Resources/CPG Finance 101/
Retail Inventory Forecasting by Retailer: How Each Account Orders and What to Model
Demand Planning & Inventory Forecasting

Retail Inventory Forecasting by Retailer: How Each Account Orders and What to Model

How mass, club, grocery, drug, specialty and natural accounts actually order from a CPG brand, and the math to put behind each one.

12 min read
Updated October 2026

Quick answer

Retail inventory forecasting predicts what each retailer will order, when, and in what quantity, then works back to production and cash. Start from sell-through (units per store per week x doors), convert it to the retailer's replenishment cadence using its target weeks of cover, then layer in promotions, deductions and payment terms. Each account orders differently: DC replenishment, direct-to-store, or through a distributor such as UNFI or KeHE.

Retail inventory forecasting means predicting what each retail account will order from you, when, and in what quantity, then working back to production, lead times and cash. Unlike DTC, you are forecasting a buyer's replenishment system, which lags sell-through and cuts POs on its own cadence. Here is how the main US and Canadian accounts typically order and what to model for each.

Why retail is not DTC with a bigger number

A shopper's purchase (sell-through) and your revenue (the retailer's PO, or sell-in) are separated by the retailer's inventory position. Forecast sell-through first and derive sell-in from it, or a pipeline fill looks like demand; the sell-in vs sell-through variance template exists for that reconciliation. Orders are lumpy, since a DC orders a case when cover drops below a threshold. Cash lags: inventory is bought 60 to 120 days before the PO and paid 30 to 90 days after shipping, less deductions (see our CPG cash flow forecasting guide).

Two formulas do most of the work. Weekly demand at retail = units per store per week x active doors. Weeks of supply = (units at the retailer's DCs + units on shelf) / weekly demand. When weeks of supply falls toward the retailer's target cover, a PO is coming; well above it, the next PO will be late or small.

The three ordering models

  • DC replenishment. Most mass, grocery, drug and specialty accounts. You ship to the retailer's DCs and its system cuts POs from DC inventory, store demand and target cover. Fill rate and on-time delivery are measured; shortfalls are typically charged back.
  • Direct-to-store (DSD). Beverages, chips, bread and some dairy and frozen, delivered to stores by your trucks or a DSD distributor. The forecast is doors x units per door per week, adjusted for displays and promotions.
  • Distributor (UNFI, KeHE, Dot Foods, regional specialists). The retailer orders from the distributor, the distributor orders from you, and you see demand only through movement reports: a large stocking order, margin and fees between gross and net, and a lagging signal. The UNFI template and KeHE template model both.

Mass and club

We cover Target and Costco in dedicated guides and the Walmart math in our retail launch financial model.

Sam's Club

Sam's Club is Costco's logic on Walmart's systems: club-pack sizing, roughly 600 US clubs and a tight item count, so each SKU carries real volume per door. Replenishment and compliance run through Walmart's supply chain, so on-time, in-full delivery to the DC matters. Model pallet-level orders, an initial fill across your clubs, and the real chance the item rotates out after a cycle.

BJ's Wholesale

BJ's is the East Coast club, a few hundred locations from New England to Florida. It typically carries more SKUs per category than Costco and accepts manufacturer coupons, which changes promo math. Orders flow through BJ's DCs in club packs. Model fewer doors with higher units per door than grocery, plus events that pull demand forward. Confirm continuous versus in-and-out before sizing production.

Dollar General

Dollar General has the largest US store count, well over 19,000 small-format, mostly rural stores. Shelf space is tiny, so planogram placement is everything, and price points are typically channel-specific with smaller sizes. Replenishment runs through its own DCs. Model very low units per store per week across a huge door base, treat the first PO as a pipeline fill several times steady-state demand, and confirm DC assignments.

Family Dollar

Family Dollar, part of Dollar Tree until its 2025 sale, follows the dollar-channel pattern: thousands of small stores, tight planograms, channel-specific packs and DC replenishment. Expect regional rollouts tied to planogram windows and a buyer focused on price point and velocity per linear foot. Model the pipeline fill, then low, steady velocity. Systems have been in transition, so confirm the current portal and terms.

Five Below

Five Below is a trend-driven value retailer with roughly 1,800 stores and an assortment that turns over seasonally. Buys often behave like one-time programs rather than continuous replenishment: a defined quantity, a set in-store window, limited reorder. Model a program buy with a hard end date, size production to the committed quantity plus a small buffer, and keep ongoing velocity out of the base forecast until a reorder arrives.

Meijer

Meijer is a privately held Midwest supercenter chain with roughly 250 stores in six states. It orders through its own DCs for most center-store and frozen items and accepts DSD in the usual categories. Regional and private, it leans on buyer relationships and shares less data than national mass. Model a small door count with supercenter-level velocity, weekly ad spikes and seasonal resets.

