Strategy & Planning

How to run a pricing analysis as CPG Founders

Strategy & PlanningFor CPG Founders3 apps11 steps~22 min to set up

You're setting shelf prices and negotiating co-packer contracts without a clear picture of your actual unit economics. Your COGS is split across a QuickBooks bill here, a Plaid transaction there, and a co-packer invoice in someone's email. You know roughly what a case costs to produce, but landed cost after freight, broker fees, and slotting allowances is a number you rebuild in a spreadsheet every quarter — and it's always out of date by the time you need it. Meanwhile your retail buyer wants to talk about promotional pricing on a call tomorrow and you're eyeballing margins that might be off by four points.

Build this on Starch → See the apps used

Strategy & PlanningFor CPG Founders3 apps11 steps~22 min to set up

Outcome

What you'll set up

A live pricing model that pulls actual COGS components from QuickBooks and Plaid and computes landed unit cost, gross margin by channel, and break-even volume — updated automatically, not rebuilt by hand each quarter

A scenario analysis workspace where you can test price changes, promotional depth (20% off for a 4-week feature), and volume assumptions side-by-side to see which scenarios blow your margin covenant with your lender

A weekly digest that flags when your spending on freight, ingredients, or co-packing drifts from baseline so your pricing model stays accurate without manual reconciliation

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

Scenario Analysis Transaction Insights Growth Analyst

Data sources & config

Starch syncs your QuickBooks data on a schedule (invoices, bills, vendor payments, journal entries) and syncs your Plaid bank and credit-card transactions on a schedule. The Scenario Analysis app uses both as the baseline for modeling. Transaction Insights runs on the same Plaid sync. Growth Analyst connects to PostHog from Starch's integration catalog for DTC channel data. No manual CSV exports required.

Prompts to copy

Build me a pricing analysis dashboard for my CPG brand. Pull COGS line items from QuickBooks — ingredients, co-packing fees, freight, packaging — and Plaid transactions to catch anything not yet in QuickBooks. Compute landed cost per case, gross margin by channel (DTC Shopify, Amazon FBA, wholesale), and break-even units at my current fixed overhead. Let me adjust price per unit, case volume, and promotional discount depth and show me how margin and runway change across those scenarios.

Show me all Plaid transactions in the last 90 days categorized as freight, co-packing, or ingredients. Flag any vendor charge that's more than 20% above its trailing 3-month average and send me a weekly email summary so I know when my input costs are drifting.

Every Monday, email me a pricing health digest: landed cost vs. last month, gross margin by channel vs. my 40% target, and any new vendor that charged my account in the last 7 days that I haven't categorized yet.

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

Walkthrough

Step-by-step

1 Connect QuickBooks as a scheduled-sync provider. Starch pulls bills, vendor payments, and journal entries so every co-packer invoice and ingredient PO is in the system without manual entry.

2 Connect your business checking and credit card accounts via Plaid as a scheduled-sync provider. This catches freight charges, one-off ingredient purchases, and co-packing deposits that hit the bank before they're fully categorized in QuickBooks.

3 Open the Scenario Analysis starter app and describe your cost structure: 'My base case is a 12-unit case of [SKU]. Landed cost components are ingredients ($X/case), co-packing ($Y/case), inbound freight ($Z/case), and outbound FBA prep ($W/case). My fixed overhead is $[monthly]. Build me a model that lets me test price per case, promotional discount %, and monthly volume.'

4 Set your channel split. Tell Starch: 'I sell 60% through Amazon FBA at a $18.00 MSRP, 25% through Shopify DTC at $22.00, and 15% through distributors at a $10.50 net price after distributor margin. Show me blended gross margin and per-channel margin separately.'

5 Run your first scenarios. Test at least three: current pricing, a 15% price increase on DTC only, and a 4-week Amazon promotion at 25% off. Scenario Analysis shows runway, burn rate, and break-even volume for each — side by side.

