Zen Arbitrage: From Dumpster Diving to $90K/Month — a Book-Flipping SaaS, a 3-Year Plateau, and a Sale to a $66M Roll-Up

It starts with secondhand books pulled from a dumpster in high school, and ends with a SaaS earning $90,000 a month (about ¥13.5M), grown over seven years, and sold to a roll-up company that had raised $66M. Peter Valley’s story is a fun one on its own. But the real value of this case lies in the fact that he openly shares the failures of the sale process, a miscalculated churn rate that cut the price, ignorance of his own traffic sources, and doubts about his own indispensability. For any indie developer who might one day sell a SaaS, this reads not as a textbook but as an actual, rare-to-find failure log.

The Business’s Journey

From Dumpster to SaaS — Where the $14,000 in Development Money Came From

Valley’s origin story starts in high school, finding thousands of books in a large dumpster outside a thrift store, an episode that ends with him locked in by an employee and the police being called. He’s been a used-book flipper for 20 years since. The turning point was learning how to “repackage knowledge”: he learned the basics of business from CD courses by internet marketers, turned his own flipping know-how into an ebook earning $100 a day, then expanded into video courses. In 2015, he put about $14,000 from this information-product income into hiring a developer, introduced through a friend, to build Zen Arbitrage. Zero management experience, zero outside funding, zero technical background. That’s the state he was starting from, in his own words.

The launch was a pre-sale aimed at the audience he already had, his readers and students. 100 slots, limited, at $97/month. It sold out in 90 seconds. That initial burst only happened because the audience existed before the product did, and it’s also why the $14,000 development cost was recovered almost immediately.

Stuck at $30K for Three Years — the Real Cause Was Churn

After hitting $30,000 in monthly revenue within six months, the numbers barely moved for about three years. New customer acquisition hadn’t stalled, the outflow simply matched the inflow. Valley admits he “didn’t understand customer retention well enough.” His target audience joined expecting to “make money at the click of a button,” and left disappointed by the actual amount of work required, a mismatch in expectations more than a problem with the tool itself.

The fixes he applied to break the plateau were mundane: introducing a 14-day free trial, beefing up support with Intercom, improving onboarding, and building a Facebook group where users could learn from each other. The team grew gradually too, around 2019 he added a support hire and two developers, and hired a manager for affiliate marketing. Even so, at the time of sale the team was still just 1 employee plus 2 part-time developers, running $90,000 in monthly revenue, 750 paying members, and $1.1M in annual revenue on that minimal headcount.

Three Holes Exposed During Due Diligence

The sale went through Quiet Light Brokerage, with Carbon6, a company buying up Amazon-seller tools, as the buyer. But the negotiation didn’t go unscathed. Valley reveals three failures:

• A miscalculated churn rate: he’d been conflating involuntary churn (like expired cards) with voluntary churn in his reporting. Buyers care about voluntary churn specifically, and when this surfaced during due diligence, it led to a significant cut from the initial offer (the final sale price is undisclosed)
• Blind to traffic sources: “I knew it was organic, but I hadn’t really nailed down where the traffic was actually coming from”
• Guru dependence: the business relied on Valley’s own personal brand (a well-known figure in the book-flipping world), a structure he himself admits raised buyer doubts about “whether the marketing would still work once the founder was gone”

Looking back on the whole sale process, he calls it “a steep learning curve. There were a lot of mistakes I could have avoided with the right mentor or more experienced advice.” His own summary: “The business had far more potential than I was allowing it. My inexperience, my poor management of technical talent, and my own bandwidth were putting a lid on growth.”

Reading Between the Numbers

A churn-definition error directly cuts the sale price. Separating involuntary from voluntary churn, tracking retention by cohort, Usersnap’s “keep the data room always current” is precisely the preventive measure for this failure. Getting your numbers in order is worthwhile even with no plan to sell, as a way of staying “sellable at any moment.”

The three-year plateau was a churn ceiling. New customer acquisition was working, but the outflow matched it. In recurring-revenue businesses, stagnation is more often a retention problem than an acquisition problem. That investing in community helped break the plateau is the same mechanism as Nomad List’s “community prevents churn”.

The “pickaxe for book flippers” got bought up as part of a market-wide land grab. Carbon6 acquired 16 companies in the 16 months starting in May 2021, and as another sale to the same Carbon6, Order Tagger, shows, tools inside the Amazon economy are prime targets for roll-up acquirers. If a solo founder lands a hit tool inside an ecosystem, the roll-up companies operating in that ecosystem belong on their exit list.

The Limits of This Case

With the final sale price undisclosed, we can’t verify the valuation multiple against $90K in monthly revenue, and the size of the “discount” remains a matter of qualitative testimony. Also, Zen Arbitrage’s early momentum rests on an audience built over nearly a decade of content publishing, pulling out the “$14,000 in development cost, sold out in 90 seconds” figure on its own would badly misjudge how reproducible this is. A personal brand is the strongest asset during the launch phase and the biggest discount factor at sale. That duality is the skeleton of this whole case.

Conditions for Reproducing This in Japan

What generalizes is the sequence: build audience understanding and capital first through content publishing, then build a tool via outsourced development that automates that audience’s work. The same shape could work in Japan’s resale/e-commerce circles too, and the P&L on the player’s side can be checked in a first-hand account of reselling in Japan. On the other hand, an exit environment where a roll-up company like Carbon6 makes frequent acquisitions is specific to the US Amazon economy, and there’s no guarantee a comparable Japanese tool would find the same exit waiting. This is a record best read with the full seven-year timeline in mind, including the three years of plateau, not just the final headline.

Related Cases

• Radius — a vertical SaaS that likewise pushed through a long plateau
• Usersnap — data hygiene as a preventive measure
• A first-hand reselling account (Japan) — the P&L structure on the player’s side