Phez

Shanti, a 38-year-old Portland-based software developer, had spent years building side projects using Ruby on Rails. In summer 2015, he started Phez—a play on the word "fez," which he thought sounded funny. The core idea was simple yet ambitious: build a Reddit alternative that paid users in Bitcoin for quality content, with a strong emphasis on free speech. Shanti believed Reddit had too much control over discourse, silencing communities arbitrarily, and he wanted to create an alternative where the Overton window wasn't a constraint.

Shanti built the initial prototype over "a few long weekends" using Ruby on Rails, his framework of choice since its early days. He deployed on Amazon AWS with PostgreSQL, nginx, and unicorn, keeping AWS costs minimal at around $20/month. The MVP included standard Reddit features: posts, communities, voting, and comments. Shanti was transparent about his limitations: "The initial product looked terrible! Just like a boilerplate Bootstrap site." As a backend developer, he struggled with responsive front-end design across desktop and mobile. Despite these challenges, he kept iterating for several months, even building a full API for a mobile app that never materialized.

Shanti had no formal marketing strategy. He simply posted about Phez on relevant subreddits and a few other social networks. The payout structure was designed as a pyramid: the top contributor received ~$25 in Bitcoin, with ~$100 split among the next 15 contributors. He allocated roughly 1.5 Bitcoin (~$390 at the time) for monthly rewards. Initially, the market responded with skepticism—people weren't sure it was a scam—but when payments actually went out, word spread that you could earn $10–$25 per month as a top contributor.

The cryptocurrency reward proved to be a strategic mistake. Rather than attract quality contributors, it incentivized low-effort gaming and spam. "99% of posts/comments were made with just the idea of doing the bare minimum to receive a Bitcoin micropayment," Shanti recalls. Every few weeks, new spam techniques emerged—gibberish posts and comments designed to game the karma system. More fundamentally, introducing monetary rewards seemed to *reduce* intrinsic motivation; people contributed less than they would for imaginary internet points alone. Shanti monitored traffic closely, but it never grew. He had envisioned scaling to a Reddit-size audience, then monetizing via ads, premium subscriptions, and in-house sponsorships—the classic social network playbook. Instead, he got stagnation.

After several months, Shanti announced Phez's closure and shut it down. He estimates he invested over 100 hours into development. Financially, the direct loss was minimal (~$390 in Bitcoin payouts + ~$240 in AWS costs). But the real loss was opportunity cost. The 1.5 Bitcoin he paid out would have been worth ~$29,014 if cashed at Bitcoin's December 2017 peak of $19,343/BTC. In hindsight, Shanti acknowledges that a similar idea—Steemit, a blockchain-backed social network with its own cryptocurrency—became viable. But Steemit had a full team and actual cryptocurrency infrastructure; Shanti was a solo web programmer. He later launched Timewalk Ventures, attempting to do "Y Combinator for smaller side projects," and has become a vocal advocate for learning from startup failures.