Public posts on X, read for you

They already said it. Out loud.

We find the people posting the exact problem your product solves on X — and hand you the post.

We are opening in small batches. Add your product URL after joining and it is first in the scan queue — your first report lands on day 7 after your invite opens.

@mira_designsScore87

“Honeybook just raised prices again. $39/mo to send invoices and track 4 clients is insane. What are you all using?”

Why — Direct pricing complaint about a competitor and actively asking for alternatives — a buying-intent signal, not a vague gripe.

Jul 18X
@studio_marloScore79

“Dubsado has been sitting half-configured for three weeks. I just want to send an invoice and move on.”

Jul 20X
@davey.buildsScore74

“Lost a client’s feedback in a 200-message email thread again. There has to be a better way to run a 1-person studio.”

Why — Names the exact pain — feedback lost in email — and self-identifies as a solo studio, your core ICP.

Jul 19X
@nora.typeScore68

“Anyone else keep a client’s revision notes in six different places? Figma comments, email, DMs, a Notion doc.”

Jul 19X

Not another lead list. Real people, with proof.

Every name arrives attached to the public post that earned it — so you can open it, read it, and judge for yourself.

Step 01 — Analysis

Paste a URL. Skip the keyword setup.

Opus reads your product page, works out who it is for and what those people complain about, then writes the search conditions itself. Refold got 12 of them, in minutes.

  • No keyword lists to maintain
  • You can review and edit every condition before the scan starts

“Freelance designers juggling 3+ clients who track projects in Notion or spreadsheets.”

Analysisrefold.app
Ideal customer
Freelance designers juggling 3+ clients who track projects in Notion or spreadsheets.
Pain
Losing track of client feedback, unpaid invoices, and scattered files.
Search conditions · 12
honeybook too expensivenotion client trackerfreelance CRM recommendation+9 more
Step 02 — Weekly report

A shortlist you can read in one sitting.

3 curated reports per 30-day cycle, not a firehose. Each candidate comes with the original post, the date, a score and a plain-English reason they were picked.

  • Score 60–100 is promising and shown first
  • Save, mark contacted or dismiss — the report keeps your state

“Honeybook just raised prices again. $39/mo to send invoices and track 4 clients is insane. What are you all using?”

Report 3Week of Jul 15–21
@mira_designs87

“Honeybook just raised prices again. $39/mo to send invoices and track 4 clients is insane. What are you all using?”

Why we picked them — Direct pricing complaint about a competitor and actively asking for alternatives — a buying-intent signal, not a vague gripe. Names a 4-client workload, which matches the ICP.
@studio_marlo79

“Dubsado has been sitting half-configured for three weeks. I just want to send an invoice and move on.”

@davey.builds74

“Lost a client’s feedback in a 200-message email thread again. There has to be a better way to run a 1-person studio.”

Near misses

We show you the ones we almost picked.

Scores between 35 and 59 land in Near miss, and we name the exact condition they failed. You get our reasoning, not just our wins — sometimes the near miss is the one you want.

  • Every near miss states the missing signal
  • Below 35 we do not show it at all

“Lettering artist who complains about invoicing, but appears to work with only one client at a time.”

Near miss35 of 96 candidates
@lena_letters52

Lettering artist who complains about invoicing, but appears to work with only one client at a time.

Missing: Multiple-client workload signal
@ravi.makes58

Strong invoicing pain, but nothing in the post confirms a multi-client workload.

Missing: Multiple-client workload signal

Where your customers already are.

We do not build an audience for you. We read where people already say what is going wrong, out loud and in public.

X

Public posts and replies where people name the tool that just let them down — and ask the timeline what to switch to.

Example candidates: @mira_designs · @davey.builds

Public posts only. Every candidate links straight back to the post it came from. Reddit and more sources come after the beta.

One plan for the beta.

We open with a single one-time plan: a full 30-day cycle with 3 weekly reports. No subscription — pay once, read your reports, decide for yourself.

First 10+Beta
$9one-time

A single cycle to test the water on one product.

Join the waitlist
  • Posts scanned 750
  • Typical candidates 12–15
  • Weekly reports 3
  • Suggested approaches Included

Candidate counts are estimates based on typical results — not a guarantee. One product URL per cycle. Waitlist members are invited first; larger plans come after the beta.

Questions, answered plainly.

When can I get in?
We are opening in small batches so every early report gets human attention. Joining the waitlist takes an email address; adding your product URL puts you in the scan queue, so the moment your invite opens the analysis is ready to run — not starting from zero.
How is this different from a lead list?
A lead list gives you companies and job titles. WarmList gives you individuals and the public post that shows they have the problem right now — with the date, the platform and a link to the original. You are not guessing whether they care; you can read them saying it.
Do you post, reply or DM for me?
No. Auto-posting is not a setting we offer — it does not exist to turn on. WarmList sends 0 messages on your behalf. Each candidate comes with a suggested approach — for example, Reply publicly with a genuine recommendation. No pitch — share what you’d use and offer to help. — and you decide whether and how to act.
Do I need to pick keywords?
No. You paste your product URL and Opus derives the search conditions itself — 12 of them for refold.app. You can review and adjust them before the first scan, but nothing is required from you to get started.
How many candidates will I get?
It depends on how much your audience posts. On the beta plan a typical cycle surfaces 12–15 candidates. That is an estimate, not a guarantee — we would rather send you a short honest list than pad it.
What happens to the candidates who nearly qualified?
You see them. Anything scoring 35–59 appears in Near miss with the exact condition it failed, such as Missing: Multiple-client workload signal. Below 35 we leave it out entirely rather than waste your time.
When do the reports arrive?
The analysis runs the moment your invite opens. Posts are fetched on days 3, 7, 10, 14, 17 and 21, and reports are published on days 7, 14 and 21. Reports live on screen; email is just a short note that a new one is ready.
Can I cancel?
There is nothing to cancel. The beta is a single one-time cycle — you pay once, the First 10+ cycle runs its 30 days with every report included, and it simply ends. No subscription starts unless you choose to run another cycle.

The post is already up. Be there when we open.