Ops & Supply

How to forecast product demand as DTC Brand Founders

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

Your demand forecast is a Shopify export pasted into a Google Sheet, adjusted by whoever remembers that you sold out of your best SKU two Novembers in a row. Meta ad spend spikes drive pull-through your 3PL hasn't planned for, and your co-packer needs 8-week lead times while you're still guessing whether the Q4 promo is going to hit. You find out you're understocked when a customer emails asking where their order is. You find out you're overstocked when you're staring at 4,000 units of a flavor that didn't move and your bank balance reflects it. No tool in your current stack — Shopify, Meta Ads, Klaviyo, a 3PL portal — talks to the others, so the forecast is always someone's best guess, never the whole picture.

Build this on Starch → See the apps used

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

Outcome

What you'll set up

A demand plan built from your actual Shopify order velocity, seasonal patterns, and promotional calendar — not last month's numbers plus a gut adjustment

Reorder triggers that account for your co-packer lead times and safety stock thresholds, so you stop making the call too late

A single dashboard showing inventory across your 3PL and warehouse locations, with shelf-life rotation and per-channel allocation, so you never oversell a SKU that's already committed elsewhere

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

Demand Planner Inventory Planner

Data sources & config

Connect Shopify from Starch's integration catalog — the agent queries your orders, products, and inventory live when the demand plan runs. Starch syncs your bank transactions on a schedule via Plaid so cash impact of reorder decisions shows up in the same view. Your Meta Ads data connects from Starch's integration catalog so promotional spend drives the lift model rather than a manual input. 3PL portals without a direct API are automated through your browser — no API needed.

Prompts to copy

Build me a demand forecast using my Shopify order history for the last 18 months. Factor in a 20% promotional lift for our Black Friday campaign and a Q4 seasonal spike. Show reorder dates by SKU assuming a 7-week co-packer lead time and 3 weeks of safety stock.

Give me a live inventory view across our 3PL in New Jersey and our Chicago warehouse. Flag any SKU with fewer than 4 weeks of cover based on current velocity. Show me anything expiring in the next 90 days and surface which channel is allocated to each lot.

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

Walkthrough

Step-by-step

1 Connect Shopify from Starch's integration catalog. The agent pulls your full order history — quantity by SKU, order date, channel, and discount code — so the forecast has real velocity data underneath it, not a spreadsheet you updated last quarter.

2 Connect Meta Ads from Starch's integration catalog so your promotional spend calendar feeds into the lift model automatically. When you schedule a campaign, the demand plan sees it and adjusts forward projections without you needing to input it manually.

3 Connect Klaviyo from Starch's integration catalog so email campaign send dates and open rates contribute to expected pull-through. A 40% open-rate blast to 80k subscribers hits your 3PL differently than a 12% open-rate to a cold segment.

4 Starch syncs your bank transactions on a schedule via Plaid. This grounds the reorder recommendation in what you can actually spend — the demand plan knows your cash position isn't infinite.

5 If your 3PL or co-packer has a web portal but no direct API, tell Starch to automate it through your browser. Starch pulls current stock levels and open PO statuses directly from the portal — no API needed, no manual export.

6 Use the Demand Planner (currently in beta — request access) to generate a rolling 12-week forecast by SKU. Tell Starch your co-packer lead times, minimum order quantities, and any known promotional windows. It builds the reorder calendar from those constraints, not from defaults.

7 Use the Inventory Planner (currently in beta — request access) to set channel allocations across DTC, wholesale, and retail. If you've committed 500 units to a Target replenishment order, those units get reserved and your Shopify storefront doesn't oversell against them.

8 Set shelf-life parameters for any perishable SKU. The Inventory Planner applies FEFO rotation logic and surfaces any lot approaching its best-by date so you can run a promo or redirect it before it's a write-off.

9 Tell Starch: 'Every Monday morning, pull last week's Shopify sell-through by SKU and compare it against the forecast. Flag any SKU where actuals are more than 15% above or below plan and Slack me a summary.' This runs automatically each week — you don't need to pull it.

10 Before a buyer meeting or board update, ask Starch to generate a demand summary: actual vs. forecast by SKU, current inventory cover by location, upcoming reorder dates, and cash outlay required. It pulls from the live data across all connected sources and builds the summary for you.

11 When a promo is confirmed, update the promotional calendar in Starch and ask it to re-run the forward forecast. The new reorder dates, safety stock requirements, and cash impact recalculate automatically against your co-packer lead times.

12 Review the reorder recommendations weekly. When you're ready to place a PO, Starch can draft the purchase order details based on the quantity and timing the demand plan recommends — you confirm, and it either sends or gives you the formatted output to send to your co-packer directly.

