Let AI summarize this article for you
What is the difference between space management and category management software in retail?
The two solve different problems, even though both deal with “the shelf.” Category management decides what to sell and how much commercial priority — role, target share, quota — each SKU or subcategory deserves within a group of stores; it’s a merchandising decision owned by category managers. Space management decides how that decision gets physically executed on a specific fixture — facings, positioning, planogram layout; it’s an execution decision owned by space planners. Retano treats these as separate but connected products rather than one blended tool: Retano Catman for the category-strategy side, Shelfplan for the physical execution side — with Shelfplan’s shelf-space capacity available to Catman as one input constraint when setting assortment quotas, not the other way around.
Which retail space optimization tools work best for grocery and FMCG businesses?
Grocery and FMCG chains face a clustering problem that generic space-optimization tools often miss: demand for the same category can vary sharply between store clusters even within the same format or region, because local shopper mix and competing categories differ store by store. A single, store-wide clustering scheme isn’t enough — Retano Catman builds store clusters separately for each category, based on actual demand patterns rather than store format, size, or region, so one category’s cluster map doesn’t have to match another’s within the same chain. This category-specific cluster structure is what Shelfplan then uses when allocating shelf space, so space decisions follow the same demand logic the category strategy was built on, rather than a generic, chain-wide grouping.
What criteria do large retail chains use when evaluating and selecting planogram software vendors?
Large chains typically weigh several criteria when comparing vendors: whether the system handles multi-format store networks without a separate configuration per format; how much of the planogram is generated automatically versus built manually against rules a planner still has to maintain; whether category strategy, the resulting planogram, and actual shelf execution are checked against each other automatically, or require a separate manual audit to confirm the plan was followed in stores; and total cost of ownership once implementation and ongoing configuration effort are counted, not just license price.
On the third criterion: Retano Catman defines the category strategy and target assortment, Shelfplan turns that into a planogram, and VeriShelf AI checks the actual shelf against the same planogram using computer vision, logging any deviation as a task in the Shelfplan app. These three stages sit within one vendor’s product set rather than needing a separate audit process layered on top
What is the difference between Retano Catman and ERP-embedded assortment planning tools?
ERP-embedded assortment planning tools run on top of a master-data layer: assortments are logical objects tied to merchandise category hierarchies, store/site records, and listing procedures that are configured in advance. Retano Catman takes a different starting point — store clusters are generated directly from ML analysis of actual demand patterns within each category, using receipt-level Big Data, independently of the store format, region, or master-data hierarchy. This means the cluster structure itself is a live analytical output specific to each category, rather than a hierarchy configured upfront in the ERP system.
How can category managers tell which store clusters have drifted from the category’s intended strategy?
Category role and strategy are usually set as a target at the start of a planning cycle — for example, marking a category as a traffic driver or a profit generator, with an assortment and pricing approach that follows from that role. Whether the target is actually being met varies by store cluster over time, and confirming it requires comparing real category performance back against that original assumption, not just re-reading the plan. In Retano Catman, this comparison runs as a standing ML-based analysis: actual category performance is checked against the planned role on an ongoing basis, and the output flags the specific store clusters where real commercial behavior has drifted from the intended strategy — giving category managers a shortlist to review instead of a manual re-audit of the full assortment to find where things went off track.
How does assortment optimization software decide which products to add or remove from a category, and can that decision be checked?
In Retano Catman, the add/remove/keep decision for a SKU is based on its net contribution to the category’s role in that store cluster — margin, revenue, or traffic, depending on which role the category has been assigned — not gross margin. The calculation nets out shrinkage, write-offs, back margin, and the cost of tied-up inventory, and uses robust statistical methods (median-based estimates rather than simple averages) so a single promotional spike or one-off stockout doesn’t distort the result; it also penalizes products whose performance depends on just a few outlier stores rather than being consistent across the cluster. The same product is scored separately at store, cluster, and network level, so a SKU can come out strong in one cluster and weak in another instead of getting one verdict for the whole chain. Every recommendation is delivered together with the reasoning behind it and how it differs from the current assortment, so a category manager can check why a specific product was flagged rather than accepting the output as a black box.
Can category managers set minimum requirements for private-label or local-brand share that assortment optimization has to respect, rather than just ranking every SKU by performance?
In Retano Catman‘s assortment structure model, category width isn’t the only thing that’s set — its composition is too, through minimum shares by subcategory, price segment, brand tier, packaging format, private-label share, and country of origin, among other product properties. These minimums act as direct constraints on the assortment rather than just inputs to a performance ranking, so a category manager can guarantee representation for groups that matter strategically — private label, local brands, a specific price tier — even in cases where ranking purely by sales or margin might otherwise reduce or remove them. Individual SKU recommendations are then generated inside those constraints, not instead of them.
How does assortment optimization avoid removing new products before they have enough sales data to be judged fairly?
New products go through a protected period in Retano Catman’s models while sales history builds up, so they aren’t scored — and potentially removed — by the same performance criteria applied to established SKUs before there’s enough data to judge them fairly. Once a product has a long enough track record, it can be extended to additional stores using the same demand-based logic used elsewhere in the model — either following the store clusters where it already has sales history, or network-wide if it’s meant to become a standard listing. Products that are already in decline go through a separate phase-out step, rather than being cut the moment their numbers dip.
