How to Forecast Inventory for Chewy: Autoship, Fulfillment Centers and the Ramp

Forecasting inventory for Chewy is closer to forecasting a subscription business than a retail shelf. Chewy buys your product wholesale, holds it in its own fulfillment centers and sells it to pet parents online, and the large majority of its sales come through Autoship, its recurring-order program. That means your demand signal is a subscriber base multiplied by a replenishment frequency, your POs are issued per fulfillment center rather than per store, and the biggest forecasting risk is getting the ramp wrong in the first six months while the Autoship base builds. This guide covers how Chewy orders, the subscriber math, the promotional and fill-rate realities, and a fictional treats brand carried through to cash.

How Chewy orders from a brand

Chewy operates as a first-party retailer: it issues purchase orders, takes title to inventory and sets the retail price. Product ships to Chewy's network of fulfillment centers across the US, and each fulfillment center is replenished based on its own on-hand and the regional demand it serves, so a single item typically generates a stream of smaller POs to several locations rather than one national order. POs typically arrive by EDI with routing and appointment requirements, and Chewy measures fill rate and on-time delivery with chargebacks for shortfalls. Chewy has also begun adding third-party marketplace selling alongside its first-party model; confirm which arrangement you are on, because the forecasting problem is different (marketplace means you hold the inventory and the demand risk).

What makes Chewy distinct from a brick-and-mortar pet retailer:

• Autoship is the engine. By Chewy's own reporting, roughly three-quarters of its net sales come through Autoship. For a consumable, that means most of your Chewy volume is predictable repeat demand from a known base of subscribers, and the forecast should be built from that base rather than from a weekly velocity guess.
• No doors, so no pipeline fill in the usual sense. The opening order is Chewy's initial stock across fulfillment centers, typically a few weeks of expected demand per location. It is smaller relative to run rate than a 1,000-store planogram fill, which makes the ramp, not the launch, the cash event.
• Demand is national and visible. Chewy shares sales and inventory data through its vendor portal, and because there is no store-level variability, the aggregate numbers are stable week to week outside of promotions.
• Returns are low for consumables, but damages happen. Shipping heavy bags and cans through parcel networks produces damages that are typically deducted; build an allowance.

Payment terms are negotiated and typically fall in the 30 to 60 day range, with deductions for shortages, damages and promotional funding netted against remittances. Confirm yours and model to the week.

Subscriber math instead of door math

For a consumable on Chewy, replace "units per store per week x doors" with:

Autoship weekly units = active Autoship subscribers / average weeks between shipments

Total Chewy weekly units = Autoship weekly units / Autoship share of your item's sales

Active subscribers next month = active subscribers this month + new subscribers - (active subscribers x monthly churn rate)

Chewy PO forecast = total weekly units, split across fulfillment centers by their share of demand, with each location reordering when it drops below a few weeks of cover

The subscriber base is the number to watch. New Autoship sign-ups come from Chewy's search placement, your own marketing and the Autoship discount on first orders; churn comes from pets switching, price sensitivity and product issues. A brand that tracks adds and churn monthly can forecast Chewy six months out with more confidence than almost any other retail account. For the broader method, our guide to SKU-level demand forecasting covers subscriber-driven channels alongside retail.

Promotions and how they move inventory

Chewy's recurring Autoship discount (a first-order discount plus a smaller ongoing one) is generally Chewy's pricing, but brands are often asked to fund promotional pricing, bundles or featured placement for specific windows, and holiday and seasonal events (pet birthdays, back to school for pets, Black Friday) bring real spikes. Because Chewy ships from its own stock, every promotion has a build PO ahead of it, sized to the expected lift across fulfillment centers. The specific inventory risk at Chewy is a first-order discount that pulls in subscribers who churn quickly: the build PO and the first month look great, then the base erodes and Chewy's fulfillment centers are over cover for weeks. Model promotional cohorts with a higher churn rate than organic ones.

A worked example: Fetch and Field treats

Take a fictional dog treat brand we will call Fetch and Field, launching one SKU on Chewy. Assumptions: retail $18.99, wholesale $11.00, landed cost $5.50, average Autoship frequency of 5 weeks, Autoship at 70% of the item's sales, 1,200 new Autoship subscribers a month, 6% monthly churn, six fulfillment centers each holding 4 weeks of cover, co-manufacturer lead time 6 weeks, terms 45 days, damages allowance 2% of gross.

• Subscriber ramp: month 1 ends with about 1,200 subscribers, month 3 with about 3,400, month 6 with about 6,200, and the base levels off near 20,000 subscribers if adds and churn hold (1,200 / 0.06).
• Weekly demand at month 6: 6,200 / 5 = 1,240 Autoship units, divided by 0.70 for about 1,770 total units a week, about $19,500 gross.
• Weekly demand at steady state: 20,000 / 5 = 4,000 Autoship units, 5,700 total units a week, about $62,700 gross, more than three times the month 6 run rate.
• Opening order: Chewy stocks roughly 4 weeks of month 1 demand (about 340 units a week) across six locations, about 1,400 units, a $15,400 PO. Small, which is the point: the launch is not the cash event.
• Production planning: with a 6-week lead time and demand tripling between month 6 and steady state, Fetch and Field must order production against the subscriber forecast, not against last month's POs. Ordering to trailing POs at month 6 means short-shipping Chewy by month 9, with the chargebacks and lost Autoship subscribers that follow. Ordering to steady state at month 3 means roughly $31,000 a week of inventory arriving for a business consuming about $5,300 a week.
• Cash: the working capital tied up at any point is roughly lead time plus terms, about 12 weeks of landed cost. At month 6 that is 12 x 1,770 x $5.50, about $117,000; at steady state about $376,000. Both numbers should be in the plan before launch, along with the month in which each becomes true.

For the PO-to-cash mechanics see purchase order forecasting. The closest analogue to Chewy among other accounts is Amazon's first-party program, covered in our guide to Amazon Vendor Central inventory forecasting, and for how Chewy compares with PetSmart, Petco and other pet accounts see the retail inventory forecasting hub.

Where Drivepoint fits

Drivepoint is the AI finance platform for consumer brands, and subscriber-driven retail accounts like Chewy are where a connected model earns its keep. It connects Chewy sales and PO data alongside Shopify, Amazon and your accounting system through 100+ integrations, forecasts from the subscriber base at the SKU level inside your Excel model, and carries production, POs and remittances through to the P&L, balance sheet and cash flow. AI scenario planning lets you test a higher churn rate, a funded promotion or a second SKU in minutes, not days, so production orders follow the forecast instead of trailing it.