Arvid Kahl: How He Built a One-Person Business

by TFE Editor · 13 min read
Arvid Kahl: How He Built a One-Person Business

Arvid Kahl is a software engineer, author, and podcaster who co-founded FeedbackPanda with Danielle Simpson and sold it in 2019. He later founded Podscan, took a six-figure investment he called bootstrapper-compatible funding, and wrote that Audiohook acquired it in 2026. He matters to a solopreneur because he published the record, including the co-founder and the money.

A solopreneur starts and runs a business without a standing staff. FeedbackPanda was two people, and Podscan later had an investor and sales help, so those names belong on the page. A freestyle entrepreneur builds the business around a life and still has to get paid. Every dollar below is dated and attributed. The figures are his, or taken from a page fetched on November 11, 2026, and they are unaudited.

Arvid Kahl at a glance

Item What the sources support
Known for FeedbackPanda, with Danielle Simpson. Podscan. Zero to Sold. The Embedded Entrepreneur. Find your Following. The Bootstrapped Founder
Model A teacher-feedback SaaS sold in 2019 with no employees and no outside money. Then a podcast-intelligence SaaS with a 2024 SAFE, sold to Audiohook in 2026
Team FeedbackPanda: Arvid Kahl and Danielle Simpson, and no employees. Podscan: an unnamed sales helper and “the team,” then Audiohook. The Calm Company Fund is an investor, not staff
Notable work The two books, the free Zero to Sold guide, Find your Following (a Twitter course on its own site), the podcast, and the two products. He also names Permanent.link and Podline.fm
Where to follow The About page links the podcast and X. The footer on that page links GitHub and the book and course sites. On November 11, 2026 the newsletter archive said the newsletter was paused and there were no email signups

The story, from a weekend prototype to two sales

One unpaid chore, then a company of two

On November 9, 2019 he published the FeedbackPanda story. In 2017, he wrote, Danielle was teaching English as a second language online to Chinese children, in 12-hour days, while he worked as a software engineer for a German company in the IoT space. After teaching she would work two extra unpaid hours writing student feedback in Excel sheets and Word documents.

I started working on a prototype on the weekends.

Arvid Kahl, November 9, 2019. They called it FeedbackPanda. Two years later they sold it to SureSwift Capital for what he calls a life-changing amount of money. That post prints no price. It does say they never hired an employee, never raised money, got profitable within a few weeks after they made it public, and had over five thousand customers. Marketing ran almost exclusively through the teachers Danielle had gathered. He uses lifestyle business for the version you keep, and says they still made the day-to-day work replaceable.

We never hired a single employee.

We never raised any money either.

The free Zero to Sold guide, fetched November 11, 2026, calls itself a living document and has no publication date. He built the first system in a week or two. Danielle was at the helm of product and design. He managed the technical and infrastructural work. They shared customer service. A Facebook comment from her, with the link, is when he says teachers started asking. The guide’s hours differ: 10-hour days, and almost two hours of feedback. The guide says Chinese students. The November 9 post says Chinese children.

FeedbackPanda was just Danielle and me.

We sold FeedbackPanda in mid-2019, just under two years after founding the company.

Arvid Kahl, in that guide. Monthly recurring revenue had just reached $55.000 at the sale. That is his punctuation for $55,000. He still gives no sale price. The About page names co-founder Danielle Simpson, dates the sale to 2019, and says their MicroConf Europe 2019 talk came just after. The same guide says he was in businesses where they never finished building the product, co-founded startups that fizzled because they did not know how to market the service, and was involved in companies that did not find a way to monetize popular products, then pivoted. He does not name them.

Books, a show, and a paused list

The About page calls him a software engineer, bootstrapped founder, author, and podcaster. The books are Zero to Sold and The Embedded Entrepreneur. The course is Find your Following. The course site calls it a Twitter course. The guide says it was the inspiration for the book, which is four times as extensive and was professionally edited. His bookshelf lists Company of One by Paul Jarvis. The Growth Reads review is Company of One. Post, article, and page sitemaps checked the same day had no review of his own books. The newsletter archive said the newsletter was paused and there were no email signups. More founder pages are on the solopreneurs hub.

A SAFE he still called bootstrapping

The About page says he later founded and bootstrapped Podscan.fm, a podcast intelligence platform. The April 5, 2024 post is where outside money shows up. Two months earlier he had started tinkering. A wire from the Calm Company Fund had just arrived. He calls the amount six figures and does not print the exact check.

In the April 5 post he asks whether this can still be called bootstrapping, and writes that he believes so. He was already invested in the fund. The deal, as he describes it, is a SAFE with a side letter. If an event touches or converts equity, the fund gets its ratio of shares.

Until then, shares are fully owned by me.

If I never sell, they never own any part of the business.

