SITE-202
Six days of campaign before a single feature shipped. Then 4,000 messages on the first day.
Site-202 "New World" was an SCP roleplay community I founded and owned. I ran development, wrote the lore, built the roleplay and community architecture, and directed the pre-launch campaign — publishing seventeen numbered dev logs along the way.
Most servers open the doors and then hope. Site-202 inverted it: the audience was built before the product was. Goals and trailer went up on day one of development — before channels, before roles, before lore.
By launch there were fifty people already waiting, and they had been watching it get built for six days.
Build in private, then announce.
A finished server drops into an empty room. The first members arrive to a place with no history, no rhythm and nobody to talk to.
Publish the promise first.
Goals and trailer channels shipped before any feature. Then six days of teasers, dev logs and reaction polls, with an invite-driven reason to bring people in.
4,000+ messages, day one.
Fifty members waiting at launch, boost level 2 reached the same day, and a hundred members inside the first week.
A launch is a campaign, not a date.
Every entry below is taken from the server's own dated announcement and dev-log record.
Publish the promise, not the product
Development opens with the goals and trailer channels — what the project is for, who it is for, and what it refuses to be. No roleplay channels, no roles, no lore yet. Anyone arriving can read the commitment before there is anything to join.
Community shell, then security
Hubs, self-roles and sub-community forums go up. Raid, spam and ping protection are built in the same pass — before the first member, not after the first incident.
First teasers
Forum structure, the nine-point ruleset and the character guide are teased publicly. Reactions are enabled on the teaser and news channels so interest can be measured before launch instead of guessed at.
A chat channel, and a reason to stay
A pre-release chat opens for suggestions and concerns. Everyone present is promised a permanent pre-release role at launch — the first retention mechanic, shipped before the server exists.
Alpha, and a second pair of hands
The project enters detailing, lore and mechanics. A manager joins the staff team and takes partnerships, which is where most of the early member growth came from.
Fifty members, waiting
The full lore teaser goes live on YouTube. Fifty people are in the server for a product that has not opened yet.
Launch — and a nine-minute fix
Doors open at 12:00. Pre-release channels are archived; goals, trailer and teasers stay permanently readable. Permission errors surface across community channels at 12:01 — acknowledged publicly and fixed by 12:10.
4,000+ messages on the first day
Launch-day volume passes four thousand messages. The server reaches boost level 2 the same day off member boosts.
One hundred members in a week
A hundred members inside seven days of opening, off partnerships and the pre-launch audience rather than paid reach.
Distribution
Listed on three public server directories, with the community asked directly to review and bump rather than the listings being left to work on their own.
Whitelisted character intake
A template hub, a copy-the-template flow and a submissions forum with three explicit outcomes: accepted, denied, or edits requested. Written up as a step list and a video so nobody had to ask.
Nine-point ruleset
Deliberately short. Conduct, platform terms, doxxing, age policy, staff impartiality, and a closing clause that absence from the list is not permission.
Seventeen numbered dev logs
Every addition, removal, move and change published under a running number, including the unglamorous permission fixes. The build was auditable by members while it happened.
Public roadmap
Milestones tied to member counts rather than dates, so progress was measured by the thing that actually mattered and nobody was promised a deadline that could slip.
Factions & senate
Six groups of interest with their own logos, banners and theme music, plus a facility senate with ten department representatives — a political layer members could actually hold positions in.
Geography as channels
The site, the surrounding region and the pocket dimensions built as a navigable channel tree, so the map was something you moved through rather than a document you read.
Open applications, published results
Moderator and partnership-manager applications decided in public, accepted and rejected alike, with a reason attached. Costly, and the reason nobody argued about favouritism.
Booster perks, split by permanence
Rewards divided into permanent in-roleplay benefits and temporary out-of-roleplay ones, so supporting the server bought something lasting instead of only a coloured name.
Operations as story beats
Multi-day operations run as lore chapters with briefings, after-action reports and outcomes that changed canon — the recurring reason to come back.
The owner account was compromised. The server did not stop.
Three weeks after launch the founder account was compromised and disabled by the platform. Support could not restore access. The account holding ownership of a community of several hundred people was simply gone.
Announced it publicly the same day, in plain terms, with a specific instruction: any message from that account is not me, do not click anything. A full server backup was completed the next day, an ownership transfer was requested, and the second account already held full permissions — so moderation, events and roleplay never paused.
The community lost nothing. That is only true because the permissions were distributed before they were needed.
Disclose immediately.
Public announcement the same day with a concrete warning, rather than a quiet fix that leaves members exposed to whoever holds the account.
Secure the asset.
Full server backup completed and posted, so the community existed independently of any one account.
Operate unchanged.
A second account with full permissions was already in place. Ownership transfer was requested and the server ran normally in the meantime.
This is the same argument as the Crimson case study, arrived at from the other direction: a community that depends on one account is as fragile as one that depends on one person.