Grocery

Grocery mixes all three models, sometimes within one store. Our Kroger inventory forecasting guide covers the largest traditional grocer.

Albertsons Companies

Albertsons operates more than 2,200 stores under Safeway, Vons, Jewel-Osco, Shaw's, Acme and other banners. Buying mixes national category management with division-level decisions, so a national authorization does not put you in every division. It replenishes through its own DCs, with DSD in beverage, snack and bakery. Model authorized doors by division, track scan data per division, and build in the weekly ad calendar.

H-E-B

H-E-B is a privately held Texas grocer with around 400 stores, a loyal shopper base and strong private label. It typically orders through its own DCs and runs its own transportation, with DSD in standard categories. Velocity per store runs high relative to door count, so 400 H-E-B stores can out-ship 1,000 elsewhere. Model per-store velocity carefully, plan around H-E-B's own events, and confirm data access.

Hy-Vee

Hy-Vee is an employee-owned Midwest grocer with roughly 280 mostly large-format stores across eight states. It replenishes through its own distribution with the usual DSD categories, and store directors hold more assortment autonomy than at national chains, so authorization and placement can diverge. Model a smaller door count with store-level variability, confirm how many stores have actually set the item, and treat store data as the signal.

Harris Teeter

Harris Teeter is a Kroger-owned Southeast and Mid-Atlantic banner with roughly 250 stores in higher-income trade areas. It has historically kept some buying separate from the broader Kroger divisions, so authorization is often its own conversation, and replenishment runs through its own DCs. Model a premium regional grocer: strong natural and specialty velocity, e-VIC promotions, seasonal resets. Confirm whether data arrives through Kroger's systems.

Loblaws

Loblaw is Canada's largest grocer, with Loblaws, No Frills, Real Canadian Superstore, Zehrs, Provigo and Shoppers Drug Mart. Buying is organized by banner and division with its own DCs and strong private label. Model door counts by banner and province, factor in bilingual labeling and duty on US-made goods, and expect listing fees and promotional commitments as part of authorization. PC Optimum promotions drive lift.

Sobeys

Sobeys, owned by Empire, runs roughly 1,500 Canadian stores under Sobeys, Safeway, IGA, Foodland, FreshCo and Farm Boy. It replenishes through regional DCs, and buying keeps a regional character in Quebec and Atlantic Canada. Model it province by province rather than as one national account, include listing fees and the Scene+ promo calendar, and confirm which banners the authorization covers.

Metro

Metro is a Quebec and Ontario grocer and pharmacy operator with roughly 950 food stores under Metro, Super C, Food Basics and Adonis. Replenishment runs through its own DCs, and French-language packaging compliance in Quebec is non-negotiable. Model a two-province account with heavy weekly flyer promotion, confirm listing fees and the promotional calendar, treat the first order as a pipeline fill, and ask what movement reporting exists.

Drug and convenience

Small shelves, high door counts, planogram-driven resets, low velocity per door, and an opening PO that can be a large share of first-year volume.

CVS

CVS runs roughly 9,000 pharmacies and replenishes through its own DC network, with planogram resets on a seasonal or semiannual schedule. Front-store buyers manage tight assortments, and ExtraCare promotions drive much of the lift. Model low units per store per week across a large door base, size the pipeline fill to planogram facings, and build chargeback exposure for fill rate and labeling compliance into the plan.

Walgreens

Walgreens operates around 8,000 US stores with DC replenishment and scheduled reset windows. Its front-store strategy has been in flux, with closures and assortment rationalization, so door counts can move during your program. Model authorized doors conservatively, revisit store count quarterly, and use myWalgreens promotion windows in the forecast. Expect compliance chargebacks for shortages and late deliveries, and confirm terms in writing.

Rite Aid

Rite Aid went through bankruptcy proceedings and store closures from 2023 through 2025, and its footprint and operations changed substantially. Start any model with what currently exists: confirm active store count, DC structure and payment reliability with the buyer or your broker before shipping. If you have an active program, treat it as a small regional drug account with pipeline fill and low velocity, and keep credit exposure tight.

7-Eleven

7-Eleven is the largest US convenience chain, with many franchisee-operated stores that hold discretion over part of their assortment. Corporate authorization gets you on the recommended list; actual placement depends on franchisee adoption and the distributor (often McLane or Core-Mark) servicing the stores. Model adoption rate x stores x low units per store per week, expect single-serve packs, and read demand through distributor movement.

Specialty beauty and pet

Fewer SKUs, higher velocity per door, and more asked of the brand in testers, gratis and marketing funds. Sephora, Ulta and Chewy have their own guides.

