Ops & Supply

How to set reorder points and safety stock as CPG Founders

Ops & SupplyFor CPG Founders2 apps12 steps~24 min to set up

You're running reorder points off a spreadsheet that was 'good enough' when you had two SKUs and one 3PL. Now you've got 12 SKUs, a co-packer in New Jersey with a 6-week lead time, FBA inventory, a handful of wholesale accounts, and a DTC Shopify store — and your 'safety stock' is whatever number felt comfortable last quarter. You've stocked out on your hero SKU during a Whole Foods promo window and eaten the lost POs. You've also over-produced on a seasonal SKU and watched $40K of product expire at your 3PL. The math isn't that hard. The problem is assembling it across five data sources fast enough to matter.

Build this on Starch → See the apps used

Ops & SupplyFor CPG Founders2 apps12 steps~24 min to set up

Outcome

What you'll set up

A per-SKU reorder point that automatically factors in your co-packer lead time, average daily velocity by channel, and any upcoming promotions — so you know exactly when to place a production run before you're out of stock

Safety stock targets that account for demand variability and lead time variability, not just a flat 'two weeks of sales' buffer that made sense once and has been copy-pasted ever since

Automated reorder alerts that fire when on-hand inventory (across your 3PL, co-packer, and FBA) crosses the threshold — so you're not refreshing a spreadsheet every Monday hoping someone updated the numbers

The Starch recipe

Apps, data, and prompts

The combination of Starch apps, the data sources they pull from, and the prompts you use to drive them.

Apps used

Inventory Planner Demand Planner

Data sources & config

Inventory Planner and Demand Planner connect to Shopify from Starch's integration catalog (the agent queries it live when your inventory view runs), and to Amazon via browser automation — no Amazon Seller Central API required. Your 3PL and co-packer inventory feeds can be pulled in via Google Sheets connected from Starch's integration catalog, or Starch can automate pulling reports through your browser if your 3PL has a portal but no API. Starch syncs your Stripe data on a schedule to cross-reference revenue by SKU.

Prompts to copy

Set reorder points for all active SKUs based on co-packer lead time, current sell-through rate by channel, and 30% safety stock buffer. Alert me when any SKU crosses its reorder point.

Show me which SKUs are at risk of stocking out in the next 45 days given current velocity and co-packer lead times. Flag anything where FBA and DTC inventory combined is below safety stock.

Calculate the right safety stock for each SKU using actual demand variability from the last 90 days of Shopify and Amazon sales, and build in a 2-week buffer for co-packer lead time variance.

Run these in Starch → or paste them into your favorite agent

Walkthrough

Step-by-step

1 Connect your Shopify store from Starch's integration catalog so the agent can pull daily unit sales by SKU, broken out by channel (DTC, wholesale, marketplace).

2 Connect your 3PL inventory feed — if your 3PL has a portal, Starch automates pulling the inventory export through your browser with no API needed. If they send CSV reports, connect Google Sheets from Starch's integration catalog.

3 Wire in your FBA inventory data. Starch automates your Amazon Seller Central dashboard through your browser so you can see FBA on-hand, reserved, and inbound units alongside your 3PL stock.

4 Enter your co-packer lead time per SKU — Inventory Planner stores this and uses it to back-calculate how far in advance you need to trigger a production run.

5 Pull the last 90 days of sales by SKU and channel into Demand Planner. Ask it to calculate average daily velocity and standard deviation so you're working off real demand variability, not gut feel.

6 Tell Demand Planner about any upcoming promos, retail resets, or seasonal windows ('Whole Foods reset is in 6 weeks, expect 3x velocity on SKU-004 for that month'). It builds the lift into the reorder calculation.

7 Have Inventory Planner calculate the reorder point for each SKU: reorder point = (average daily units × lead time in days) + safety stock. Safety stock = Z-score × standard deviation of demand × square root of lead time.

8 Review the per-SKU reorder points and adjust any that look off — for example, if a new product doesn't have 90 days of history yet, you can manually set a conservative baseline and let the system learn over time.

9 Set the alert threshold: Starch notifies you (via Slack, which it connects to on a schedule, or via email through Gmail) when any SKU's combined inventory across 3PL and FBA drops below its reorder point.

10 Run a 45-day forward projection once a month — Demand Planner shows you which SKUs are going to hit their reorder point before the next scheduled production run so you can decide whether to pull the run forward.

11 Track actuals against the model each week. When a SKU's real sell-through deviates from the forecast by more than 20%, Starch flags it so you can manually review whether to adjust the reorder point or safety stock buffer.

12 At the end of each quarter, review which SKUs triggered reorder alerts on time versus which ones still stocked out or over-produced — and use that to calibrate safety stock percentages for the next quarter.

See this running on Starch

Connect your tools, describe what you want, and the agent builds it. Closed beta is free.