Arvid Kahl, April 5, 2024. After a certain time, he wrote, dividends he pulls out send “somewhere around ten percent-ish” to the fund, and he can pay himself a six-figure salary before that. A substantially similar side project in the first year or so would belong to Podscan. The podcast, blog, and newsletter were left out. He opened a Delaware C-Corp and called himself a Canadian resident. That is his account of his own setup, general information, not professional advice. He names Permanent.link (also written PermanentLink) and Podline.fm as products where he would not have known where to put a six-figure check. The spend was for GPUs. He was nearing almost 2 million hour-long episodes ingested. The daily new-episode count he wrote was 10.000 (10,000).

On May 3, 2024 spend so far was around $10.000, all GPU compute. That figure is money he said he had already spent. He wrote that the ramp would likely cost around $15,000 to $20,000 a month, and that after the backlog infrastructure would cost no more than $2,000 a month. Those two lines are forecasts.

I’ve just crossed the $1000 Monthly Recurring Revenue mark last week.

He wrote that churn was zero percent, retention from month one to month two was 100 percent, and he was adding one or two subscriptions a day. Over 20 customer conversations. He called them calls when he wrote about recording them. He named Otter.ai and Grain for post-processing those calls. Cold outreach, personal emails to podcast booking agencies, he wrote, “Didn’t go far.” Design and UX help, he wrote, would be project-based and had not started. He wrote that he answered customer messages himself.

Two months of profit, then Audiohook

On April 4, 2025, almost a year after the funding, he wrote that Podscan was profitable. That day’s database was 27 million transcribed episodes and 3.7 million podcasts, a different count from April 2024. The next goal was $15,000 to $20,000 in monthly recurring revenue, so he could pay infrastructure, business expenses, and an engineer. He wrote that he had to hit $15,000 first. A serious seven-figure offer could be emailed to him. That is an invitation, not a price. He would stay solo for a couple of months, which he called the default.

On June 20, 2025 one major customer had churned.

And it was—but only for two months.

Around $10,000 a month in expenses, somewhere around $6,000 in monthly recurring revenue, $4,000 short of breaking even. The same post also says expenses fell from $30,000 to just under $10,000, and it describes $10,000 plus labor, $6,000 a month in revenue, and over 100 customers. An unnamed person was setting up a sales pipeline, outreach, and demo booking. The old $500 plan became the second tier. The top tier was $2,500 a month. He wrote that 50,000 new podcast episodes come out every day. He says “solo founder” and “the team” in the same post.

The About page says that in 2026 Audiohook acquired Podscan and that he joined to keep working on it. Audiohook’s press page, fetched November 11, 2026, has no dateline and no price. It says Podscan ingests more than 70,000 podcast episodes every day, and that he oversees the technical transition. A Podnews copy dated September 30, 2026, datelined Ogden, Utah, via Audiohook, says the deal closed August 24, 2026. No fetched page prints a sale price. His daily new-episode line is 10.000 on April 5, 2024 and 50,000 on June 20, 2025. The 70,000 figure is Audiohook’s ingest line, not a third reading of that same count. No monthly revenue dollar after June 20, 2025 was in the pages fetched for this check, including the February 13, 2026 post.

The playbook: five moves he actually dated

Our words, tied to the record above, and to the difference in solopreneur vs entrepreneur. The other names stay attached.

1. Start from a chore you can time

The 2019 post starts with hours he could see, then a weekend prototype. The guide’s build took a week or two. He stayed on student feedback until that niche could pay two people.

2. Put every other name on the page

Danielle Simpson is on the About page. The 2019 post says no employees and no outside money. April 5, 2024 names the fund. June 20, 2025 names a helper and a team. The 2026 release names Audiohook.

3. Write the month the number breaks

April 4, 2025 says profitable. June 20, 2025 says that lasted two months, then one customer left. Copy the second post.

4. Mark a forecast so it stays a forecast

The May 3, 2024 line of about $15,000 to $20,000 a month, and the April 4, 2025 revenue goal, are forecasts. The $10.000 GPU figure that day is spend he said was already done. $55.000 in the guide is monthly recurring revenue at the FeedbackPanda sale, in his punctuation, and neither sale has a price here.

5. Keep a list you still send

He built the blog, the podcast, the course, and a newsletter. On November 11, 2026 that newsletter was paused. We like Kit for this: it’s free up to 10,000 subscribers. Partner link · T&Cs apply. Kit’s help center, updated September 23, 2026 and fetched November 11, 2026, says the Free plan is $0 a month under 10,000 subscribers, with unlimited broadcasts. Sequences and visual automations are on paid plans. Our recommendation. Nothing fetched here says he uses Kit.