Credo Beauty

Credo is a clean beauty specialty retailer with under 20 stores and a meaningful online business. It enforces a strict ingredient standard, and orders are small, frequent and closely tied to sell-through. Model a low-door, high-touch account: modest opening order, replenishment in small waves, dotcom often outselling stores, testers and gratis as a real cost line. Confirm the return and markdown policy for slow items.

PetSmart

PetSmart operates roughly 1,600 North American stores and replenishes through its own DC network, with scheduled planogram resets and strong private label. Model DC replenishment with store-level velocity, size the pipeline fill to facings per store, and plan for its promotional calendar and Treats loyalty events. Consumables repeat steadily, so velocity stabilizes faster than in beauty or snacks. Confirm all-store versus cluster authorization.

Petco

Petco runs roughly 1,500 pet care centers, ships from its own DCs, and offers a vendor portal for POs and performance data. Its assortment leans premium and natural, and Vital Care membership drives repeat on consumables. Model it much like PetSmart: DC replenishment, pipeline fill by facings, steady repeat on food and treats, promotional spikes around events. Confirm whether petco.com inventory is planned with stores.

Tractor Supply

Tractor Supply is a rural lifestyle retailer with more than 2,200 stores and a large pet and animal category that often beats pet specialty on bag sizes and per-store food volume. It replenishes through its own DCs with a mix of continuous and seasonal programs. Model a large-door rural account with bigger packs, the Neighbor's Club calendar and seasonal swings. Confirm which store formats carry your item.

Grove Collaborative

Grove is an online retailer of household, personal care and wellness products, built on subscription and now operating as a curated marketplace with recurring orders. It buys wholesale and ships from its own fulfillment centers, so POs arrive as inventory draws down. Model a small online 1P account: modest opening order, replenishment tied to on-hand and sales rate, recurring orders that steady demand. Confirm wholesale versus marketplace terms.

Thrive Market

Thrive Market is a membership-based online natural grocer that buys wholesale and ships from its own fulfillment centers, with curated natural assortments and strong private label. Membership and auto-reorder steady demand. Model an online 1P account with POs to a few fulfillment centers, sell-through-driven replenishment, and pushes around membership events and seasonal boxes. Confirm case pack requirements and which fulfillment centers receive your items.

Natural and distributor-fed accounts

In natural grocery you usually sell to a distributor, so the forecast has two layers: what shelves consume and what the distributor orders. Our Sprouts template and Whole Foods template cover two retailers fed this way; the CPG template library holds the gross-to-net models.

Fresh Thyme Market

Fresh Thyme is a Midwest natural grocer with roughly 70 stores, concentrated in Illinois, Indiana, Ohio, Michigan and nearby states. Most center-store and frozen items reach it through UNFI or KeHE, with DSD in bakery and some perishables, so your PO usually comes from the distributor, not Fresh Thyme. Model a small door base with natural-channel velocity, demo commitments, and the distributor lag between shelf and order.

UNFI

UNFI is the largest natural and specialty distributor in North America and the primary distributor for Whole Foods. You sell to UNFI, UNFI sells to retailers, and you see demand through weekly movement by DC. Expect an initial stocking order several times steady-state demand, margin and program fees (fill allowances, promotional allowances, pass-through slotting) between gross and net, and ordering that lags the shelf by weeks.

KeHE

KeHE is the other national natural, specialty and fresh distributor, serving Sprouts, conventional grocers' natural sets and independents. Its pattern mirrors UNFI's: DC-level stocking orders, weekly replenishment from movement, and a gross-to-net stack of margin, promotions and administrative fees. KeHE's own promotional calendar and trade shows create order spikes. Model each DC's movement, plan allowances as a percentage of gross, and reconcile deductions monthly.

Dot Foods

Dot Foods is a redistributor: it buys truckloads from manufacturers and sells mixed less-than-truckload loads to other distributors, foodservice operators and retailers. You ship large, infrequent orders to Dot's DCs and see demand only indirectly as its customers pull. Model a lumpy, DC-level account with long gaps between POs, never read a Dot order as consumer demand, and ask which Dot DCs stock you.

Rainforest Distribution

Rainforest is a regional natural and specialty distributor serving the Northeast, especially the New York metro area and independent natural retailers. Orders are smaller and more frequent than a national distributor's, and store-level buyer relationships carry more weight. Model a modest opening order, weekly replenishment tied to the stores it serves, and promotions negotiated directly. Ask for movement by retailer, your only view of sell-through.

Dora's Naturals

Dora's Naturals is a Northeast distributor focused on refrigerated and perishable natural products, with DSD-style delivery to stores in and around New York and New Jersey. Cold chain and frequent delivery mean small, regular orders, and shelf life dominates the forecast. Model weekly orders tied to store count and velocity, hold limited safety stock, and confirm minimums, delivery days and how spoilage is shared.

