Customer Support

How to run customer qbrs as Small Customer Success Teams

Customer SupportFor Small Customer Success Teams3 apps12 steps~24 min to set up

Your three-person team runs QBRs for 250 accounts every quarter. That means pulling usage data from PostHog, deal history from HubSpot, and support ticket counts from Intercom — then stitching it together in a Google Slides deck you're building from scratch each time. A single QBR deck takes 3-4 hours: exporting CSVs, reformatting numbers, writing the narrative, and chasing down the account executive for context on the last upsell conversation. Multiply that by 20 strategic accounts a quarter and you've just spent 60-80 hours on slide assembly. Nothing is wrong with your tools. What's missing is a surface that reads all of them together.

Build this on Starch → See the apps used

Customer SupportFor Small Customer Success Teams3 apps12 steps~24 min to set up

Outcome

What you'll set up

A QBR generator that pulls live account data from HubSpot, Intercom, and PostHog and drafts a full presentation in minutes — not hours

A meeting notes workflow that captures every QBR call in real time, extracts action items, and archives them against the account so nothing falls through

A reusable QBR template you describe once in Starch and run for any account, with the data automatically populated from the connections you've already set up

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

Presentation Agent Meeting Notes CRM

Data sources & config

Starch syncs your HubSpot data on a schedule (contacts, companies, deals, owners). Connect Intercom from Starch's integration catalog — the agent queries it live when your QBR app runs. Connect PostHog from Starch's integration catalog the same way. Starch syncs your Gmail and Google Calendar data on a schedule so meeting context and call history are automatically attached to accounts. The Presentation Agent assembles the deck from all of these sources when you run the prompt.

Prompts to copy

Build me a 10-slide QBR deck for Acme Corp. Pull their deal history and open opportunities from HubSpot, their support ticket volume and CSAT from Intercom for the past 90 days, and their weekly active user count from PostHog. Include a slide on usage trends, a slide on open support issues, a slide on our expansion proposal, and a next-steps slide with action items from our last call.

Transcribe my QBR call with Meridian Health starting now. After the call, generate a summary with key decisions, extract all action items with owners, and save everything under the Meridian Health account in our CRM.

Show me all accounts where support ticket volume has increased more than 20% quarter-over-quarter and usage has dropped — flag those as churn risk in a list I can review before our next team standup.

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

Walkthrough

Step-by-step

1 Connect HubSpot — Starch syncs your contacts, companies, deals, and owner data on a schedule. This becomes the backbone of every account record: renewal date, ARR, last activity, open opportunities.

2 Connect Intercom from Starch's integration catalog. When a QBR app runs for a given account, the agent queries Intercom live for ticket volume, open issues, CSAT scores, and conversation history for that account over the trailing 90 days.

3 Connect PostHog from Starch's integration catalog the same way. The agent pulls weekly active users, feature adoption, and session counts per account on demand.

4 Starch syncs your Google Calendar on a schedule. Any meeting with an account's domain is automatically associated with that account's record — so your QBR prep context includes every call from the past 12 months.

5 Open Presentation Agent and type your QBR prompt for the first account. Name the account, specify the slide structure you want, and tell Starch which data points matter: usage, tickets, expansion, next steps.

6 Presentation Agent generates the full deck in minutes. Review it, iterate on any slide in plain language ('make the usage trend chart show month-over-month instead of week-over-week'), and export to PowerPoint or a shareable link to send the customer before the call.

7 On the day of the QBR, activate Meeting Notes. It transcribes the call in real time so you're present in the conversation instead of typing. After the call, it generates a summary with key decisions and extracts action items with owners.

8 Meeting Notes archives the transcript and summary against the account record in your Starch CRM automatically — so six months from now when someone asks 'what did we agree on last time?' you can search for it.

9 After the call, review the extracted action items. Assign any that need follow-up, set a reminder, or drop them into your Task Manager app if you've set one up.

10 At the start of each new quarter, run the churn-risk prompt across your full account list: accounts with declining usage and rising ticket volume get flagged before you build the QBR schedule, so you prioritize the right accounts first.

11 Once you've run five or six QBRs, tell Starch to save your slide structure as a reusable template — 'save this as my standard QBR template so I can run it for any account with the same layout.' Every future QBR starts from that baseline.

12 Share the template with your two teammates. Anyone on the team can now run a polished, data-populated QBR deck for any account by typing the account name and a one-line prompt.

