38 new customers in September, MQL to SQL down to 29%
Leads rose 10% to 1,310 and customers to 38. MQL to SQL fell from 33% to 29%: sales accepted 118 of 412 MQLs.
On the 1st: last month's visits, leads, MQLs, SQLs and customers from HubSpot, each step's conversion rate, and the step that slipped.
Reads
You get
The card on the right is a sample. Yours shows your own numbers and updates in place every run.
What lands on your board
Leads rose 10% to 1,310 and customers to 38. MQL to SQL fell from 33% to 29%: sales accepted 118 of 412 MQLs.
The monthly funnel slide takes longer than it should. Lifecycle stage counts come from one report, traffic from another, and someone divides them in a spreadsheet that never quite matches last month's version. By the time it is done, nobody asks which step changed.
The agent counts how many contacts entered each lifecycle stage last month, straight from HubSpot's stage dates, sets them against website visits and the month before, and points at the step whose rate moved most. Marketing sees whether lead volume held; sales sees whether the hand-off did.
For each stage, POST /crm/v3/objects/contacts/search with one filter on hs_v2_date_entered_<stage> (lead, marketingqualifiedlead, salesqualifiedlead, opportunity, customer) using operator: "BETWEEN" from the first to the last millisecond of last month, and limit: 1. Read the count from total.
Call GET /analytics/v2/reports/totals/total?start=YYYYMMDD&end=YYYYMMDD for last month and read visits. If the HubSpot tracking code is not on every page, take GA4 sessions for the same month instead and label the row.
Run the same counts for the previous month. If you use custom lifecycle stages, list their internal IDs once from GET /crm/v3/properties/contacts/lifecyclestage and use those in the property names.
Visit to lead, lead to MQL, MQL to SQL and SQL to customer, each as the later count over the earlier one in the same month. Flag a step whose rate changed by more than 15% relative to last month.
A table of stage counts and step rates for both months, then one line naming the step that moved and the likely reason from the data (fewer MQLs accepted, a lead source that dried up).
With BRIEFKIN_TOKEN set, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shellcurl -fsSL https://briefkin.dev/skills/hubspot-monthly-funnel-report/install | shAsk Claude Code:
promptRun the hubspot-monthly-funnel-report skillOr paste the prompt instead of installing:
promptUsing the HubSpot CRM API, count how many contacts entered each lifecycle stage last month and the month before: search contacts with a BETWEEN filter on hs_v2_date_entered_lead, hs_v2_date_entered_marketingqualifiedlead, hs_v2_date_entered_salesqualifiedlead and hs_v2_date_entered_customer (limit 1, read total). Get website visits for both months from HubSpot's analytics API (totals report) or, if that is not available, from GA4 sessions. Show a table of visits, leads, MQLs, SQLs and customers for both months with each step's conversion rate and its relative change. Name the step whose rate changed most and suggest why.1st of the month at 9:00, from cron on any machine where Claude Code is signed in:
crontab0 9 1 * * claude -p "Run the hubspot-monthly-funnel-report skill"Add it with crontab -e. A laptop asleep at that time runs it when it wakes if you use launchd or Task Scheduler instead; GitHub Actions works for the API route.
Connect Briefkin once with curl -fsSL https://briefkin.dev/install | sh and every run lands on your board as a report card. How setup works
SKILL.md---
name: hubspot-monthly-funnel-report
description: "On the 1st: last month's visits, leads, MQLs, SQLs and customers from HubSpot, each step's conversion rate, and the step that slipped. Use when the user asks for this, or when it runs on a schedule."
---
# HubSpot monthly funnel report
On the 1st: last month's visits, leads, MQLs, SQLs and customers from HubSpot, each step's conversion rate, and the step that slipped.
The monthly funnel slide takes longer than it should. Lifecycle stage counts come from one report, traffic from another, and someone divides them in a spreadsheet that never quite matches last month's version. By the time it is done, nobody asks which step changed.
The agent counts how many contacts entered each lifecycle stage last month, straight from HubSpot's stage dates, sets them against website visits and the month before, and points at the step whose rate moved most. Marketing sees whether lead volume held; sales sees whether the hand-off did.
## Two ways to get the data
Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used.
## Route A: API or MCP server
### What you need
- A HubSpot user or private app that can read contacts (`crm.objects.contacts.read`), plus access to HubSpot's analytics for visits.
- The HubSpot MCP server or connector, or a private app token in `HUBSPOT_ACCESS_TOKEN` with read scopes only.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **Count stage entries** (HubSpot): For each stage, `POST /crm/v3/objects/contacts/search` with one filter on `hs_v2_date_entered_<stage>` (`lead`, `marketingqualifiedlead`, `salesqualifiedlead`, `opportunity`, `customer`) using `operator: "BETWEEN"` from the first to the last millisecond of last month, and `limit: 1`. Read the count from `total`.
2. **Get website visits** (HubSpot): Call `GET /analytics/v2/reports/totals/total?start=YYYYMMDD&end=YYYYMMDD` for last month and read `visits`. If the HubSpot tracking code is not on every page, take GA4 `sessions` for the same month instead and label the row.
3. **Repeat for the month before**: Run the same counts for the previous month. If you use custom lifecycle stages, list their internal IDs once from `GET /crm/v3/properties/contacts/lifecyclestage` and use those in the property names.
4. **Work out the step rates**: Visit to lead, lead to MQL, MQL to SQL and SQL to customer, each as the later count over the earlier one in the same month. Flag a step whose rate changed by more than 15% relative to last month.
5. **Write the report**: A table of stage counts and step rates for both months, then one line naming the step that moved and the likely reason from the data (fewer MQLs accepted, a lead source that dried up).