A note on distributor data systems. Two names from beverage distribution are not distributors or retailers but the software distributors run on; what matters is the data they give you.

VIP (Vermont Information Processing)

VIP is the route-accounting and ordering system many beer, wine and spirits distributors run on. If yours do, the useful outputs are depletions (distributor sales to each retail account), distributor inventory by SKU, and your shipments in. Depletions are your closest proxy for sell-through, shipments are sell-in, and the gap is distributor inventory, best expressed as days on hand. Forecast from depletions by account and hold days on hand at the distributor's target. Access typically depends on your supplier agreement, so confirm what you can see.

Encompass (Encompass Technologies)

Encompass is another beverage distribution ERP and data platform, used by distributors instead of or alongside VIP. The forecasting logic is the same: pull depletions by retail account, distributor on-hand by SKU and inbound orders, then model your sell-in as the quantity that holds distributor days on hand at target as depletions move. Access typically runs through the distributor or a supplier portal, and field names and cadence differ from VIP, so confirm what is available before building around it.

What to model for any retail account

Seven inputs cover every account above: authorized doors by division and set date; units per store per week by SKU; the pipeline fill (facings x doors plus DC safety stock); steady-state demand and the retailer's target cover; the promotional calendar with lift and pull-forward; deductions and distributor fees as a percentage of gross; and payment terms plus lead time, so each PO lands on the cash view on the right dates. See purchase order forecasting for PO-to-cash mechanics and SKU-level demand forecasting for the method underneath.

Where Drivepoint fits

Drivepoint is the AI finance platform for consumer brands, built for this many-account, many-channel reality. It connects retailer and distributor data alongside Shopify, Amazon and your accounting system through 100+ integrations, rolls store and DC movement into a SKU-level forecast inside your Excel model, and carries every PO through to the P&L, balance sheet and cash flow. Its 140+ retail-specific reports include sell-in versus sell-through and weeks of supply by account, and AI scenario planning lets you test a new authorization or a distributor switch in minutes, not days.

Frequently asked

Questions, answered

How is retail inventory forecasting different from forecasting DTC sales?

In DTC, a sale and the revenue happen together and you see demand instantly. In retail, shoppers buy from the retailer's inventory and the retailer orders from you later, in case or pallet quantities, based on its own cover targets. You forecast consumer sell-through first, then translate it into the retailer's lumpy PO pattern, and finally into cash, which arrives 30 to 90 days after shipping, less deductions.

What is the difference between sell-in and sell-through?

Sell-in is what the retailer or distributor orders from you. Sell-through is what shoppers buy off the shelf. They diverge whenever the retailer builds or drains inventory: an opening order is pure sell-in with no sell-through yet, and a slow quarter can show strong sell-through but no reorders if the DC is overstocked. Reorders follow sell-through, so forecast that first.

How do I forecast orders when I sell through a distributor like UNFI or KeHE?

Model two layers. First, retail consumption: authorized stores x units per store per week, from the distributor's movement reports. Second, the distributor's own ordering, which starts with a stocking order several times steady-state demand, then replenishes each DC as movement draws it down, typically lagging the shelf by a few weeks. Book distributor margin and program fees between gross and net so the forecast lands on net revenue.

How big should my first PO be for a new retail account?

Size it to the pipeline fill: facings per store x authorized stores, plus the safety stock the retailer's DC holds, which is typically a few weeks of expected demand. That opening order is often several times your steady weekly volume and will not repeat, so plan production for the fill plus the first replenishment cycles rather than annualizing it as if it were ongoing demand.

What is a good weeks of supply target at a retailer?

It depends on the retailer's lead time and how fast the item moves, but most replenishment systems aim to hold a few weeks of cover at the DC plus store shelf stock. Calculate it as (DC units + shelf units) / weekly sell-through. When it drops toward the target, expect a PO; when it climbs well above it, expect a gap. Ask the buyer what cover the system is set to.

Why do retailer POs not match my sell-through?

Because the retailer's system orders in case or pallet multiples when its cover crosses a threshold, not continuously. Promotions pull orders forward and leave a dip afterward, planogram changes trigger fills and returns, and DC rebalancing moves inventory without any shopper involvement. Track DC and store inventory separately, and the gap between POs and sell-through becomes a predictable inventory movement rather than a mystery.

See what Drivepoint
looks like for your brand.

Book a demo and see how quickly Drivepoint gets your complete financial model up and running — connected to your data, built for your channels, ready for your next big decision. Whether you're planning a retail launch, preparing for a raise, or replacing a spreadsheet that only one person can touch.

Book a demo