Ross Stores opened 51 new locations across 17 states in September and October, completing a 115-store growth plan for fiscal 2026 and lifting its total footprint to 2,377 stores. The off-price operator says it is positioned to reach 2,900 Ross Dress for Less stores and 700 DD's Discounts locations over time. The expansion lands on top of a 10% second-quarter comparable sales gain and a 68% jump in net earnings to $851.3 million. For supply chain and operations teams, the harder question is not how many doors open, but whether replenishment clocks can keep pace with them.
Key Conclusions
Off-price growth changes the economics of replenishment before it changes the economics of selling. A network that adds 115 doors in a single year spreads the same inventory pool across more shelves, so the cadence of allocation decisions matters more than the size of any single buy. Second-quarter comparable sales rose 10% and net earnings climbed 68% to $851.3 million (Retail Dive), which gives the retailer room to fund distribution capacity. Funding, however, is not the same as sequencing, and capacity that arrives a quarter late shows up as markdown risk rather than missed sales.
The second conclusion concerns assortment variability. Off-price formats buy opportunistically, which means the assortment in a new store is rarely a copy of an existing one. That makes planogram standardisation less useful and demand sensing more important. Retailers that forecast a new store from the chain average will systematically under-forecast the first two quarters, because there is no store-level history to regress against and the assortment itself is still settling.
The third conclusion is about format adjacency. Seasonal operators are converting temporary footprints into year-round concepts, and Spirit Halloween will expand its Christmas format to 44 locations this season (Retail Dive). The same physical site may therefore carry two different replenishment profiles within a single calendar year. Store networks are becoming less homogeneous, and the data model has to reflect that shift rather than assume every door behaves like the chain median.
Inside The Expansion: What 115 Doors Change
Growth at this scale is usually described in terms of square footage and store counts, but the operational constraint sits one layer down. Every new door requires an opening inventory package, a replenishment lane and a returns path. When 115 doors open within two quarters, those three requirements compete for the same distribution capacity that existing stores depend on, and the competition is resolved by priority rules that are often implicit rather than designed.
Allocation Before Assortment
The first discipline to install is an explicit allocation order. New stores need opening packages, high-volume stores need fast replenishment and low-volume stores need fewer, larger deliveries. Writing that order down and reviewing it weekly converts a set of ad hoc decisions into a repeatable policy. Retailers that skip this step usually discover the problem only after a peak week, when the fast replenishment lane has been consumed by opening packages for stores that have not yet opened.
The First Two Quarters Problem
A new store has no comparable history, which breaks the standard forecasting loop. The workable substitute is cluster history: group the new store with existing stores that share catchment characteristics, format and assortment mix, then use the cluster curve as the prior and blend in actuals as they accumulate. The blend weight should shift toward actuals over roughly eight to twelve weeks, and the cluster assignment itself should be revisited once the first quarter of data is complete.
Best Practices
Sequence Distribution Capacity First
Before committing to a store opening calendar, map the distribution capacity required by opening packages against the capacity available after base replenishment. If the two curves cross, either the calendar or the capacity plan has to move. Publishing both curves to commercial and operations teams at the same time prevents the common failure mode in which openings are committed commercially and then renegotiated operationally under time pressure.
Replace Store History With Cluster History
Cluster-based forecasting is only as good as the cluster definition. Build clusters from measurable attributes such as catchment income band, competitive density, format size and assortment breadth, then validate them by checking whether historical store performance within a cluster actually co-moves. Clusters that do not co-move should be split rather than averaged, because averaging hides exactly the variation that new-store forecasting depends on.
Both practices depend on the same underlying asset, which is a clean store master record. Store attributes, opening dates, format labels and assortment tiers should live in one system and be referenced by every downstream process. When those attributes are duplicated across planning, distribution and reporting systems, cluster definitions drift apart and the forecasting improvement quietly disappears within a couple of quarters.
