How to write updates that make people want to help
Founder updates get help when they pair clear proof with one forwardable ask that makes the next action obvious.
Thesis: People do not help because they like you. They help when the ask is concrete enough to act on in two minutes and the progress is credible enough that forwarding it costs them no reputation. Most founder updates fail both tests at once.
You send the monthly update. It is well written. It has the metrics, the wins, the "asks" section at the bottom. Forty people receive it. Three reply "great progress!" and zero actually do anything. You assumed the update was the work. The update was only the invitation. The work was making it effortless to say yes, and you skipped it.
Here is the thing founders get backwards. They write the update to look impressive. They should write it to be easy. An impressive update makes the reader feel good about you for ten seconds. An easy update makes the reader take an action while they are still in their inbox, because the action is sitting right there, pre-built, costing them almost nothing. Impressiveness is for you. Easiness is for them. Help comes from easiness.
Why people don't act, even when they want to
Assume your reader genuinely likes you and wants you to win. They are still not going to help, and not because they are flaky. Three frictions stop them, and your update usually trips all three.
The first is effort. Your ask is "we'd love intros to anyone in fintech." That is not an ask, it is homework. Now the reader has to scan their memory, decide who fits, figure out how to frame you, write the intro, and hope it lands. You handed them a blank page. People do not do homework for acquaintances. They do homework for their own kids and their own deadlines.
The second is risk. When someone forwards you to a person they respect, they are spending a small amount of their own reputation. If you look unproven, vague, or like you might embarrass them, they keep their wallet shut. They will not say this. They will just go quiet. Every forward is a tiny bet on you, and a credible update is the only thing that shortens their odds.
The third is ambiguity. "Let me know if you can think of anyone" has no edge to grab. There is no clear action, no deadline, no specific name to react to. Ambiguous asks get filed under "later," and later never arrives. The reader needs exactly one obvious thing to do, not a menu.
The fix is not to ask for less. It is to do the helper's first three steps for them, inside the update, before they ever decide whether to help.
The proof-plus-ask combo
Every update that generates help carries two things welded together: a piece of proof and a specific ask the proof has already justified. Separately, each is weak. The proof alone is a brag. The ask alone is a cold favor. Together they form a single move: "here is a fact that makes me worth your reputation, and here is the small, exact thing to do about it."
Proof is not "we're growing fast." Proof is a concrete, verifiable fact that a stranger would find credible: a named customer who signed, a usage number that crossed a line, a person who joined, a revenue figure, a logo your reader will recognize. The test is whether your reader could repeat the fact to someone smarter than them without feeling exposed. "We doubled" fails. "We went from 4 to 9 paying design partners, including [recognizable name]" passes.
The ask is the part founders truncate. A real ask names the exact person or the exact type, and it carries the framing the helper would otherwise have to invent. The reader should be able to forward your words verbatim. If they have to write anything from scratch, you have lost most of them. You are not asking them to help. You are asking them to do unpaid copywriting under social pressure, and they will decline by doing nothing.
What a forwardable ask actually looks like
Here is the same ask, before and after, so the difference is unmistakable.
Before (homework):
Asks: We're always looking to connect with great fintech investors and potential customers. If anyone comes to mind, we'd love an intro!
This forces the reader to generate names, decide fit, and write the pitch. Most do none of it.
After (forwardable):
One specific ask this month: we're raising a seed round and would love a warm intro to [Partner Name] at [Fund], who led [comparable company]'s seed. If you know them, here's a forwardable blurb you can paste:
"Wanted to intro you to [Founder], who's building [one line]. They just signed [recognizable customer] as a paid design partner and grew paying accounts from 4 to 9 last quarter. They're raising a seed and I thought of you because of [comp]. Worth a quick look, happy to connect you both."
The reader does one thing: decide yes or no, then paste and hit send. You did the recall, the framing, and the writing. All that is left is the part only they can do, which is spending the reputation. That is the only part you cannot outsource, so make it the only part you ask for.
