We read the threads your readers argue in, find the question that keeps coming back, and write the piece around it.
Start a run$ peddler run "why teams abandon their design system" [01] mining demand412 threads [02] clustering6 clusters [03] convergence0.71 GO [04] keywords18 ranked [05] outline7 sections [06] writing1,840 words artifacts: html, markdown, seo meta, json-ld done in 9m 12s
Every run does the same six things in the open. You watch each one land, and every number it leans on stays on the page.
We read public discussion threads, video pages and open-web pages on your topic, and pull the whole conversation rather than the titles a keyword tool would show you.
Hundreds of scattered complaints collapse into the handful of questions people keep circling back to. That cluster becomes the piece, not your best guess.
The convergence figure, and the threads it was measured over. You see how much of the conversation actually circles your topic before a word gets written.
Keywords come out of the words real people used, ranked by how often they turned up next to real intent. No volume tool, no wishful padding.
Headline options and a section-by-section outline land before the draft does, so you can overrule the direction while it is still cheap to change.
The full draft in your voice, plus HTML, Markdown, SEO meta and JSON-LD, every claim still linked to the thread it came from.
Runs are metered in credits, and you only pay for the ones that finish. A run that fails costs you nothing, we absorb it. Plans are monthly subscriptions in USD, billed by Paddle as merchant of record, and they renew until you cancel.
Nothing is held back on the small plan: the same demand mining, the same convergence call, the same publish-ready artifacts. Top up any time at 7,200 credits per dollar.
Choose Starter$12 per month, renews automatically until you cancel. Cancel any time from Billing.
About a piece a day if you want one. Built for the writer or small team publishing weekly, with upgrades that grant the full credit difference straight away.
Choose Growth$29 per month, renews automatically until you cancel. Cancel any time from Billing.
For teams publishing most days of the week. The same pipeline and the same evidence attached to every piece, with enough credits that the calendar stops being the constraint.
Choose Scale$99 per month, renews automatically until you cancel. Cancel any time from Billing.
Most content tools start with a blank box and a confident tone. Ours starts with the argument already happening: the thread nobody answered properly, the question asked four hundred times, the answer everyone half-knows. We read that first, put a number on how much of it circles the same problem, and only then write. You get a finished piece and the paper trail underneath it: the threads, the excerpts, the reasoning, all still linked.
Start a runContent Peddler is for people who publish under their own name and cannot afford to be wrong or dull. Three kinds of people keep coming back to it, and the run works the same way for all three.
See a sample run
You are the product, the support desk and the content team, and the blog is the first thing to slip. A run takes the topic you were already going to write about, checks whether anyone actually cares, and hands back a piece you would have needed a free Sunday to research. You still edit it. You just stop starting from nothing, and you stop publishing into silence.
The questions your users ask in public repeat, and they are usually not the ones your roadmap is answering. A run reads those threads, so the guide you publish addresses what people actually got stuck on. Every claim keeps a link back to the thread it came from, which makes review with engineers short.
A calendar to fill, no research budget, and a stakeholder who wants to know why this post and not another. Runs give you both halves: enough finished pieces to keep the calendar honest, and a demand figure per topic you can put in front of whoever signs off.
We kept writing posts that nobody had asked for. The research was the expensive part, the guessing was the demoralising part, and the tools on offer skipped straight past both. So we built the thing that does the reading first and shows its work, then writes.
Read the method