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 QBR — Meridian Health (April 3, 2026)

Sample numbers from a real run

Meridian Health — current ARR	48,000
Meridian Health — open expansion opportunity	18,000
Meridian Health — support tickets Q1 vs Q4	34
Meridian Health — weekly active users (peak Q1)	412
Time saved vs manual deck build (hours)	3

Meridian Health is a $48,000 ARR account up for renewal in June with an $18,000 expansion opportunity sitting in HubSpot. Normally, preparing this QBR would mean pulling the Q1 ticket data from Intercom (34 tickets, up from 22 in Q4 — worth addressing), finding the WAU trend in PostHog (peaked at 412 in February, dipped to 380 in March), and writing a narrative that connects the two. Instead, you typed a single prompt into Presentation Agent at 9 a.m. on April 2nd. By 9:12 you had a 10-slide deck: usage trends, ticket summary, a slide calling out the February spike and explaining it was tied to a new feature rollout, the expansion proposal framed around the two departments not yet on the platform, and a next-steps slide pre-populated with the three open action items from your January call — which Meeting Notes had archived automatically. The QBR call ran for 45 minutes. Meeting Notes transcribed it, flagged that the Meridian VP asked about SSO support twice, and extracted four action items. Your teammate got the summary in Slack before you'd finished your post-call notes. Total prep time: 25 minutes. Total QBR materials: deck, transcript, summary, action items, all archived under Meridian in Starch.

Measurement

How you'll know it's working

QBR prep time per account (target: under 30 minutes)

Percentage of strategic accounts with a QBR completed on schedule each quarter

Action item completion rate 30 days post-QBR

Accounts flagged as churn risk before QBR season vs. accounts that churned without warning

Expansion opportunities identified in QBR that converted within 90 days

Comparison

What this replaces

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

Gainsight or ChurnZero

Purpose-built for CS but starts at six figures annually and requires a CS-ops person to configure; Starch costs a fraction of that and you describe what you want in plain language instead of learning a platform.

Manual Google Slides + CSV exports

No recurring cost but consumes 3-4 hours per QBR deck and produces inconsistent output that degrades when someone is out sick or changes the template.

HubSpot reporting + Intercom separately

Both tools are strong in isolation but neither surfaces a cross-system health view — you still have to open four tabs and synthesize the story yourself before every call.

Notion AI or Google Slides with Gemini

Can help with writing and formatting but have no live connection to HubSpot, Intercom, or PostHog, so you're still doing the data gathering manually before AI touches the document.

On Starch RECOMMENDED

One platform — presentation agent, meeting notes, 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

We use HubSpot as our CRM — does Starch replace it or sit on top of it?

Starch sits on top of it. Starch syncs your HubSpot data on a schedule — contacts, companies, deals, owners — and uses it as the source of truth for account records. Your team keeps working in HubSpot for anything that lives there. Starch builds the surfaces HubSpot doesn't ship: the QBR generator, the cross-system health view, the churn-risk flag that combines HubSpot activity with Intercom ticket trends and PostHog usage in one place.

Does Starch connect to Intercom and PostHog?

Yes. Connect Intercom and PostHog from Starch's integration catalog — the agent queries each one live when your app runs. That means ticket volume, CSAT, and usage data are pulled fresh at the moment you generate a QBR deck, not from a stale weekly export.

What if the account uses Zendesk instead of Intercom?

Connect Zendesk from Starch's integration catalog the same way. The agent queries it live. You'd update your QBR prompt to reference Zendesk ticket data instead of Intercom — everything else stays the same.

Is the Presentation Agent available right now?

It's currently in development. You can request beta access to get notified when it launches. In the meantime, you can build a QBR data summary app in Starch — pulling HubSpot, Intercom, and PostHog into a structured account brief — and drop that output into Google Slides or your existing deck template. The Meeting Notes app is available today.

Can Starch handle 250 accounts, or does it slow down at that scale?

Starch is built around scheduled-sync connections that store your HubSpot, Gmail, and Google Calendar data in its own database — so querying across 250 accounts is fast. Live-queried apps like Intercom and PostHog are called on-demand when you run a specific account's QBR, not all 250 at once. You'd typically run the churn-risk scan across all accounts and then generate individual decks for the 15-20 accounts you're actually meeting with that quarter.

What are Starch's honest limits for a CS team?

A few worth knowing: Starch is not SOC 2 Type II certified yet, which may matter if your accounts ask about your vendor security posture. There's no on-prem option. Live-queried data from apps like Intercom and PostHog is not stored in Starch — it's pulled fresh each time, so Starch is not a historical data warehouse for those sources. And the Gmail OAuth consent screen currently shows the underlying connector's name rather than Starch — a verified client is on the roadmap.

Our CS team is three people. Is this overkill?

It's designed for exactly this. The tools built for CS operations — Gainsight, Catalyst, ChurnZero — assume a CS-ops person to configure them and a budget to match. Starch assumes a small team that knows what it wants but doesn't have time to build it from scratch or configure an enterprise platform. You describe the QBR workflow you want in plain language and Starch builds it. If it doesn't look right, you describe what to change.