Internal Comms & Meetings

How to draft a slack announcement as Small Marketing Teams

Internal Comms & MeetingsFor Small Marketing Teams2 apps12 steps~24 min to set up

Your three-person team runs announcements through a process that would embarrass a larger org: someone writes a draft in Notion, pastes it into Slack, realizes the campaign numbers are stale, goes back to HubSpot to pull the right MQL count, checks GA4 for traffic, opens the Meta Ads dashboard for spend, assembles the real numbers in a Google Doc, rewrites the Slack message, tags the CEO for approval, and by the time it posts the numbers are from yesterday. Campaign launches, pipeline updates, MQL reports to the broader team — each one is a 45-minute research-and-assembly job dressed up as a two-paragraph Slack message. Nobody on a 3-person team should spend that long on internal comms.

Build this on Starch → See the apps used

Internal Comms & MeetingsFor Small Marketing Teams2 apps12 steps~24 min to set up

Outcome

What you'll set up

A Starch app that pulls live campaign data from HubSpot, GA4, and your ad platforms and drafts a complete Slack announcement in your team's voice — ready to post in under 5 minutes

A recurring weekly trigger that auto-generates a pipeline-contribution summary (MQLs sourced, revenue influenced, paid spend vs. pipeline) every Monday morning before standup

A searchable archive of every announcement you've sent, cross-referenced with the underlying data, so you can answer 'what did we say about that campaign?' without digging through Slack history

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

Knowledge Management Email Agent

Data sources & config

Starch syncs your HubSpot data on a schedule (contacts, companies, deals, owners). Connect GA4 and Meta Ads from Starch's integration catalog — the agent queries them live when the announcement app runs. Connect Google Ads from Starch's integration catalog the same way. Notion syncs on a schedule if you're storing campaign briefs there. Slack is available through Starch's integration catalog for posting the finished draft.

Prompts to copy

Build me a Slack announcement app for my marketing team. It should pull this week's MQL count and deal pipeline influenced from HubSpot, session volume and top acquisition source from GA4, and total paid spend from Meta Ads and Google Ads. Draft a 3-paragraph Slack message in a direct, non-corporate tone — one paragraph on what we launched, one on what the numbers say, one on what we're doing next. Let me edit before it posts.

Every Monday at 8 AM, pull last week's HubSpot MQLs, GA4 organic sessions, and combined Meta + Google ad spend. Draft a weekly pipeline-contribution summary I can paste into the #marketing Slack channel. Flag anything that moved more than 20% week-over-week.

Create a knowledge base entry for every Slack announcement we post. Tag it with the campaign name, date, MQL count at the time of posting, and a one-sentence summary. Make it searchable so I can find what we said about any campaign in under 10 seconds.

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

Walkthrough

Step-by-step

1 Connect HubSpot — Starch syncs your contacts, deals, and pipeline data on a schedule, so MQL counts and revenue-influenced figures are always current when an announcement runs.

2 Connect GA4 from Starch's integration catalog. The agent queries it live when your announcement app runs, so session volume and acquisition source data reflect the actual period you're announcing.

3 Connect Meta Ads and Google Ads from Starch's integration catalog. Live queries mean your spend figures match what's in the platform at the moment of drafting — no manual exports.

4 If you keep campaign briefs in Notion, Starch syncs those on a schedule. The announcement app can pull the campaign name, target audience, and stated goal directly from the brief.

5 Describe what you want: tell Starch the structure of a typical announcement your team sends (launch context, key metrics, next action), and Starch builds the drafting app around that template.

6 Set the trigger — either manual (you kick it off when a campaign wraps) or scheduled (every Monday at 8 AM for a weekly pipeline contribution summary). Both work in the same app.

7 When the app runs, Starch queries HubSpot for MQL count and influenced pipeline, GA4 for sessions and top source, and the ad platforms for spend — all in one pass.

8 The draft surfaces in Starch with the numbers filled in. You review the paragraph that contextualizes the MQL movement, edit the tone if needed, and approve.

9 Connect Slack from Starch's integration catalog. Starch posts the approved draft to the channel you specify — #marketing, #company-wide, or a campaign-specific channel.

10 The Knowledge Management app logs every posted announcement with the campaign name, date, metric snapshot, and a one-sentence summary. Next time someone asks 'what did we say about the Q1 webinar campaign?' you search and find it in seconds.

11 If a number moves more than 20% week-over-week — say, MQLs drop sharply — the app flags it in the draft so you don't post a routine-sounding update when the number actually needs explaining.

12 Fork the base announcement app for different use cases: one version for external-facing launch announcements in a company-wide channel, one for internal marketing team standups with more granular channel-level breakdown.