Try it on Starch →

Worked example

Q1 2026 Reorder Planning — 3-SKU Scenario

Sample numbers from a real run

SKU-001 (Hero Bar, 12-count case) — reorder point	1,800
SKU-001 — safety stock at 30% buffer	540
SKU-003 (Seasonal Flavor, limited run) — reorder point	600
SKU-003 — safety stock accounting for 2x promo lift	360
SKU-007 (New Launch, 60 days history) — conservative reorder point	400

You make three SKUs and your co-packer runs on a 5-week lead time. SKU-001, your hero bar, moves about 120 cases a week across DTC and Whole Foods. At a 5-week lead time (35 days × 120/7 = 600 cases) plus 30% safety stock (180 cases), your reorder point is 780 cases on hand. Demand Planner finds that Q1 Whole Foods resets historically push velocity to 180 cases/week for six weeks — it automatically bumps the reorder point to 1,080 cases for that window and tells you to place the production run by January 12 to have product on shelf for the February reset. SKU-003 is a seasonal flavor you're worried about over-producing: Demand Planner's 90-day variance shows high standard deviation, so Starch recommends a tighter safety stock (two weeks instead of three) and smaller production runs with a confirmed reorder trigger rather than a large upfront bet. SKU-007 launched in November and has only 60 days of sales history, so Starch flags the reorder point as 'low confidence' and applies a manual conservative buffer you set yourself — it'll update automatically as more data accumulates. Total: you've got clear go/no-go numbers for three production runs before end of January instead of a spreadsheet tab that someone last updated in October.

Measurement

How you'll know it's working

Stockout rate by SKU — percentage of weeks where on-hand inventory hits zero before a production run arrives

Days of inventory on hand by channel (FBA vs. 3PL vs. co-packer in-process) at the end of each month

Overproduction write-offs — dollar value of expired or unsellable inventory per quarter

Reorder point accuracy — how often an automated alert fires with enough lead time to place a production run before hitting zero

Safety stock utilization — how often safety stock is actually consumed (if never consumed, you're holding too much; if always consumed, you're under-protected)

Comparison

What this replaces

The other ways teams handle this today, and how the Starch version compares.

Cin7 or Fishbowl

Purpose-built inventory management with solid reorder features, but starts at $700–1,000/month, assumes a warehouse team doing physical counts, and doesn't connect natively to co-packer portals without custom integrations you have to build yourself.

Inventory Planner (standalone SaaS)

Good at Shopify and Amazon replenishment forecasting but priced per-connection, doesn't know about your co-packer lead times unless you manually enter them every time, and lives entirely outside your financial and demand data.

Google Sheets with VLOOKUP formulas

Free and infinitely flexible, but someone has to update it manually after every 3PL report and Amazon inventory pull — which means it's always two days out of date when you actually need it.

NetSuite with demand planning module

Genuinely excellent for companies with 30+ SKUs and a supply chain team, but costs six figures to implement, takes six months to onboard, and is overkill for a founder still doing their own co-packer calls.

On Starch RECOMMENDED

One platform — inventory planner, demand planner all running on connected data. Setup in plain English; numbers stay current via scheduled syncs and live agent queries.

Try it on Starch →

FAQ

Frequently asked questions

My 3PL doesn't have an API. Can Starch still pull my inventory data?

Yes. If your 3PL has a web portal, Starch automates pulling inventory reports through your browser — no API needed. If they email you CSV reports, you can pipe those into Google Sheets and Starch queries Google Sheets live from its integration catalog. The WeWork-style pattern of 'API data in, browser automation out' is how Starch handles most co-packer and 3PL integrations.

Does Starch store my historical inventory data, or does it query live every time?

Starch syncs data from scheduled-sync providers (like Stripe and Google Sheets) on a schedule and stores it. For live-query apps like Shopify, data is queried fresh when your app runs rather than archived. If you need deep historical trend analysis across years of data, Starch is built for live operational surfaces, not a long-horizon data warehouse — that's an honest limit worth knowing before you sign up.

The Demand Planner and Inventory Planner apps are listed as in development. What can I actually use today?

Both apps are currently in beta and you can request access on the Starch site. In the meantime, you can describe exactly what you need to Starch in plain language — 'build me a dashboard that shows reorder points by SKU based on Shopify velocity and a lead time I set per SKU' — and the agent builds a custom app from scratch. The pre-built apps are a starting point; they're not the only way to get this workflow running.

Can Starch factor in a planned promotional lift when calculating reorder points?

Yes. You tell the Demand Planner about an upcoming promo window and expected velocity increase ('Whole Foods endcap in March, expect 2.5x velocity for 4 weeks') and it builds that into the reorder point calculation for that period. You can also trigger a one-time manual recalculation any time you get a large unexpected wholesale order.

Is Starch SOC 2 certified? My co-packer and retail buyers sometimes ask about data security.

Not yet — Starch is not SOC 2 Type II certified. If your retail buyer or co-packer has a formal vendor security review requiring SOC 2, that's a real constraint to name upfront. It's on the roadmap, but we won't claim it before it's done.

What if I have shelf-life constraints that affect how I think about safety stock?

Shelf-life management is built into Inventory Planner — it tracks lot-level expiration dates and applies FEFO (first-expired-first-out) rotation logic. When you're calculating how much safety stock to hold, you can set a maximum days-on-hand cap so you don't build a buffer so large that product expires before it ships. This is especially relevant for anything with a 12-month or shorter shelf life going into FBA, where Amazon's stranded inventory rules can compound the problem.