How We Use HYR Studio to Market HYR Studio

By Aran P Ploshansky

Every social asset HYR publishes is generated from our own tool. Full behind-the-scenes: what goes in, what comes out, and what we changed because of it.

There's a simple test for any content tool: does the team that built it actually use it? At HYR, the answer isn't just yes — it's *exclusively*. Every carousel, quote card, thread, and newsletter we publish is generated through [HYR Studio](https://hyrstudio.com) from our own source material.

This post is the full behind-the-scenes.

## What goes in

Our raw material is unglamorous: founder voice notes, recorded product walkthroughs, customer calls (with permission), and the occasional long rant about generic AI content. We treat each one the way our customers treat a podcast episode — as [source material for a 30-day content batch](/blog/turn-one-podcast-episode-into-30-days-of-content), not a finished thing.

The rule we hold ourselves to: **never write a social post from scratch.** If an idea is worth posting, it's worth recording first. That forces the insight to exist in our real voice before any tooling touches it.

## What comes out

A typical 20-minute recording becomes:

- 2 carousels (framework-style breakdowns) - 4–6 quote cards (verbatim phrasing, our brand system) - 1 thread (the narrative version) - 1 newsletter section - 1 blog post — often the one you're reading

That batch covers roughly two weeks of publishing. Two recordings a month covers the calendar.

## What we've learned by dogfooding

**1. Voice preservation is a feature you notice by its absence.** Early internal builds occasionally "smoothed" our phrasing. The output was grammatically nicer and completely dead. We rebuilt the pipeline to bias toward verbatim extraction — if Aran said it weird, the quote card says it weird. Engagement went up when the polish went down.

**2. Batching changes your relationship with the calendar.** When content is generated one post at a time, you're always behind. When it arrives as [a full extraction batch](/blog/turn-one-podcast-episode-into-30-days-of-content), you're a curator instead of a producer. The job shifts from "what do I post today?" to "which of these 25 assets earns a slot?"

**3. The kill rate is the quality bar.** We cut roughly a third of every batch. That's healthy. A repurposing system should overproduce so that you can be ruthless. If you're publishing 100% of what a tool generates, your bar is the tool's bar, not yours.

**4. The blog post is the sleeper asset.** Social posts spike and vanish. The blog version of each recording keeps collecting search traffic months later. We now treat the blog post as the *primary* output and the social batch as its distribution.

## The honest part

Dogfooding also shows you the rough edges. Some recordings produce thin batches — usually because the source material was thin. The tool can extract your voice; it can't invent your insight. Garbage in, on-brand garbage out. That constraint shaped our onboarding more than any feature request.

If you want to see the system's output in the wild, you already are. This post started as a nine-minute voice memo.

**Turn your next episode into a content batch →** [hyrstudio.com](https://hyrstudio.com)