Recruiting Staff Without Ending Up With the Wrong People

The first admin you appoint is usually the person who asked. That is the mistake, and it is close to universal, because asking is easy and the people who ask are not the people you want.

Most communities that fall apart do not fall apart over rules or hosting or a game update. They fall apart because someone was given authority they should not have had, or because the one person doing all the moderation stopped.

Why you need anyone at all

Be honest about the threshold before you start recruiting, because premature staff cause more problems than they solve.

You do not need staff while you can be present at peak yourself and incidents arrive at a rate of a few a week. A single owner covers a server of thirty comfortably.

You do need staff when your peak has hours you are never awake for, when incidents outpace your evenings, or when your own enjoyment of the server has become administration. That last one is the signal people ignore until they burn out, burnout and handover is the destination if you keep ignoring it.

The specific test, are there hours every week when something could go badly wrong and nobody with authority would see it? If yes, you need coverage. If no, you need a person to talk to, which is not the same thing and does not require giving them commands.

The people who ask

Applications for staff arrive constantly on any server with a population. The base rate on unsolicited applicants is poor, for a reason worth naming: on most servers, the ability to kick and ban is the only visible status available, so it attracts people who want status.

That does not make every applicant unsuitable. It means the application is weak evidence, and you need better evidence from somewhere else.

The strongest signal is behaviour you already saw. Somebody who explained the rules to a newcomer without being asked. Somebody who reported a problem accurately instead of arguing about it in chat. Somebody who was present through a quiet month.

The second strongest is time. Not playtime as a number, but months of being around. A person who has been on the server for six months has demonstrated something a fortnight cannot.

The weakest is enthusiasm. It is also the thing most applications lead with.

What to look for

Four qualities, in this order.

Calm under provocation. The job is dealing with people at their worst. Someone who escalates in ordinary chat will escalate with a ban command. You have already seen how each of your regulars behaves in an argument; that is the interview.

Willingness to say "I'll check". Moderators who need to be right immediately make bad decisions fast. The ones who say they will look into it are the ones you can leave alone.

Consistency over judgement. A moderator who applies the written rule the same way every time is more valuable than one with better instincts and no pattern. Players can live with a rule they dislike; they cannot live with a rule that depends on who is online. Rules that work is what makes consistency possible in the first place.

Coverage you do not have. A brilliant candidate in your timezone is worth less than an adequate one in the eight hours nobody watches. Look at your peak-hours chart and recruit against the gap rather than against the queue of applicants.

The mechanics that prevent most problems

Recruit privately. Approach people rather than opening applications. The moment you post a form you are choosing from self-selection, and you have committed to rejecting people publicly.

Give the smallest permissions that do the job. Someone who needs to mute and kick does not need ban, file access, or the panel. Most communities give everyone the same tier because it is easier, and then have no way to limit the damage when one person is wrong. Security basics covers the account side, and the principle is identical.

Start with a trial. A month, stated as a month, with a conversation at the end. This makes stepping someone back down an expected outcome rather than a demotion, which is the single thing that makes it possible to do at all.

Two, not one. A single moderator is a single point of failure and, more importantly, has nobody to check a decision against. Two people who can disagree privately produce better outcomes than one who cannot.

Never appoint for money. A donor with a ban command is the fastest route to a community that does not trust its own moderation. Funding a community server covers the models that do not do this.

Never appoint to keep someone. Promoting a valuable regular who threatened to leave sets a price on the role and everyone learns it.

Writing down what they may do

Most staff problems are not misconduct. They are a person making a call nobody had defined, then being told they were wrong.

Before the first appointment, write:

• Which offences get a warning, a temporary ban, a permanent one
• Which decisions need a second opinion
• What to do when unsure at 3am with nobody online
• Where decisions are discussed, and where they are not
• Who can overturn whom

Half a page. It removes most of the disputes and, more usefully, it means a new moderator can be wrong without it becoming personal, they missed a line rather than exercised bad judgement.

Moderating a game community covers the day-to-day, and player support without a helpdesk covers routing reports so they do not all land in your DMs.

Removing someone

This is the part nobody plans and everybody eventually needs.

Do it privately, and do it once. A quiet conversation, access removed the same day.

Do not announce why. "X is no longer staff, thanks for their time" is the whole message. Explaining invites a debate you cannot win and drags the community into a personnel matter.

Remove access immediately. Panel, RCON, Discord roles, the lot. Not because you expect the worst but because access that lingers is how the worst happens, and the majority of griefing-by-admin stories involve a former admin whose permissions nobody revoked.

Expect to lose them from the community. Most people who are stepped down leave, and that is a cost you are accepting. It is smaller than the cost of a moderator you cannot rely on.

The thing that matters more than any of it

Staff take their register from the owner. A community where the owner argues with players has moderators who argue with players. A community where the owner says "let me check" has moderators who say it too.

You cannot train that in and you cannot write it into a document. It is the one part of this that is about how you behave on your own server, and it is the reason some communities can appoint almost anyone and be fine.

The first month with a new moderator

Appointing somebody is the easy half. The month afterwards is what determines whether it worked.

Week one: watch, do not act alone. A new moderator should be present while you handle incidents rather than handling their own. This is the fastest way to transfer how decisions get made, and it costs nothing.

Week two: act, then report. They handle things and tell you what they did. Not for approval: for calibration. Two or three conversations here prevent most of the drift that shows up later.

Week three: check the edges. Ask what they were unsure about. The answer tells you which parts of your written rules are ambiguous, and it is usually the most useful documentation review you will get.

Week four: decide. Confirm, extend the trial, or step them back. Saying it out loud at a date you named in advance is what makes stepping back possible without it being a demotion.

The failure mode here is appointing somebody and then not speaking to them for two months, which produces a moderator applying rules they invented and a community noticing before you do.

When you have several

Two dynamics appear once there is a team, and both are worth pre-empting.

Disagreement in public. Two moderators contradicting each other in front of players undermines both, and every decision afterwards. A private channel for staff discussion is the whole fix and it costs nothing.

Escalating for approval on everything. The opposite problem, and it means the written rules are too thin. If every decision comes to you, the answer is more documentation rather than more supervision.

Moderating a game community covers running the team day to day, and player support without a helpdesk covers routing reports so they arrive in one place rather than five.

The one-person version

Most community servers have exactly one person with authority, and the advice above assumes a team you may never have.

If it is only you, three things matter more than recruitment.

Write the rules down anyway. Not for staff, for yourself. Consistency across months is harder alone than it looks, and a written policy is what keeps a decision in March matching one in September.

Set hours you are not on duty. A single owner who is always available is a single owner who burns out, and burnout and handover is where that ends.

Find one person you trust before you need them. Not as staff — as somebody who can hold the keys if you disappear for a fortnight. The servers that survive an owner's absence are the ones where that conversation happened in advance.