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

Q4 2025 holiday restock for a 6-SKU snack brand doing $4.2M ARR

Sample numbers from a real run

Shopify trailing-90-day velocity (hero SKU)	3,200
Units currently at NJ 3PL (hero SKU)	1,100
Units on open PO (arriving week 8)	2,000
Projected demand Nov 15–Dec 31 with Black Friday lift (20%)	5,400
Cover gap requiring emergency reorder	2,300
Co-packer MOQ	2,500
Cash required to close gap at $1.80/unit landed cost	4,500

In early October, Starch pulls the trailing 90-day Shopify velocity for the brand's top-selling jalapeño SKU: 3,200 units sold in the last quarter, accelerating. There are 1,100 units at the NJ 3PL and a 2,000-unit PO arriving in week 8 (late October). The demand plan flags that a 20% Black Friday promotional lift — pulled automatically from last year's Meta Ads spend pattern and the confirmed campaign budget in Meta Ads Manager — puts projected November 15–December 31 demand at 5,400 units. Available supply through the period: 3,100. Cover gap: 2,300 units. The co-packer's MOQ is 2,500, and they need 7 weeks. The reorder deadline to hit shelf by November 15 was October 1 — three days out. Starch surfaces the alert in Slack that morning with the math attached. The founder places the PO the same day. Cash required: $4,500 at $1.80/unit landed. Plaid confirms the operating account has $38k in it. The reorder is feasible and the brand doesn't stock out on its highest-velocity SKU the week that matters most.

Measurement

How you'll know it's working

Forecast accuracy by SKU (actuals vs. plan, % variance week-over-week)

Weeks of inventory cover by location and channel

Stockout rate (SKUs with zero available units / total active SKUs in a period)

Reorder lead time compliance (POs placed on time vs. reorder deadline)

Cash tied up in slow-moving or near-expiry inventory

Comparison

What this replaces

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

Google Sheets demand model

Free and infinitely flexible, but your Shopify data doesn't update automatically, the promo lift is whatever you type in, and the model breaks every time someone changes a formula — usually right before you need it.

Inventory Planner (standalone SaaS)

Purpose-built for Shopify reorder planning and genuinely good at it, but it doesn't know about your Meta ad spend, your Plaid cash balance, or anything else outside its own data — so the reorder recommendation still ignores half of what actually drives your decisions.

Cin7 or Brightpearl

Enterprise inventory management with deep feature sets, but implementation takes months, pricing scales steeply, and they're built for retailers with warehouse staff — not a 4-person DTC brand where the founder is also doing the reorder.

Shopify's built-in analytics

Shows you what sold, but gives you no forward forecast, no reorder logic, no lead-time awareness, and no view into what's happening at your 3PL — it stops at the Shopify boundary.

On Starch RECOMMENDED

One platform — demand planner, inventory 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

Can Starch actually connect to Shopify for order data, or does it just pull from what I export?

Starch connects to Shopify from its integration catalog and queries your orders, products, and inventory levels live when your demand plan runs. You don't export anything manually. The agent reads directly from your store data.

The Demand Planner and Inventory Planner apps say 'in development' — what can I use right now?

Both apps are in beta and you can request early access to get notified when they launch. While you're waiting, you can describe what you want to Starch in natural language and it will build you a custom demand forecasting app or inventory dashboard from scratch using your Shopify, Plaid, and ad platform connections — the App Store templates are starting points, not the ceiling.

My 3PL uses a web portal with no API. Can Starch still pull stock levels from it?

Yes. Starch automates your 3PL portal through your browser — no API needed. It logs in, navigates to your inventory view, and pulls current stock levels on whatever schedule you set. If the portal is web-accessible, it's reachable.

Can Starch factor in my co-packer lead times and minimum order quantities?

Yes. When you describe your demand plan to Starch, you tell it your lead time (e.g., 7 weeks) and MOQ (e.g., 2,500 units). The demand plan builds the reorder calendar backward from those constraints, so the 'order by' date reflects the actual deadline to hit your target in-stock date — not a generic buffer.

Is Starch SOC 2 certified? My 3PL asks about security before we share portal credentials.

Starch is not SOC 2 Type II certified today. That's worth knowing upfront if your partners have formal security review requirements. It's on the roadmap.

I have historical Shopify data going back 3 years. Does Starch use all of it for the forecast?

Starch queries Shopify live from its integration catalog rather than storing a persistent historical archive in Starch's database. For most demand planning use cases — rolling seasonal patterns, velocity trends, promo lift modeling — this is fine. If you need a multi-year data warehouse with point-in-time snapshots for statistical modeling, that's a different tool category and worth naming honestly.

Can Starch send me an automatic weekly reorder alert instead of me having to log in and check?

Yes. Tell Starch: 'Every Monday, check my Shopify inventory levels against my forecast. If any SKU has fewer than 3 weeks of cover, Slack me the SKU name, current stock, velocity, and the reorder date.' That automation runs on a schedule with no manual trigger from you.