Finance & FP&A

How to model financial scenarios and sensitivities as CPG Founders

Finance & FP&AFor CPG Founders3 apps12 steps~24 min to set up

You're running financial scenarios in Google Sheets, which means every time your co-packer raises tolling fees, your broker commission structure changes, or a retail buyer wants 90-day payment terms instead of 30, you're rebuilding formulas from scratch. Your baseline numbers are already stale because they came from a QuickBooks export you did two weeks ago. You've got one tab for 'optimistic,' one for 'base,' one for 'holy shit,' and none of them agree with your bank account. A finance hire would fix this, but you don't have one — and you need to know today whether to pull the trigger on that $180K production run.

Build this on Starch → See the apps used
Finance & FP&AFor CPG Founders3 apps12 steps~24 min to set up
Outcome

What you'll set up

A live baseline scenario wired to your actual Stripe revenue and Plaid bank transactions — so when your burn changes, the model updates without a manual export
Side-by-side scenario comparisons showing runway, net burn, and break-even under different tolling cost, trade spend, and revenue growth assumptions
A quarterly budget layer that tracks actual co-packer, broker, and freight spend against plan, with variance flags the moment a category goes off-track
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 Runway Analysis Budgeting
Data sources & config

Starch syncs your Stripe data on a schedule (charges, payouts, subscription revenue) and your Plaid bank transaction data on a schedule (categorized expenses, balances). These two live feeds form the baseline for every scenario and the actuals layer for your budget. No manual exports — when your bank account updates, the model updates. If you run payroll through Gusto, Rippling, or ADP, you can connect those from Starch's integration catalog for precise labor cost inputs; Starch syncs ADP on a schedule and queries Gusto and Rippling live when your apps need the data.

Prompts to copy
Build me a scenario model with three cases: base (revenue grows 8% MoM, tolling costs stay flat), optimistic (revenue grows 15% MoM, we land the Sprouts account in Q3), and downside (revenue flat, co-packer raises fees 12% in July). Pull the baseline burn from Plaid and revenue from Stripe. Show me runway and break-even month for each case.
Show me my current net burn and 24-month cash projection using my Plaid transactions and Stripe payouts. Break expenses into co-packer/manufacturing, freight and 3PL, broker commissions, trade spend, and overhead.
Set up a quarterly budget for Q3 2026. Use my last two quarters of Plaid spending to auto-suggest category allocations. I want to track co-packer costs, freight, broker fees, trade spend, payroll, and marketing separately. Flag any category that goes more than 10% over pace.
Run these in Starch → or paste them into your favorite agent
Walkthrough

Step-by-step

1 Connect Plaid to Starch — Starch syncs your bank transactions on a schedule, categorized by vendor. Your co-packer wires, freight invoices, and broker payments show up as line items automatically.
2 Connect Stripe — Starch syncs your Stripe charges, payouts, and any subscription revenue on a schedule, giving every scenario model a real revenue baseline instead of a number you typed in last month.
3 Open the Runway Analysis app and verify that your trailing 6-month burn and revenue figures look right. Check that your co-packer payments and 3PL charges are categorized correctly — relabel any miscategorized Plaid transactions before you build scenarios on top of them.
4 Open the Scenario Analysis app and type your first prompt. Define at least three cases: base, upside (Sprouts or UNFI deal closes, velocity increases), and downside (co-packer cost increase, a retail buyer deductions dispute delays cash by 60 days).
5 Adjust the tolling cost assumption in your downside scenario to reflect your co-packer's actual contract renewal terms. If fees go up 10-15% annually, model it — most CPG founders underestimate manufacturing cost creep.
6 Add a trade spend variable. Promotional allowances, slotting fees, and MCB deductions are often 15-25% of gross revenue for a brand in retail. Model what happens to runway if your trade spend rate increases by 5 points as you push into new doors.
7 Add a receivables lag assumption for any scenario involving wholesale or distributor revenue. Net-60 or Net-90 terms from UNFI or KeHE meaningfully change your cash timing — model the gap between recognized revenue and cash in hand.
8 Open the Budgeting app and set up your Q3 category budget. Let Starch auto-suggest allocations from your Plaid history, then adjust co-packer and freight line items to match your contracted rates for the quarter.
9 Set a pace alert for trade spend — this is the category CPG founders most reliably blow past. Flag anything over 10% of plan variance so you catch overspend before a quarterly close.
10 After your monthly close, re-run your scenarios using the updated Plaid and Stripe actuals. Check whether your real burn is tracking closer to base or downside — most brands drift toward downside in Q1 due to post-holiday velocity dips.
11 Before your next investor update or board call, export your scenario comparison showing runway under each case. Walk investors through the specific levers you control (pricing, production volume, trade spend) versus the ones you don't (velocity, macro).
12 Revisit and rebuild scenarios any time a major assumption changes — new co-packer contract, a lost retail account, or a fundraise closing at a different amount than planned. Because the baseline auto-updates from Plaid and Stripe, you're only changing the delta, not rebuilding from scratch.