6 Layer in the Transaction Insights app. Tell Starch: 'Flag any Plaid transaction categorized as freight or co-packing that is more than 20% above the trailing 90-day average for that vendor.' This catches input cost inflation before it silently erodes the margin your pricing model assumed.

7 Set a weekly alert. Tell Starch: 'Every Friday at 8am, send me a digest showing (a) total ingredient and co-packing spend this week vs. the same week last month, (b) any new vendor that has charged my account this week, and (c) my current blended gross margin vs. my 40% target.' This is your early-warning system for cost drift.

8 Validate your Amazon channel margin. Tell Starch: 'Pull all Stripe payouts tagged to Amazon settlements and compare them to my FBA-channel units shipped this month. Back-calculate net revenue per unit and show me gross margin after FBA fees, which I'll enter manually as $X per unit.' Starch syncs your Stripe data on a schedule so this recalculates automatically each month.

9 Model a distributor deduction scenario. Tell Starch: 'Add a scenario where I receive 8% invalid deductions on wholesale channel revenue — show me how blended gross margin changes and at what deduction rate my wholesale channel goes contribution-negative.'

10 Export a one-page pricing summary for your retail buyer or board. Tell Starch: 'Generate a clean summary table showing current price per unit, landed cost per unit, gross margin %, break-even volume, and the financial impact of a 20% promotional price for 4 weeks — in a format I can paste into a slide deck or email.'

11 Schedule a monthly model refresh. Tell Starch: 'On the first of every month, re-pull QuickBooks bills and Plaid transactions from the prior month, recalculate my landed cost per case with the new actuals, and update the baseline in my pricing model so I'm always working from real numbers, not estimates I typed in six months ago.'

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 Pricing Review — 12-ct Granola Bar SKU

Sample numbers from a real run

Ingredients (oats, honey, chocolate chips)	4.2
Co-packing fee (per case)	3.8
Inbound freight (per case, averaged over 500-case run)	1.1
Outbound FBA prep + inbound to Amazon FC	0.95
Amazon FBA pick-and-pack fee (per case)	2.4
Allocated packaging (box, inserts)	0.65
Total landed cost per case	13.1
Amazon channel net revenue per case (at $18.00 MSRP minus 15% referral fee)	15.3
Amazon channel gross margin per case	2.2
Amazon channel gross margin %	14.4
DTC Shopify net revenue per case (at $22.00 minus Shopify + payment fees)	20.68
DTC gross margin per case	7.58
DTC gross margin %	36.6
Blended gross margin (60% Amazon / 25% DTC / 15% wholesale at $10.50 net)	22.1

When the Starch model ran in early February, the founder assumed blended gross margin was around 30%. The actual number came back at 22.1% — because freight had climbed 18% since the last manual estimate (flagged by Transaction Insights when a $2,100 freight charge appeared vs. the $1,780 trailing average), and the Amazon FBA fee increase from January wasn't reflected in the old spreadsheet. The scenario comparison showed that a $1.50 price increase on the Shopify DTC channel — from $22.00 to $23.50 — would push blended gross margin to 27.4% with essentially no modeled volume impact, because DTC buyers are less price-sensitive than Amazon browsers. The distributor deduction scenario showed the wholesale channel goes contribution-negative if deductions exceed 11% of channel revenue — a number the founder now brings to every distributor conversation. Total time to build the model, connect QuickBooks and Plaid, and run all three scenarios: about two hours, versus the founder's prior estimate of a half-day in Excel.

Measurement

How you'll know it's working

Landed cost per case by SKU (ingredients + co-packing + freight + packaging, updated monthly from actuals)

Gross margin % by channel (Amazon FBA vs. Shopify DTC vs. wholesale/distributor net)

Break-even case volume at current fixed overhead — and how many months of inventory you need to sell to hit it

Input cost drift vs. trailing 90-day average by vendor category (freight, ingredients, co-packing) — the early warning that your margin assumptions are going stale

Promotional margin impact: gross margin change per point of promotional discount depth, so you know the real cost of a 20% Amazon coupon before you run it

Comparison

What this replaces

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

Excel or Google Sheets pricing model

Free and infinitely flexible, but the model is only as current as the last time you manually updated the cost inputs — which for most CPG founders is quarterly at best, right before a board meeting or buyer negotiation.

