Agentwork
D

Draft your monthly investor update

Pull real metrics from Stripe, PostHog and Linear and draft the update for your approval.

Operations Saves 3h/month Scheduled

Automated investor updates solve a specific failure mode: the update is late because assembling the numbers is tedious, and the numbers are approximate because they were assembled at 11pm. This workflow pulls the real figures from Stripe, PostHog and Linear on the first of the month, drafts the update in your voice, and hands it to you for the part only a founder can write.

What does this workflow do?

On schedule, the agent gathers the month's facts: revenue, growth and churn from Stripe, active usage from PostHog, shipped features from Linear, hiring changes, and highlights it finds in your team's Slack (a big customer win, a notable press mention).

It drafts the update in your established format: metrics table with month-over-month movement, shipped and upcoming, wins, lowlights, and asks. The draft arrives with every number sourced. You edit the narrative, approve, and the workflow sends it through Gmail to your investor list or hands you the final doc.

How does it work?

  1. The month closes and the workflow runs. All metric definitions are yours: how you count MRR, active users and churn is configured once and applied consistently.
  2. The agent assembles the numbers. Stripe for revenue and churn, PostHog for usage, Linear for shipped work. Every figure links to its source.
  3. It drafts the narrative. Grounded in what happened: the metrics, the ships, the notable events. Written in the voice of your past updates, which it has learned.
  4. It asks you for the human parts. The agent sends you two or three targeted questions ("What's the ask this month?", "Anything to say about the churn uptick?") so the draft reaches you without blanks.
  5. You approve, it sends. Final review in Google Docs, then delivery by Gmail to your list, or you send it yourself.

Why automate the assembly and not the judgment?

Investors read updates for the founder's thinking. The metrics table, though, is pure assembly work, and it's where errors creep in when it's built by hand from three dashboards. The split this workflow makes: machines assemble and verify the facts, you supply the judgment, and an approval gate sits between the draft and your investors' inboxes.

Consistency compounds too. Because the same definitions run every month, your update never has the "wait, is this the same churn number as last quarter?" problem. When a number moves in a way the agent can't explain from the data, it asks the person who'd know before the draft reaches you.

Works with

Stripe, PostHog, Linear, Gmail, Google Docs, Slack. Runs monthly; a quarterly board-pack variant uses the same sources.

StripePostHogLinearGmailGoogle DocsSlack

Frequently asked questions

Which metrics can it pull automatically?

Revenue, MRR movement, churn and new customers from Stripe; active users, retention and feature usage from PostHog; shipped issues and cycle progress from Linear. Custom metrics work if the source is connected.

Can I keep my existing update format?

Yes. Give it two or three past updates and it adopts your structure and tone. Edits you make to drafts carry forward to future months.

Does anything send without my approval?

No. The draft always stops for your review. Sending is a separate explicit step, and you can keep sending manual entirely.

How does it handle a bad month?

It reports the real numbers and flags the movement, and it asks you how you want to frame it rather than spinning it for you. Honest updates are the point.

Can it also do board packs?

The same workflow extends to a quarterly variant with deeper sections. Most teams run monthly updates and reuse the accumulated material each quarter.