Sales & CRM

How to forecast quarterly revenue as Chief of Staff and Founder's Office

Sales & CRMFor Chief of Staff and Founder's Office3 apps11 steps~22 min to set up

Every quarter, you spend three to five days assembling a revenue forecast that should take three hours. You're pasting HubSpot pipeline exports into a Google Sheet, cross-referencing Stripe MRR data you had to ask someone to pull, texting the sales lead for a weighted close probability update, and then manually reconciling all of it against the QuickBooks actuals the CFO sent in a different format. By the time you've built the model, half the underlying data is already stale. The CEO wants a number for the board deck Thursday. You have a spreadsheet with seventeen tabs and a nagging feeling you missed a deal that closed last week.

Build this on Starch → See the apps used

Sales & CRMFor Chief of Staff and Founder's Office3 apps11 steps~22 min to set up

Outcome

What you'll set up

A live revenue forecast that pulls directly from HubSpot pipeline data and Stripe subscription metrics — updated automatically, not whenever someone remembers to export a CSV

A scenario model that shows the CEO three versions of Q3 in 10 minutes: best case, base case, and 'what if two enterprise deals slip' — built on your actual numbers, not made-up assumptions

A repeatable quarterly close workflow so the next board deck takes two hours instead of two days, with the same structure every time

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

Sales Agent CRM Scenario Analysis Investor Reporting

Data sources & config

Starch syncs your HubSpot contacts, companies, deals, and owner data on a schedule — so pipeline stage and deal value are always current without a manual export. Starch syncs your Stripe charges, subscriptions, and invoices on a schedule for recognized revenue actuals. Starch syncs your Plaid transactions and balances on a schedule to ground burn rate. QuickBooks entities (invoices, bills, payments, journal entries) also sync on a schedule for accounting actuals. These four data sources combine into a single dataset your forecast app reads against.

Prompts to copy

Build me a quarterly revenue forecast view that pulls our open pipeline from HubSpot — weighted by stage probability — and our current Stripe MRR, then projects total recognized revenue for Q3 broken out by new business, expansion, and renewal. Flag any deal over $50k that hasn't had an activity update in 14 days.

Create three Q3 scenarios side by side: base case using current pipeline at weighted close rates, upside case assuming our two largest enterprise deals close in July, and downside case where those deals slip to Q4. Show projected revenue, burn rate, and runway for each. Pull actuals from Stripe and Plaid so the baseline isn't made up.

Draft a Q2 close summary for the board update: actual revenue versus forecast, top three deals that closed, top three that slipped with reason codes, and one-paragraph narrative on what it means for the Q3 number.

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

Walkthrough

Step-by-step

1 Connect HubSpot in Starch — deals, stages, owners, and close dates sync automatically on a schedule. You do not export a CSV again.

2 Connect Stripe in Starch — MRR, new subscriptions, expansions, churned revenue, and payouts sync on a schedule. Your revenue actuals are live.

3 Connect Plaid in Starch — categorized transactions and balances sync on a schedule. This is your burn denominator for any runway calculation.

4 Connect QuickBooks in Starch — invoices, bills, payments, and journal entries sync on a schedule. This is your reconciliation layer when Stripe and accounting numbers diverge.

5 Open the Sales Agent CRM starter app and tell Starch: 'Customize this for a 150-person SaaS company. Add a field for deal type (new business / expansion / renewal), expected close quarter, and last activity date. Flag deals over $50k with no activity in 14 days.' Starch builds the view; you don't configure a single field manually.

6 Open the Scenario Analysis starter app and tell Starch: 'Build a Q3 forecast with three scenarios. Base case: current HubSpot pipeline weighted by stage probability plus current Stripe MRR. Upside: add $240k from the two enterprise deals in final negotiation. Downside: those deals slip to Q4.' Starch models all three against your real Stripe and Plaid baseline.

7 Review the scenario outputs with the CEO in your weekly sync — not a 45-minute spreadsheet walkthrough, a 10-minute decision conversation about which scenario to plan hiring against.

8 Build a pipeline health dashboard on top of the same connected data: 'Show me a table of every open deal expected to close this quarter, sorted by weighted value, with stage, owner, last activity date, and a flag if the deal has slipped from its original close date.' This replaces the HubSpot report you were screenshotting into slides.

9 At quarter close, open the Investor Reporting starter app and tell Starch: 'Draft the Q2 section of our board update. Pull actual revenue from Stripe and QuickBooks, compare to the forecast we set in April, list the top five deals that closed and the top three that slipped. Write a one-paragraph narrative in the voice of our past updates.' Starch drafts it; you edit for 20 minutes instead of writing from scratch for two hours.

10 Save all three apps — forecast, pipeline health, board draft — as a named workspace called 'Quarterly Revenue Close.' Next quarter, open it, confirm the data connections are current, and run the same workflow in a fraction of the time.

11 Set an automation: 'Every Friday at 8am, pull this week's HubSpot pipeline changes — new deals added, stage changes, deals marked closed-won or closed-lost — and Slack me a summary with the net change to our Q3 forecast number.' You stop asking the sales lead for weekly pipeline updates.