Common Mistakes
The first mistake is treating store growth as a real estate problem rather than an inventory problem. Signing leases is a commercial activity with its own timeline, and that timeline rarely matches the distribution build-out. The second mistake is assuming that a proven replenishment model scales linearly with door count, when in practice the marginal cost of a new door rises once the network exceeds the density that the existing distribution footprint was designed for.
A third mistake is under-investing in the returns path. Off-price formats carry higher return rates in some categories, and a new store without a configured returns lane will push that volume back through the standard replenishment channel. The result is congestion precisely where the network can least afford it. Retail coverage of store operations consistently shows that the operational details of a new door, rather than its opening day, determine whether it contributes to chain margin in its first year.
Three Operating Metrics To Watch
The first metric is opening package fill rate, measured per store and reported weekly through the first quarter. A fill rate below target signals that allocation priority is being overridden by base replenishment, and it usually precedes a visible assortment gap that customers notice before planners do.
The second metric is replenishment lead time dispersion, not just its average. A stable average can hide a long tail of stores that wait twice as long as the median. Off-price retail already operates with lumpy supply, and Retail Dive shows how quickly a distribution or labour decision at one node propagates across the network. Reporting the ninetieth percentile alongside the mean makes that tail visible.
The third metric is sales density per square foot relative to the cluster prior. New stores that underperform their cluster for two consecutive quarters are usually signalling a catchment mismatch rather than an execution problem. Industry forecast pages, including NRF, provide the seasonal baseline needed to separate a weak store from a weak quarter, which matters when the whole network is absorbing a large opening programme.
Summary
Ross completing 115 openings in a single fiscal year is a clear statement about where off-price demand sits. The operational consequence is that replenishment stops being a background process and becomes a constraint that shapes the opening calendar. Sequencing distribution capacity, replacing store history with validated cluster history and publishing the right metrics are the three levers that make a large opening programme financially coherent.
None of these levers requires new technology. They require a store master record that every system references, an explicit allocation order that is reviewed weekly, and a reporting layer that shows dispersion rather than averages. Retailers that build those three things early can keep adding doors without degrading the service level that made the format attractive in the first place.
Data Sources
- Retail Dive: Ross opens 51 new stores, October 12, 2026.
- PR Newswire: Ross Stores opens 51 new locations, October 12, 2026.
- Retail Dive: Retail Dive operations coverage, October 2026.
- Retail Dive: Spirit Halloween to transform dozens of locations to Spirit Christmas, October 2026.
- Retail Dive: Inside Scheels' sporting goods wonderland, October 2026.
- NRF: NRF retail sales and holiday forecasts, October 2026.
FAQ
Why does store count change replenishment economics?
A: More doors spread the same inventory pool across more shelves, so allocation cadence and priority rules matter more than the size of any single buy, and late capacity shows up as markdown risk.
How should a new store be forecast without history?
A: Use validated cluster history as the prior, blend in actuals over roughly eight to twelve weeks, and revisit the cluster assignment once the first quarter of store data is complete.
What is the most useful replenishment metric?
A: Lead time dispersion rather than the mean, reported at the ninetieth percentile, because a stable average can hide a long tail of stores waiting far longer than the median.
Why does format adjacency complicate planning?
A: A single site can host a seasonal format and a year-round concept in the same year, which means two different replenishment profiles share one physical location and one delivery window.
What is the minimum data asset required?
A: A single store master record holding attributes, opening dates, format labels and assortment tiers, referenced by planning, distribution and reporting rather than duplicated across them.
References
- https://www.retaildive.com/news/ross-opens-51-new-stores/832721/
- https://www.prnewswire.com/news-releases/ross-stores-opens-51-new-locations-302903890.html
- https://www.retaildive.com/topic/operations/
- https://www.retaildive.com/news/spirit-halloween-store-openings-spirit-christmas-2026/832675/
- https://www.retaildive.com/news/scheels-store-experience-retail-best-kept-secret/831211/
- https://nrf.com/research-insights/forecasts