QuickBooks P&L reports

QuickBooks shows you historical COGS in aggregate but doesn't compute per-unit landed cost by channel, doesn't let you run price scenarios, and requires you to pull and interpret the reports yourself; note that QuickBooks report views (P&L, Transaction List) are temporarily unavailable in Starch pending a connector fix, though entity-level data like bills, invoices, and vendor payments syncs normally.

Cin7 or Katana (inventory + COGS)

Purpose-built for CPG inventory costing and gives you strong lot-level COGS tracking, but neither tool includes financial scenario modeling or connects your cost data to a pricing decision framework without significant manual export work.

Mosaic or Runway (financial modeling SaaS)

Well-designed for startup finance teams doing headcount and burn modeling, but not built for CPG-specific cost structures like co-packing runs, freight averages, or channel-specific margin splits; also priced for teams with a finance hire, not a solo operator founder.

On Starch RECOMMENDED

One platform — scenario planning, transaction insights, growth analyst 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 pull my actual co-packing invoices and ingredient bills, or do I have to enter costs manually?

If your co-packer bills go through QuickBooks, Starch syncs those bills automatically on a schedule — line items, vendor, amounts, and payment status. Same for ingredient POs billed through QuickBooks. If some charges hit your bank directly (a deposit to a small co-packer, for example), Starch syncs those from Plaid and you can categorize them once. After that, the model uses real invoice data, not numbers you typed in.

My COGS changes every production run because ingredient costs and freight rates move around. Will the model go stale?

That's exactly what Transaction Insights is watching for. You set a threshold — say, any freight or ingredient charge more than 15% above the trailing 90-day average for that vendor — and Starch flags it automatically and emails you. You can also schedule a monthly model refresh that re-pulls the last month's actuals from QuickBooks and Plaid and updates the baseline. You won't catch every change the day it happens, but you won't be running a six-month-old cost assumption into a buyer meeting either.

Can I model Amazon promotional pricing, including the FBA fee structure?

Yes. You tell Starch your FBA fee per unit or per case (Amazon publishes these; you enter them as a fixed input), your MSRP, and your promotional discount depth. The scenario model computes net revenue after referral fee and FBA fee, subtracts your landed cost, and shows you gross margin at each promotional price. You can compare a 15% coupon vs. a 25% coupon vs. no promotion side-by-side. What Starch won't do automatically is pull your FBA fee schedule from Seller Central — you'd enter that number, or Starch can read it from your Amazon Seller Dashboard app, which connects via browser automation.

Does Starch store my QuickBooks and bank data, or does it query it fresh each time?

Both QuickBooks and Plaid are scheduled-sync providers — Starch syncs the data on a regular schedule and stores it in Starch's database. That means your pricing model runs against real data without waiting for a live API call every time you open it. The tradeoff: Starch is designed for live operational surfaces, not long-horizon archiving. If you need a seven-year historical cost database, a data warehouse is the right tool. For a rolling 12-month pricing model, the scheduled sync is exactly right.

I use Shopify for DTC and sell through a couple of wholesale marketplaces. Can Starch pull revenue by channel?

Shopify is reachable from Starch's integration catalog and the agent queries it live when your app runs. For wholesale marketplace revenue, it depends on the platform — if it's a web-based portal you log into, Starch can automate it through your browser without needing an API. You'd tell Starch: 'Log into my KeHE Connect portal, pull last month's settlement report, and add that net revenue to my wholesale channel total.' That's a first-class Starch pattern.

Is Starch SOC 2 certified? I'm nervous about connecting my bank accounts and accounting data.

Starch is not SOC 2 Type II certified yet — that's worth knowing before you connect sensitive financial data. Plaid and Stripe, the data bridges, have their own security certifications. If your company policy requires SOC 2 Type II from every vendor in your stack, Starch isn't there yet and we'd rather tell you that clearly than oversell it.