Sales & CRM

How to forecast quarterly revenue as Small Finance Teams

Sales & CRMFor Small Finance Teams2 apps12 steps~24 min to set up

Every quarter, your three-person finance team rebuilds the revenue forecast from scratch. You pull Stripe MRR into a Google Sheet, manually reconcile it against QuickBooks invoices, add a pipeline haircut from a HubSpot export that's already two days stale, and then your VP of Sales changes the close dates on six deals and the whole model breaks. The board wants a 90-day revenue forecast with scenario assumptions baked in. You have Friday afternoon and a spreadsheet. The ERP captures what happened; it has no opinion on what's coming. You end up spending more time stitching data between systems than actually thinking about the numbers.

Build this on Starch → See the apps used

Sales & CRMFor Small Finance Teams2 apps12 steps~24 min to set up

Outcome

What you'll set up

A live quarterly revenue forecast that pulls actuals from Stripe and QuickBooks on a schedule, so your baseline is always current without a manual refresh

A scenario layer where you can adjust assumptions — new logo adds, churn rate, deal close timing — and see how each scenario affects quarterly revenue, burn, and runway side by side

An investor-ready output that formats the forecast narrative automatically, so the board deck section that used to take you half a day drafts itself

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

Data sources & config

Starch syncs your Stripe data on a schedule (charges, invoices, subscriptions) and syncs your QuickBooks data on a schedule (invoices, payments, journal entries — 20+ entity types). Plaid connects on a schedule for cash and bank transaction context. For pipeline data, connect your CRM from Starch's integration catalog and the agent queries it live when the forecast runs. The Scenario Analysis app uses the Stripe and Plaid scheduled sync as the baseline; Investor Reporting pulls from the same synced data to draft the narrative output.

Prompts to copy

Build me a quarterly revenue forecast model that uses my Stripe MRR and QuickBooks invoices as the actuals baseline. I want three scenarios: base case, upside (20% more new logos, 1% lower monthly churn), and downside (15% slower new logo growth, 1.5% higher churn). Show me projected revenue, net new ARR, and cumulative burn for each scenario through end of Q3 2026.

Every month, pull the latest Stripe and Plaid data and draft the revenue section of my investor update. Include MRR, net new ARR, projected quarterly revenue vs. plan, and a two-sentence narrative on what's driving the variance. Send it to me as a draft before it goes to the investor list.

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

Walkthrough

Step-by-step

1 Connect Stripe in Starch — Starch syncs your charges, invoices, subscriptions, and payouts on a schedule, so MRR and recognized revenue are always current without you touching them.

2 Connect QuickBooks in Starch — Starch syncs invoices, payments, vendors, and journal entries. This gives you the accrual-basis actuals to cross-reference against Stripe cash.

3 Connect Plaid in Starch — Starch syncs your bank transactions and balances on a schedule. This is your cash layer: real burn rate, not just what's in the P&L.

4 Open the Scenario Analysis starter app from the Starch App Store and fork it. The template comes pre-wired to Stripe and Plaid; you'll extend it to pull QuickBooks invoices as well.

5 Tell Starch what your forecast structure looks like: 'My quarterly revenue model tracks new logo ARR, expansion ARR, churned ARR, and professional services revenue separately. QuickBooks is the system of record for billed revenue; Stripe is the system of record for subscription MRR.'

6 Define your three scenario assumptions in plain language. Starch builds the scenario branches — you're not writing formulas, you're describing what changes between base, upside, and downside.

7 If your pipeline lives in HubSpot or another CRM, connect it from Starch's integration catalog. Tell Starch: 'Pull open deals from HubSpot with close date in Q3 2026, apply a 70% probability haircut to anything past 60 days old, and add the weighted total to my new logo ARR projection.' Starch queries HubSpot live each time the forecast refreshes.

8 Review the scenario outputs — revenue, net new ARR, burn, runway — and adjust any assumption that looks off. This is where you actually think about the numbers instead of building the infrastructure around them.

9 Open the Investor Reporting app and tell it to pull the quarterly revenue forecast outputs and draft the financial section of your next board update. Give it your preferred narrative tone from a past update so it matches how you write.

10 Set a schedule: every Monday, Starch refreshes the Stripe and QuickBooks sync, reruns the scenario model with updated actuals, and Slacks you a summary of how Q3 tracking is trending vs. plan.

11 When close week arrives and the VP of Sales moves deal dates, update the pipeline assumption in one prompt — Starch reruns the model and the board deck section updates automatically rather than cascading through a fragile spreadsheet.

12 Before the board meeting, export the scenario comparison and the narrative draft. You spend an hour reviewing and editing rather than three days building.