What did not work, and the caveats

  • No sale price in these pages. FeedbackPanda is a life-changing amount. Podscan’s price is absent from the About page, the Audiohook page, and the September 30, 2026 release.
  • The hours disagree. November 9, 2019 says 12-hour days, two extra hours, and Chinese children. The guide says 10-hour days, almost two hours, and Chinese students.
  • Bootstrapped, and also a SAFE. The About page says he bootstrapped Podscan. April 5, 2024 describes a six-figure SAFE. Shares stay his until an event touches or converts equity. After a certain time, dividends send somewhere around ten percent-ish to the fund, after a six-figure salary. FeedbackPanda, he wrote, raised nothing.
  • Solo has a roster. Danielle Simpson. Then the fund, a design-help plan, an unnamed sales helper, “the team,” and Audiohook.
  • Profit was a two-month window. June 20, 2025: somewhere around $6,000 in monthly recurring revenue, around $10,000 in expenses, also “just under $10,000” and “$10,000 plus labor,” and $6,000 a month in revenue in the handoff sentence, over 100 customers.
  • The daily counts are not one series. April 5, 2024: 10.000 new episodes a day, in his punctuation. June 20, 2025: 50,000 new episodes coming out every day. Audiohook: Podscan ingests more than 70,000 podcast episodes every day.
  • May 3, 2024 is an early month. $1,000 of monthly recurring revenue and zero percent churn are not the June 2025 figures. Cold outreach had not gone far. The newsletter was paused on the archive page fetched November 11, 2026. The August 24, 2026 close is on the Podnews copy. The Audiohook page does not print that date.

The AI version of this playbook

This week is ours. On May 3, 2024 he named Otter.ai and Grain for post-processing customer calls. On April 4, 2025 he wrote that where APIs might change, he uses AI systems to extract the data. Those are dated examples. You decide the offer. A tool drafts. You check it against a page you opened.

Job You decide AI drafts You check
The chore One unpaid task you have watched someone repeat A short description of a tool that does only that task The hours match what that person actually does
The names Who else owns, funds, or sells An About line that includes every name The page shows the co-founder, the fund, or the helper
The number One month, and whether the line is revenue, cost, or a forecast A sentence that uses only figures you typed A forecast cannot be read as money already received
The call The questions you will ask Notes from a transcription tool The notes match what the person said
The list What you send, and the day A draft from your outline You send it. A paused list is labeled paused

A first week, if you already know the chore. This plan is our analysis.

  1. Day 1. Write the chore in the other person’s words, with their hour count.
  2. Day 2. List every name that belongs on an About page.
  3. Day 3. Time one step. Stop when the clock you set is done.
  4. Day 4. Send one note where those people already talk.
  5. Day 5. Put one price on a page and date it.
  6. Day 6. Open a list you will send. Write the first note yourself.
  7. Day 7. Publish the month: money in, money out, and anyone who left.

Lessons for your one-person business

  • If a second person co-founded it, put their name in the sentence. Danielle Simpson is in his.
  • Keep revenue, expenses, and a sale price on different lines. He printed the first two on June 20, 2025.
  • Publish the month the number breaks. Two months of profit, then a lost customer, is the Podscan chapter to copy.
  • If you take money and still say bootstrapped, publish the instrument. He did that on April 5, 2024, and left the exact check off the page.

FAQ

Who is Arvid Kahl?

A software engineer, bootstrapped founder, author, and podcaster, on his About page. He co-founded FeedbackPanda with Danielle Simpson, sold it in 2019, founded Podscan, and wrote that Audiohook acquired Podscan in 2026. The April 5, 2024 post calls him a Canadian resident. Ogden, Utah, on the September 30, 2026 release, is Audiohook’s dateline.

What is a bootstrap founder?

On his pages, someone who builds without handing a voting investor the company. FeedbackPanda, he wrote, never raised money. On April 5, 2024 he still used the word after a six-figure SAFE, because the shares stayed his unless an equity event converted them.

Did he build FeedbackPanda alone?

He co-founded it with Danielle Simpson. The November 9, 2019 post and the guide say they hired no employees. The guide says she was at the helm of product and design, he managed the technical and infrastructural work, and they shared customer service. The November 9 post says the marketing ran almost exclusively through the teachers she had gathered.

How much money has he reported?

His numbers, unaudited. The guide: $55.000 of monthly recurring revenue at the FeedbackPanda sale, meaning $55,000, and no sale price. November 9, 2019: over five thousand customers and a life-changing amount with no dollars. April 5, 2024: a six-figure SAFE, amount unstated. May 3, 2024: about $10.000 already spent on GPUs, a forecast of about $15,000 to $20,000 a month during the ramp, and $1,000 of monthly recurring revenue. April 4, 2025: profitable, with a goal of $15,000 to $20,000 of monthly recurring revenue. June 20, 2025: somewhere around $6,000 of monthly recurring revenue against around $10,000 of expenses. No Podscan sale price was in the pages fetched here.

What happened to Podscan?

He founded it, took the April 5, 2024 funding, and on June 20, 2025 wrote that a short run of profit had already ended. His About page says Audiohook acquired it in 2026. The Podnews copy dates the announcement September 30, 2026 and the closing August 24, 2026. The price is not there.

The takeaway

Copy the public record: one chore, the other names, a number with a month on it, and a second post when the number breaks.

Grab the free One-Person Business AI Stack checklist and use it to staff that week: one chore, the other names, and one number you can still defend yourself.

Sources

Fetched November 11, 2026, unless the date is the page’s own.