## Route B: the browser
### What you need
- Signed in to HubSpot with a user that can view contacts and marketing analytics (a Marketing Hub or Content Hub seat for the traffic numbers).
If something is missing, say what and stop; do not invent numbers.
### Steps
Start at https://app.hubspot.com/contacts/. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears.
1. **Count entries into each stage**: Open HubSpot and pick the account if more than one is listed; the contacts page opens. Click **Advanced filters**, add `Date entered "Lead"` **is between** the first and last day of last month, and read the record count above the table. Repeat with `Date entered "Marketing Qualified Lead"`, `Date entered "Sales Qualified Lead"`, `Date entered "Opportunity"` and `Date entered "Customer"`, then the same five for the month before.
2. **Read website visits**: Open **Reporting > Marketing analytics** (older accounts: **Reports > Analytics tools > Traffic analytics**), **Sources** tab, set the date range to last month with frequency **Monthly**, and read the `Sessions` total. Switch to the month before. If the tracking code is not on every page, take GA4 sessions instead and label the row.
3. **Work out the step rates**: Visit to lead, lead to MQL, MQL to SQL and SQL to customer, each as the later count over the earlier one in the same month. Flag a step whose rate changed by more than 15% relative to the month before.
4. **Write the report**: A table of stage counts and step rates for both months, then one line naming the step that moved and the likely reason.
Note: Ten filters and two analytics views is a slow run; the filter counts are exact, though, and match the API totals. HubSpot renamed its reporting menus more than once, so if **Marketing analytics** is not in your navigation, look under **Reports > Analytics tools**. The web app needs a desktop-size window for the filter panel.
## The report
- **Funnel table**: visits, leads, MQLs, SQLs and customers for last month and the month before.
- **Step rates**: each stage's conversion from the one before, with the change.
- **The step that slipped** or improved most, in one line.
- A note when visits came from GA4 instead of HubSpot.
## Post it to Briefkin
When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on.
```bash
[ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env
curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \
-H "authorization: Bearer $BRIEFKIN_TOKEN" \
-H "content-type: application/json" \
-d @- <<'JSON'
{
"v": 1,
"event_id": "hubspot-monthly-funnel-report/<date of this run, e.g. 2026-10-16>",
"item_key": "hubspot-monthly-funnel-report",
"project": "Acme",
"source": "HubSpot monthly funnel report skill",
"kind": "update",
"title": "38 new customers in September, MQL to SQL down to 29%",
"summary": "Leads rose **10%** to 1,310 and customers to **38**. MQL to SQL fell from 33% to **29%**: sales accepted 118 of 412 MQLs.",
"widget": [
{
"type": "Table",
"props": {
"columns": [
"Stage",
"September",
"August",
"Step rate",
"Change"
],
"align": [
"left",
"right",
"right",
"right",
"right"
],
"rows": [
[
"Visits",
52400,
49800,
"",
""
],
[
"Leads",
1310,
1190,
"2.5%",
"+4%"
],
[
"MQLs",
412,
365,
"31.5%",
"+3%"
],
[
"SQLs",
118,
121,
"28.6%",
"-14%"
],
[
"Customers",
38,
34,
"32.2%",
"+15%"
]
]
}
},
{
"type": "Callout",
"props": {
"text": "MQL to SQL is the step that slipped: 47 more MQLs, 3 fewer SQLs. Check how many MQLs sales has not worked yet.",
"tone": "warn"
}
}
]
}
JSON
```
Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report.
## Run it on a schedule
Suggested: 1st of the month at 9:00 (`0 9 1 * *`). From cron, with the agent signed in on that machine:
```bash
0 9 1 * * claude -p "Run the hubspot-monthly-funnel-report skill"
# Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run
```
Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/hubspot-monthly-funnel-report#schedule
Source: https://briefkin.dev/skills/hubspot-monthly-funnel-report
A HubSpot user or private app that can read contacts (crm.objects.contacts.read), plus access to HubSpot's analytics for visits.
The HubSpot MCP server or connector, or a private app token in HUBSPOT_ACCESS_TOKEN with read scopes only.
https://mcp.hubspot.com. HubSpot also offers a Claude connector built on the same server; you sign in with your HubSpot user and it sees what that user can see (docs).crm.objects.deals.read, crm.objects.contacts.read, crm.objects.companies.read, crm.objects.owners.read), and store its token in HUBSPOT_ACCESS_TOKEN.The MCP server and connector can create and update records, so for unattended reports a private app with only .read scopes is the safer credential.
No. It counts entries into each stage during the month, so an SQL in September may have been an MQL in July. That is the view most monthly reviews use, and it is stable once the month is over. For true cohort conversion, follow one month's new leads forward with the same hs_v2_date_entered_ properties.
A contact set straight to customer gets a customer date but no SQL date, so step rates can top 100% for small numbers. The report shows raw counts next to the rates for that reason. If skipping is common, ask the team whether stages are set by hand or by workflows.
No. The skill runs with your own agent and gives you the report in the chat. With a Briefkin token set, each run also posts it to your board as a report card, so you can glance at it instead of asking again.
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Windsurf and Amp all read SKILL.md files. The install command puts it in ~/.claude/skills and ~/.agents/skills, where they look. ChatGPT, the Claude app and Grok take the prompt instead. Pick your agent under "Install and run it" and the commands change to match.
Yes. Switch the data source to "Your browser" and the steps change to the HubSpot pages themselves: the agent opens app.hubspot.com/contacts in a browser you are signed in to and reads the numbers off the page. Claude Code, Codex, Cursor, Claude app and ChatGPT can drive a browser out of the box; Gemini CLI, Copilot CLI and OpenCode need a Playwright or Chrome DevTools MCP server first. It is slower than the API and needs you signed in, so it suits a report you run by hand or from a scheduler on your own computer.
Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.
Free during the open beta. No card.