Sales & CRM

How to forecast quarterly revenue as CPG Founders

Sales & CRMFor CPG Founders3 apps10 steps~20 min to set up

You're running quarterly revenue forecasts in a spreadsheet that pulls from four different places: your Shopify dashboard, a distributor portal that updates twice a week, a QuickBooks export you have to manually trigger, and a Stripe dashboard you check separately for DTC. Every time a buyer at a regional chain asks for a 90-day projection to support a new door authorization, you spend a day stitching numbers together — only to realize your co-packer lead times have shifted and the whole production-capacity assumption is wrong. There's no single place where your DTC revenue, wholesale purchase orders, and broker commissions live together. So your forecast is always a snapshot of the past dressed up as a prediction.

Build this on Starch → See the apps used

Sales & CRMFor CPG Founders3 apps10 steps~20 min to set up

Outcome

What you'll set up

A live revenue forecast that combines your Stripe DTC transactions, Plaid bank deposits from distributors and co-packer payments, and QuickBooks invoice data into a single view updated on a schedule — no manual exports.

A scenario model that lets you stress-test Q3 assumptions side-by-side: what if the Whole Foods regional expansion adds 80 doors in July vs. October, and how does that change your production cash requirement and runway?

An investor-ready quarterly revenue summary you can generate in minutes at close — with your actual MRR, gross margin by channel, and channel-level growth rates already in the narrative.

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 Investor Reporting CRM

Data sources & config

Starch syncs your Stripe data on a schedule (DTC charges, subscription revenue, refunds) and your Plaid data on a schedule (bank deposits, ACH payments from distributors, co-packer debits). Starch also syncs your QuickBooks data on a schedule (invoices, bills, payments, vendor records) so wholesale POs and broker commissions are included in the model. Your broker and buyer contacts live in the Starch CRM, which connects to Gmail so email thread context stays attached to each account.

Prompts to copy

Build me a quarterly revenue forecast that pulls Stripe DTC revenue and Plaid deposits, breaks revenue into channels (DTC, Amazon, wholesale, foodservice), and lets me override the growth rate per channel for each quarter of the year.

Create a scenario comparison: Scenario A assumes Whole Foods expansion launches in July adding $45K/month in new POs by Q4; Scenario B assumes it slips to October. Show me the runway and production cash requirement difference between the two.

Draft my Q2 investor update: include total net revenue, DTC vs. wholesale split, gross margin, burn rate, and a one-paragraph narrative about what drove the variance vs. plan. Pull the numbers from Stripe and Plaid and flag anything that looks like an anomaly.

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

Walkthrough

Step-by-step

1 Connect Stripe and Plaid so Starch syncs your DTC revenue and bank-level cash activity on a schedule — this becomes the baseline for every model you build.

2 Connect QuickBooks so Starch syncs your invoices and payments, giving you wholesale PO revenue and distributor settlement history without manual exports.

3 Open the Scenario Analysis app and tell Starch how your revenue breaks down today: DTC via Shopify/Stripe, Amazon Seller Central, distributor wholesale, and any foodservice or club channel POs.

4 Set your channel-level growth assumptions for each quarter — e.g., 'wholesale grows 15% in Q3 when the Whole Foods expansion hits, then 8% in Q4' — and let Starch calculate the implied revenue, cash timing, and production cash requirement.

5 Build a second scenario with a delayed expansion and compare the two side-by-side to see the runway and burn-rate difference before you commit to co-packer capacity.

6 In the Starch CRM, create account records for your key buyers and distributors, attaching your deal stage (authorized, in-review, pitching) and projected PO volume — so the forecast has a sales-pipeline grounding, not just a top-down growth rate.

7 Ask the CRM: 'Which wholesale accounts haven't placed a PO in 60 days?' and cross-check against your forecast assumptions — accounts you assumed would reorder may already be signaling churn.

8 At quarter-close, open Investor Reporting and tell Starch to pull Q2 actuals from Stripe and Plaid, calculate your channel revenue split and gross margin, and draft the narrative section noting any variance vs. the scenario you modeled in step 4.

9 Review the draft update, add any context Starch couldn't see (e.g., 'we had a $12K deduction dispute with KeHE that's now resolved'), and approve it for send.

10 Set a recurring automation: 'Every first Monday of the month, update my revenue forecast with the latest Stripe and Plaid data, flag any channel where actuals are more than 10% below plan, and Slack me a summary.' You get a standing heads-up without logging into anything manually.

