← Studio
03
On my own
Cellopack
Launched · started with friendscellopack.xyz ↗
- Problem
- Collectors build reputation across eBay, Instagram, X, Reddit, Facebook groups, and card shows, but that reputation stays where it was earned. When a deal moves somewhere else, the buyer has to piece together whether the person on the other side is established or a scammer. One common workaround is a vouch thread, where past buyers reply to confirm successful deals. It works, but the reputation is still tied to one platform and the evidence behind it can be difficult to validate. There is no simple way for a collector to carry their identity, history, and reputation with them.
- Constraints
- The hardest part was proving that the accounts someone adds actually belong to them. Cellopack can bring reputation signals together from other platforms, but it does not control those platforms. Each one exposes different permissions: some support instant verification, some require approval, and some provide no usable access at all. That meant designing a trust model that could get stronger as integrations improved without depending on any one platform to work. I also built it solo and bootstrapped, so every product decision had to earn its complexity.
- Process
- I started with the smallest useful trust loop: create a profile, connect the places where you already collect or sell, verify ownership where possible, and gather vouches from people you have dealt with. Then I built the safeguards around those signals so reputation could not be created as easily as it could be claimed. Identity verification came later as an optional premium signal rather than something required to participate. Only after the trust loop worked did I focus on discovery, onboarding, and getting a new member to their first meaningful credibility signal.
- Outcome
- Cellopack is launched but early, code-complete and running in production, and it started with a small friends-and-family beta. Some accounts can be verified automatically; others still go through a manual review queue. The biggest remaining risk is no longer technical. It is whether collectors will trust a new product enough to put their reputation on it, and whether that profile gives someone on the other side of a deal enough value to care. I have built the product. I have not yet proven the market.
- Lessons
- I underestimated how committed people are to the process they already use. An established, pinned thread with a history of replies is real reputation, and moving off it costs more than I expected. So the opening isn’t convincing a seller to leave a system that already works. It’s reaching collectors earlier, as they start their first vouch thread or move to a new platform, before that commitment sets in.