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

Pre-UNFI pitch scenario model — September 2026

Sample numbers from a real run
Monthly Stripe revenue (base)48,000
Monthly Stripe revenue (UNFI upside)74,000
Co-packer tolling cost (monthly)21,000
3PL and freight (monthly)6,500
Broker commissions at 5% of gross2,400
Trade spend / promotional allowances9,600
Payroll (founder + 2 FTEs)18,500
Overhead and misc3,800
Current cash balance310,000

You're heading into a pitch with a regional UNFI buyer and need to show investors what the deal actually does to your financials. In your base scenario — $48K/month Stripe DTC revenue, no new wholesale — your net burn is roughly $13,800/month and your runway is 22 months on $310K cash. That's a fine business, but not the growth story investors want. You build an upside scenario in Starch: UNFI adds $26K/month in incremental revenue starting in November, but you also model the cash timing correctly — UNFI pays on Net-60, so the first two months of that revenue don't hit your bank until January. You also add 8 points of trade spend (slotting, MCB accruals, promotional allowances) on the UNFI volume, which adds $2,080/month in effective cost. The upside scenario still improves runway to 29 months, but only if your co-packer can hold tolling rates through Q2 2027. You add a third scenario: UNFI deal closes, co-packer raises fees 12% in February. Runway compresses to 24 months. Now you know exactly what to negotiate with your co-packer before you sign the UNFI contract — and you can show investors the specific levers, not just a hockey stick.

Measurement

How you'll know it's working

Net burn rate by month, broken out by manufacturing vs. trade spend vs. overhead — not blended
Runway in months under each scenario (base, upside, downside), updated monthly against actual Plaid + Stripe data
Trade spend as a percentage of gross revenue — target band vs. actual, by quarter
Cash conversion lag — days between booking wholesale revenue and receiving payment (UNFI / KeHE Net-60/90 reality)
Gross margin by channel (DTC Shopify vs. Amazon FBA vs. distributor wholesale) to stress-test which revenue mix actually improves burn
Comparison

What this replaces

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

Google Sheets with manual QuickBooks exports
Free and flexible, but your baseline is always stale — you're modeling against numbers from two weeks ago, and every co-packer fee change means a manual rebuild.
Mosaic or Runway (dedicated FP&A SaaS)
Purpose-built for scenario modeling with cleaner UI, but priced for Series A+ companies with a finance hire — overkill and overpriced for a small CPG team that also needs CRM, inventory, and ops tooling in the same platform.
QuickBooks + spreadsheet hybrid
QuickBooks gives you accurate actuals but no scenario modeling — you still end up exporting to Excel and building the 'what if' layer manually, which breaks every time an assumption changes.
Notion or Coda finance templates
Good for simple burn trackers but not built for multi-variable scenario comparison — you can't wire live Stripe and Plaid feeds without custom integrations, so it turns into another manual-update doc.
On Starch RECOMMENDED

One platform — scenario planning, runway analysis, quarterly budgeting 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 books aren't fully clean in QuickBooks yet — can I still build useful scenarios?
Yes. Starch connects your Plaid bank transactions and Stripe payouts directly, so the baseline cash model doesn't depend on your books being closed. Your Plaid feed shows actual cash out the door — co-packer wires, freight, payroll — categorized at the transaction level. It won't match your P&L exactly, but it gives you a cash-basis burn rate that's accurate enough for scenario modeling. Clean up QuickBooks in parallel; once it's current, you can add QuickBooks as a connection and layer in accrual-basis data too. Starch syncs QuickBooks on a schedule — invoices, bills, vendors, and payments — though the report views (P&L, Transaction List) are temporarily unavailable while an upstream fix is in progress.
Can Starch model gross margin by channel, not just total burn?
Yes — tell Starch what you want to see. You'd say something like: 'Build me a margin breakdown by channel: DTC Shopify, Amazon FBA, and distributor wholesale. Pull revenue from Stripe for DTC, pull Amazon Seller data for FBA, and let me manually input distributor net revenue. For COGS, use my co-packer invoices from Plaid transactions tagged to that vendor.' Starch builds the view from your description. Shopify can be connected from Starch's integration catalog and queried live; Amazon Seller data is available through the Amazon Seller Dashboard app.
How does Starch handle the trade spend and deductions that distributors take? That's a big source of variance for us.
Deductions show up in your Plaid feed as reduced remittances — a check from UNFI for less than your invoice amount. You can tag those transactions in your budget as 'trade deductions' and Starch will track them as a spend category. For modeling purposes, you'd input your expected deduction rate as an assumption in the scenario (e.g., 'assume 18% effective trade spend rate on all UNFI volume'). Starch won't automatically dispute the deductions — that's a manual process — but it will make the cash impact of your deduction rate visible in every scenario, which is usually the first step to understanding whether your distributor channel is actually profitable at current velocity.
Is my financial data secure? I'm connecting my bank account through Plaid.
Starch uses Plaid's standard bank connection flow — the same one used by most personal finance and accounting apps. One honest thing to know: Starch is not yet SOC 2 Type II certified. If that's a hard requirement for your business or your investors, it's worth knowing upfront. For most early-stage CPG founders, Plaid connectivity at this level of certification is standard practice.
Can I share scenarios with my investors or co-founder without giving them full Starch access?
You can export scenario outputs and share them as reports. Starch doesn't currently have a public share link or guest-view mode, so you'd be exporting a snapshot rather than giving someone a live link to your model. For an investor update, this usually works fine — you export the scenario comparison, walk them through it on a call. If you need a live collaborative model your co-founder can edit simultaneously, that's closer to what a dedicated spreadsheet tool does today.
My revenue is seasonal — we spike at holiday and slow in Q1. Will the baseline model handle that correctly?
The Runway Analysis baseline uses your trailing 6-month revenue trend from Stripe, which means a strong holiday quarter will inflate your forward projection if you're modeling in January. You'd want to tell Starch to adjust for seasonality: 'Model Q1 revenue at 60% of Q4 Stripe revenue, then ramp back to Q4 levels by Q3.' Starch builds the scenario with that override. The baseline is a starting point — you control the assumptions for anything you know is going to deviate from trend.

Ready to run model financial scenarios and sensitivities on Starch?

Request closed-beta access. Everything is free during beta.

You're on the list! We'll be in touch soon.