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

Q2 2026 Revenue Close — 4-SKU Better-For-You Snack Brand

Sample numbers from a real run

DTC (Shopify / Stripe)	87,400
Amazon FBA	54,200
Wholesale — natural channel (UNFI/KeHE)	118,600
Foodservice / club (Costco test)	22,000
Gross deductions (invalid chargebacks)	-9,800
Net revenue	272,400

Going into Q2 close, the founder expected net revenue around $295K based on the January scenario model — but the Starch forecast flagged a variance early in June when Plaid showed UNFI settlement deposits running $18K below the PO values in QuickBooks. Turns out KeHE had taken $9,800 in deductions marked as 'compliance fees' that were being disputed. The Starch CRM showed the KeHE buyer contact had gone 45 days without a reply to the dispute email thread, which surfaced the relationship risk alongside the financial one. The Costco test doors ($22K) weren't in the original Q2 model at all — they were a late addition that partially offset the deduction hit. At quarter-close, the Investor Reporting app pulled Stripe DTC ($87,400), Plaid net deposits (backing into the wholesale and Amazon figures), and drafted a narrative that led with the deduction story, explained the Costco upside, and showed Q3 wholesale pipeline from the CRM — all in about 20 minutes of founder time, vs. the usual half-day of spreadsheet work.

Measurement

How you'll know it's working

Net revenue by channel (DTC vs. Amazon vs. wholesale vs. foodservice) — monthly and quarterly

Deduction rate as a percentage of gross wholesale revenue — tracks distributor chargeback exposure

Forecast accuracy — actual net revenue vs. scenario-model plan, flagged when variance exceeds 10%

Cash timing gap — days between when a wholesale PO ships and when the distributor deposit clears Plaid

Wholesale account reorder rate — which buyers placed POs in the last 60 days vs. forecast assumption

Comparison

What this replaces

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

Fathom or LivePlan connected to QuickBooks

Good at QuickBooks-native P&L views but don't pull Stripe DTC or Plaid deposits into the same model, and have no CRM or investor-reporting layer.

Spreadsheet (Google Sheets / Excel) with manual exports

Fully flexible but requires a manual export cycle from every source, breaks every time a distributor changes their settlement format, and has no scenario-comparison workflow.

Salesforce + Tableau

Enterprise-grade pipeline and revenue analytics, but requires an admin to configure, costs multiples of what a seed-stage CPG brand can justify, and doesn't touch co-packer or distributor data natively.

Mosaic or Runway (FP&A SaaS)

Purpose-built financial modeling with good scenario tools, but aimed at SaaS metrics (ARR, churn, headcount) rather than CPG channel economics like deductions, PO timing, and shelf-life-constrained inventory cash.

On Starch RECOMMENDED

One platform — scenario planning, investor reporting, crm 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 in my Amazon Seller Central revenue, not just Stripe and Plaid?

Yes — the Amazon Seller Dashboard app is live, and Starch can also automate your Amazon Seller Central through your browser for any data not covered by the direct connection. If your Amazon payouts are hitting your bank account, Plaid will also capture those deposits, so you have at least two paths to get Amazon revenue into your forecast.

My co-packer invoices and COGS aren't in Stripe or Plaid — they're in QuickBooks. Will Starch see them?

Yes. Starch syncs your QuickBooks data on a schedule, including bills, payments, and vendor records. That means co-packer invoices, ingredient costs, and freight charges in QuickBooks all feed into your gross margin calculation alongside the revenue data from Stripe and Plaid.

Does Starch store historical data, or is this just a live snapshot?

Scheduled-sync providers — Stripe, Plaid, QuickBooks — are synced on a schedule and stored in Starch, so you have a running history to trend against. That said, Starch is designed for live data surfaces and decision-support, not as a long-horizon data warehouse. If you need multi-year archived analytics, that's worth naming honestly.

QuickBooks has a P&L report view — can I use that directly?

QuickBooks report views (P&L, Transaction List, Vendor Expenses) are temporarily disabled in Starch's QuickBooks connection while an upstream fix is underway. Entity-level data — invoices, bills, payments, vendors, journal entries — syncs normally, so Starch can reconstruct your P&L from the underlying transactions. This is worth knowing upfront.

My distributors don't have APIs. How do I get their deduction data into the forecast?

Two options. First, Plaid captures what actually settles in your bank account, so you can back-calculate deductions by comparing Plaid deposits against the PO values in QuickBooks. Second, Starch can automate your distributor portals through your browser — no API needed — to scrape deduction detail reports directly from the portal UI.

I'm not ready to share investor access to my financials. Can I control what goes into the investor report?

Yes. The Investor Reporting app drafts the report for your review before anything goes out. You approve the numbers and narrative, add context Starch couldn't see, and then send. Starch doesn't auto-publish to investors without your sign-off.