stores based on demand, format, and other retail outlet characteristics. As the product offer evolves, the matrix is reviewed: new SKUs are introduced, active lines are redistributed across stores, and obsolete items are delisted.
However, the final outcome depends on how effectively these decisions are executed in practice. An item must not only exist within the matrix, but also reach the designated store, feature on the current planogram, and appear on the physical shelf. It is precisely this connection that transforms the assortment matrix from a static list of SKUs into an operational tool for category management.