Nightmarks

Nightmarks is a SaaS product that maps and catalogs the world's recorded supernatural encounters. UFO sightings, ghost stories, cryptid reports, folklore legends — historically scattered across old forums, cryptid wikis, decades of out-of-print books, and regional ghost-hunting blogs. Nightmarks pulls them into a single structured, explorable place — a map, a library, and an editorial layer rolled together, with a paid membership tier on top.

Two real product problems had to be solved at the same time.

The data problem. Thousands of historical encounters, each with its own shape — dates, locations, witnesses, categories, credibility tiers — needed a schema that could hold them without flattening the details that make each one interesting. Map-able locations were essential, but so was honoring the fact that some encounters are documented to a street corner and others exist only as 'somewhere in Hertfordshire, 1678.'

The credibility problem. A database of supernatural encounters can't pretend everything in it is verified, and it can't pretend everything is fiction. A 2024 multi-witness UFO sighting with radar correlation is not the same artifact as a 17th-century woodcut about a devil-mowed field. Both deserve a place in the catalog. They don't deserve the same visual weight on the map, in search, or in editorial surfaces.

The build sat at the intersection of three different things: a content-heavy database, an interactive map, and a SaaS surface — accounts, paid tier, editorial layer. A few decisions shaped the outcome.

Map-first development. The explore view is the core experience, so I built it first and validated the data shape against it. Anything that couldn't render meaningfully on the map was a signal the schema needed rethinking before adding more entries.

Credibility tiers as a first-class filter. Verified, historical, user-reported, and unverified each get their own surface in the UI — as filters, as a date-range default, and as visual weight on the map itself. The UX argues that credibility is part of the data, not a footnote.

An editorial layer on top of the database. 'Nightmark of the Week,' curated regional collections, and a guides section give the product reasons to come back beyond a single search. The database is the substrate; the editorial layer is the habit loop.

A paid tier built around the data, not paywalled at the door. The free product is genuinely useful — full map, full database, full reports. The Explorer Pass adds advanced query tools and power-user features. Affiliate integrations for recommended gear and reading, plus curated partnerships with paranormal tour operators, round out the revenue legs.

The AI-accelerated workflow took typing speed and tedious implementation off the critical path. Architecture, schema, content model, and product calls were all still mine to make — those are the actual work. They could move at full speed because the bottleneck wasn't in the code, it was in the decisions.