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

April 2026 LinkedIn Webinar Campaign Wrap

Sample numbers from a real run

HubSpot MQLs (past 7 days)	47
GA4 sessions from webinar landing page	3,820
LinkedIn Ads spend	4,200
Google Ads spend	1,800
Pipeline influenced (HubSpot deals touched)	138,000

The webinar wrapped on Thursday. By Friday morning you'd normally spend 40 minutes pulling these numbers manually before writing a two-paragraph Slack message to send to #company-wide. Instead: Starch queries HubSpot and surfaces 47 MQLs sourced in the past 7 days and $138,000 in influenced pipeline. It queries GA4 and finds 3,820 sessions on the webinar landing page, with LinkedIn Organic as the top source (not the paid campaign — worth calling out). It queries Meta Ads and Google Ads from the integration catalog and confirms $6,000 total spend. The draft reads: 'The LinkedIn webinar drove 47 MQLs and $138K in influenced pipeline this week. Interesting note: organic LinkedIn outperformed paid on registration traffic — 1,840 sessions vs. 1,200 from the $4.2K LinkedIn campaign. We're pausing paid amplification on the replay and letting organic run. Full recording goes out in Tuesday's nurture email.' You edit one sentence, hit post. The Knowledge Management app logs the campaign with the metric snapshot. Total time: 6 minutes.

Measurement

How you'll know it's working

MQLs sourced per campaign (HubSpot, weekly)

Pipeline influenced per marketing dollar spent (HubSpot deals vs. Meta + Google Ads spend)

Time from campaign wrap to internal announcement posted (target: under 10 minutes)

Announcement searchability — can you find what you said about any campaign in under 30 seconds?

Week-over-week MQL movement flagged before it becomes a question from the CEO

Comparison

What this replaces

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

Writing the Slack message manually with tabs open in HubSpot, GA4, and Meta Ads

Works, but costs 30-45 minutes per announcement and the numbers are always slightly stale by the time you post — Starch gets you from 'campaign wrapped' to 'Slack posted' in under 10 minutes with live data.

Notion + a shared doc template for campaign recaps

Keeps narrative in one place, but someone still has to fill in the numbers manually; Starch automates the data-pull and drafting while still storing the output in a searchable knowledge base.

A BI tool like Looker Studio or Databox wired to your stack

Good for persistent dashboards, but doesn't draft the announcement or post it — you'd still write the Slack message yourself; Starch is the layer that turns the data into a ready-to-send communication.

ChatGPT or Claude with manually pasted numbers

Fast drafting once you have the numbers, but you still spend 20 minutes pulling them; Starch connects directly to HubSpot, GA4, and your ad platforms so the data-fetch and drafting happen in one step.

On Starch RECOMMENDED

One platform — knowledge management, email agent 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 actually post to Slack automatically or does it just draft?

Both. Connect Slack from Starch's integration catalog and you can configure the app to post automatically on a schedule (e.g., every Monday at 8 AM) or to surface a draft you approve before it posts. For most teams, the approval step is worth the extra 30 seconds — you want to catch the week a number looks weird before it goes to #company-wide.

We use Customer.io and Amplitude, not Mailchimp or Mixpanel. Can Starch pull data from those?

Customer.io and Amplitude are reachable from Starch's integration catalog — the agent queries them live when your announcement app runs. If for any reason a specific metric isn't accessible via the catalog, Starch can automate the web interface through browser automation. Either way, you're not blocked.

Does Starch store our HubSpot and GA4 data or just query it?

HubSpot data syncs on a schedule and is stored in Starch — contacts, deals, pipeline figures are always ready. GA4 and ad platform data is queried live from Starch's integration catalog each time your app runs, not stored persistently. For announcement drafting this is fine; for long-horizon archival analytics it's worth knowing. Starch is not a data warehouse replacement.

What if the MQL count in HubSpot is wrong because of attribution weirdness?

Starch surfaces whatever's in HubSpot — it doesn't fix underlying attribution. If your HubSpot MQL count has a known quirk (e.g., offline conversions lag by a day), you can add a note to the announcement template: 'HubSpot MQL count is as of [date]; expect +5-10% after offline sync.' That context lives in the template, not in your head.

We're not SOC 2 certified ourselves — does Starch's security posture matter for internal Slack announcements?

Worth knowing: Starch is not SOC 2 Type II certified today. For internal marketing announcements, most 120-person companies find that acceptable. If your InfoSec team requires SOC 2 Type II for any tool that touches HubSpot or ad platform credentials, that's a conversation to have before connecting those integrations.

Can we build different announcement templates for different channels — like one for #company-wide and a more detailed one for #marketing-team?

Yes. Describe the two formats to Starch and it builds two versions of the app — same data sources, different output structure. The #company-wide version might be three tight paragraphs; the #marketing-team version might include a channel-level breakdown of GA4 traffic and a line on which ad creative drove the most clicks. Fork the base app and describe what each version should look like.