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 Quarterly Revenue Forecast — March Close

Sample numbers from a real run

Stripe subscription MRR (actuals through March 31)	184,000
QuickBooks invoiced professional services (Q2 to date)	47,200
HubSpot pipeline — weighted new logo ARR (Q2 close dates, 70% haircut)	62,400
Projected expansion ARR (net of 1.2% monthly churn, base case)	18,600
Projected Q2 total revenue — base case	312,200
Projected Q2 total revenue — upside case (+20% new logos, 1% churn)	341,800
Projected Q2 total revenue — downside case (15% fewer new logos, 1.5% churn)	278,900

When the team ran this at the end of March, Stripe showed $184K in monthly subscription revenue — up $11K from February but $6K below the Q2 plan line. QuickBooks had $47.2K in professional services invoices already billed, tracking ahead of the $40K quarterly budget. The HubSpot pipeline had $89K in deals targeting a Q2 close; after a 70% haircut on deals past 60 days, the weighted contribution dropped to $62.4K. Base case Q2 total came in at $312K — about 4% below the $325K board target. The downside scenario at $279K was the conversation that actually mattered: it would push the fundraise timeline by six weeks. That's the number the CFO needed before the Monday board call, and Starch had it ready Friday morning without anyone rebuilding the model by hand.

Measurement

How you'll know it's working

Net new ARR by quarter (new logo + expansion − churn)

MRR-to-invoice reconciliation gap (Stripe cash vs. QuickBooks accrual)

Weighted pipeline coverage ratio (weighted pipeline / remaining quarterly revenue gap)

Quarterly revenue variance to plan (actuals + forecast vs. board target)

Runway under each scenario (months of cash at projected burn rate)

Comparison

What this replaces

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

Google Sheets rebuilt every quarter

You own every formula and every data pull — when Stripe or QuickBooks changes or a deal date moves, you fix it manually; Starch keeps the actuals current automatically and lets you change assumptions in plain language.

NetSuite built-in planning module

NetSuite planning is strong on the ledger side but doesn't incorporate live Stripe MRR or weighted CRM pipeline without custom implementation work your team doesn't have the bandwidth to own.

Mosaic or Drivetrain (FP&A SaaS)

Purpose-built FP&A tools have richer pre-built financial models, but they're priced and configured for teams larger than three people and won't let you build the one-off surfaces — the AR aging dashboard, the board narrative draft — that take as much time as the model itself.

HubSpot Forecasting

HubSpot's native forecast view is pipeline-only — it doesn't touch your Stripe MRR or QuickBooks actuals, so you still have to stitch the two together somewhere.

On Starch RECOMMENDED

One platform — scenario planning, investor reporting 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

Does Starch actually keep the Stripe and QuickBooks data current, or do I have to trigger a manual sync before every forecast run?

Starch syncs both on a schedule in the background — you set the cadence (daily is common for these sources) and the data is already fresh when the forecast runs. You don't pull anything manually. The one honest caveat: QuickBooks report views like the P&L summary are temporarily unavailable due to an upstream fix in progress, but the entity-level data — invoices, payments, journal entries, vendors — syncs normally, and that's what the revenue forecast model uses.

Our pipeline is in HubSpot. Can Starch pull deal data into the forecast, or do I have to export it?

No export needed. Connect HubSpot from Starch's integration catalog and the agent queries it live when the forecast runs — current close dates, amounts, and stages, not a stale CSV. You tell Starch how to weight or haircut the pipeline in plain language and it applies that logic each time.

What if our CRM is Salesforce instead of HubSpot?

Salesforce is available from Starch's integration catalog and the agent queries it live. Describe the same pipeline logic — deal stage, close date, probability weighting — and Starch pulls from Salesforce instead. The forecast model doesn't care which CRM you're in.

We track professional services revenue separately from subscription MRR. Can the forecast model handle that?

Yes. Tell Starch how your revenue breaks down — subscriptions in Stripe, professional services invoiced in QuickBooks — and describe which line items map to which revenue type. Starch builds the model to match your actual revenue structure, not a generic template. You're not forced into someone else's category scheme.

Is Starch SOC 2 Type II certified? Our CFO will ask before we connect the bank and accounting data.

Not yet. Starch is not SOC 2 Type II certified today. If your organization requires a certified vendor before connecting Plaid or QuickBooks, that's a real constraint and worth knowing upfront. It's on the roadmap.

Can the same setup generate the board-ready forecast narrative, or is that a separate step?

The Investor Reporting app handles the narrative layer. Once the scenario model is producing numbers, you wire the Investor Reporting app to pull from the same Stripe and Plaid data and draft the financial narrative in your voice. You review and edit; you don't write from scratch. For a three-person finance team, that's the difference between spending a day on the board deck and spending an hour on it.