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 Close — 150-person SaaS, $4.2M ARR base

Sample numbers from a real run

New business closed (Stripe, Q2 actuals)	312,000
Expansion revenue (Stripe, Q2 actuals)	87,000
Churned MRR (Stripe, Q2 actuals)	-41,000
Open pipeline, Q3 weighted (HubSpot)	528,000
Enterprise deal A — upside scenario (HubSpot)	140,000
Enterprise deal B — upside scenario (HubSpot)	98,000
Current monthly burn (Plaid)	-310,000

Going into the Q2 board update, you had two problems: Stripe showed $358k in net new ARR for the quarter, but the forecast built in April had called $410k. The $52k gap needed an explanation, not a number. Starch pulled the closed-lost deals from HubSpot and surfaced that three mid-market deals in the $15-20k range had slipped to Q3 — all owned by one rep who had joined the company in February. That was the story. For Q3 planning, the Scenario Analysis app showed the base case at $528k weighted pipeline, the upside at $766k if both enterprise deals closed in July, and the downside at $290k if they slipped. The CEO chose the base case for headcount planning and the downside case for cash management. That decision — which used to require a 90-minute spreadsheet session — took 12 minutes in the Thursday sync. The board draft came out of Investor Reporting: Starch pulled the $312k new business number from Stripe, the $310k monthly burn from Plaid, calculated 14 months of runway, and drafted the narrative paragraph explaining the forecast miss. You edited two sentences and sent it.

Measurement

How you'll know it's working

Forecast accuracy: actual closed revenue vs. weighted pipeline forecast, tracked quarter over quarter

Pipeline coverage ratio: total open Q3 pipeline value divided by Q3 revenue target (you want 3x minimum)

Deal slip rate: percentage of deals that miss their original committed close quarter, by rep and by segment

Weighted pipeline delta week-over-week: net change to the Q3 number from new deals added, stage changes, and closes

Days to board-ready close summary: how long from quarter end to a draft the CEO can actually review

Comparison

What this replaces

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

HubSpot + Google Sheets manual export

You already have this stack — the problem is that every forecast starts with a manual CSV export, and by Thursday's board prep the data is three days stale.

Salesforce + Excel financial model

More powerful for large sales orgs with dedicated RevOps, but at 150 people you don't have RevOps — you have you, and Salesforce admin work eats half the time you'd save.

Clari or Gong Forecast

Purpose-built for pipeline forecasting, but they sit on top of your CRM and add another tool subscription; they don't connect to your Stripe actuals or Plaid burn without a separate integration build.

Mosaic or Pigment (FP&A tools)

Strong for finance-led scenario modeling once you have a dedicated FP&A hire, but the setup time and cost assume someone will own the model — that person is currently you, and you have twelve other things due this week.

On Starch RECOMMENDED

One platform — sales agent crm, 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

Our HubSpot pipeline is a mess — deals in the wrong stages, stale close dates, missing values. Will this still work?

Starch syncs whatever is actually in HubSpot — it doesn't clean the data for you automatically. But once the sync is live, you can tell Starch: 'Show me every deal missing a close date or with a close date in the past that isn't marked closed-lost' and use that as a cleanup list. The forecast will reflect the real state of your pipeline, which is better than a manually tidied export that hides the same problems.

Can Starch pull the QuickBooks P&L view directly for the board update?

QuickBooks P&L and Transaction List report views are temporarily unavailable due to an upstream connector issue. Entity-level data — invoices, bills, payments, vendors, journal entries — syncs normally. For a board-level income statement, Starch can construct a summary from those entities. It's not a one-click P&L export, but the underlying numbers are there.

We use Salesforce, not HubSpot. Does the forecast workflow still apply?

Yes. Salesforce is available through Starch's integration catalog, queried live when your forecast app runs. You won't get the scheduled sync that HubSpot has — meaning Salesforce data is pulled fresh each time you run the app rather than stored in Starch — but for a quarterly forecast workflow where you're pulling numbers a few times a week, that's perfectly workable.

Is Starch SOC 2 certified? We have a security review process.

Starch is not SOC 2 Type II certified yet. If your company's security policy requires certified vendors before connecting production financial data, that's a real blocker and worth knowing upfront. It's on the roadmap.

How is this different from just building a better dashboard in Google Sheets with a HubSpot API pull?

The main difference is maintenance. A Google Sheets model with an API pull breaks when field names change, when someone adds a new deal stage, or when you need to add Stripe data alongside HubSpot. Starch's apps are built on top of live synced data with an agent that can re-query when you change the question. When the CEO asks 'what happens if we delay two hires,' you tell Starch — you don't rebuild the formula layer.

Can I set this up so the CEO gets a weekly pipeline summary without me having to send it?

Yes. Tell Starch: 'Every Monday at 7am, pull last week's HubSpot pipeline changes — new deals, stage changes, closed-won, closed-lost — calculate the net impact on our Q3 weighted forecast, and send a Slack message to the #exec channel with a three-bullet summary.' Starch builds and schedules the automation. Starch connects directly to Slack, so no separate webhook setup is needed.