Notice the proof is inside the forwardable blurb, not just in the update body. The fact has to travel with the ask, because the person who receives the forward never saw your update.
The artifact: a forwardable-ask kit
Build these once, reuse them every month. Three components.
1. Social-proof snippet library. Keep a running list of facts that survive forwarding, each in one verifiable sentence. Refresh monthly. Examples of the shape:
| Proof type | Weak version | Forwardable version |
|---|---|---|
| Customer | "Great traction with customers" | "Signed [Recognizable Co] as a paid design partner in March" |
| Usage | "Engagement is up" | "Active teams went from 12 to 31 since January" |
| Revenue | "Revenue is growing" | "Crossed $18k MRR, up from $7k in Q4" |
| Team | "Building a strong team" | "[Name], ex-[Company] staff eng, joined as founding engineer" |
| Validation | "Investors are interested" | "[Angel Name] wrote our first $50k check last month" |
If you cannot fill a row with a real, checkable fact, leave it out. A missing proof point is honest. A vague one is noise, and a fake one ends the relationship the day it is caught.
2. The forwardable-ask block. A fill-in template you drop into every update:
One ask this month: [the single most important thing right now].
Specifically: an intro to [exact name or exact profile].
Why them: [one line so the helper trusts the fit].
Paste-ready blurb: "[2–3 sentence intro the helper can forward verbatim, with one proof fact inside it]"
One ask per update. Two asks split attention and you get neither. If you genuinely have a second, save it for next month or send it one-to-one to the one person who can serve it.
3. The helper-effort scorecard. Before you send, score your ask 0–2 on each line. Anything under 7 gets rewritten, not sent.
- Names a specific person or tight profile, not "anyone in X" (0–2)
- Carries a paste-ready blurb the helper can forward verbatim (0–2)
- Contains one verifiable proof fact inside the blurb itself (0–2)
- Has exactly one ask, not a menu (0–2)
- Could be acted on in under two minutes without writing anything new (0–2)
The scorecard is the whole discipline in five lines. It moves the question from "is my update good" to "is helping me easy," which is the only question that predicts whether anyone moves.
Match the ask to the relationship
Do not send the same ask to all forty people. The helper's willingness to spend reputation scales with how well they know you, and your ask should scale with it.
A close backer who has already put money in can take a harder ask: a direct intro to a specific partner, a reference call, a forward to their whole network. They have skin in the game and a forward from them carries weight. A warm-but-light contact, someone who follows along but has no stake, gets a softer ask: "if [specific name] ever comes up naturally, I'd value the intro," with the blurb still attached so it stays effortless. A near-stranger on the list gets no ask at all, just the proof, so that next month they are warmer and the ask lands on prepared ground.
The general update goes to everyone. The specific, named ask goes to the handful who can actually serve it. Founders blast one ask to the whole list and wonder why it feels like shouting into a void. The void is the mismatch between the size of the ask and the size of the relationship.
Where this gets hard at scale
The system above is simple for one update to one person. It breaks when you are running a live round with forty backers, each owing a different ask, each holding a different relationship, each connected to a different slice of investors you want to reach. You cannot remember who is close enough for a hard ask, who already forwarded you to whom, or which proof fact answers which person's earlier doubt. So you default to the lazy broadcast, and the lazy broadcast is the thing that does not work.
This is the part RoundOS is built to carry. It holds the context your round already generates, your backers, the investors you are targeting, who knows whom, and what each person last heard from you, then maps the warm paths between a backer and a target and drafts the helper-specific update: the right proof fact for that reader, the named ask their network can actually serve, and the paste-ready blurb already written. You are choosing whom to ask and approving the words, instead of reconstructing forty relationships from memory at 11pm.
Make the ask easy to forward.
Take your last update and run the one ask in it through the five-line scorecard. If it scores under 7, rewrite it into a single named ask with a paste-ready blurb and send it to the three people who can actually serve it. If you want the helper-specific version drafted for each backer from the context you already have, [generate helper-specific update variants in RoundOS](#).