Skip to content
Journal

Teach With the Hands, Not With a Wall of Text

· ← Journal

The Panel Nobody Reads

Picture the screen. Everything dims. A panel slides up with four sentences explaining a system the player has not encountered yet. One button glows. The player taps it without reading, because they opened the game to play something, and a second panel appears.

We see that sequence constantly at The Evening Report, and our position on it has hardened over time. It is not a tutorial. It is a legal disclaimer for game mechanics — a way of being able to say the player was told. Teaching is a different discipline, and the games that practise it properly are recognisable within a minute.

Instruction Versus Instruction-by-Doing

The distinction we work with is simple. Instruction tells the player what a system does. Instruction-by-doing puts the player in a situation where the system's behaviour becomes obvious through use, and then optionally names it afterwards.

Consider a mechanic where objects of the same colour combine when brought together. The instruction approach shows a panel: matching items merge into a stronger item. The doing approach places two matching items adjacent to each other on an otherwise empty board with nothing else to try. The player drags one, sees the merge, and now knows the rule with a confidence no sentence can produce.

The second approach costs more to build. Somebody has to design a situation that can only resolve one way, and design it so it does not feel like a cage. That expense is why it works as a signal of quality: a team that builds teaching levels has thought about how learning happens.

The Vocabulary We Use in Notes

Teaching moment — a situation constructed so that the correct action is discoverable without explanation, and its consequence is immediately visible.

Naming — a short line of text delivered after the player performs an action for the first time, giving the thing a name so the game can refer to it later. Cheap, effective, frequently skipped.

Withholding — deliberately keeping a mechanic out of the opening so the player can master a smaller system first.

Reference — a permanently available explanation of the rules, reachable from a menu at any time. The single most requested and least implemented onboarding feature we encounter.

Front-loading — introducing several systems before any of them has been used. Almost always a symptom of a design that has not decided what its core is.

When to Withhold, and for How Long

Withholding is the part designers get wrong in both directions.

Hold back too little and the opening becomes a briefing: four resources, two characters and a map, all before the first meaningful input. Nothing sticks, and the player spends the next hour rediscovering things they were technically shown.

Hold back too much and the opening is thin. If the first session presents one verb and never complicates it, curiosity runs out before the game reveals what it is. We have abandoned genuinely good games because they took forty minutes to become themselves.

The pattern we find most reliable: one verb in the first minute, a complication of that same verb in the second or third, and a genuinely new element before the first session ends — but not before the first one is automatic. A player who still has to think about how to move cannot also learn what to do with a new resource.

A Comparison We Keep Returning To

SituationWeak handlingStrong handling
First inputModal explains the control, then a glowing targetThe control is the only thing that works, and it works
Second mechanicIntroduced in the same panel as the firstIntroduced after the first is used under mild pressure
FailureImmediately reversed with an apology messageAllowed to happen, with the cause made visible
Complex systemExplained fully in text before useRevealed in stages across several encounters
Forgetting a ruleNo way to look it upReference available from any menu
Optional depthBuried in a wall of stat descriptionsSurfaced when the player first has a reason to care

Failure Is a Teaching Tool

One of the more counter-intuitive things we look for is whether an opening allows the player to fail. A great many tutorials do not — the game intervenes, the sequence rewinds, and the player never finds out what happens when they get it wrong.

That protection has a cost. Understanding a system usually requires seeing its boundary. A player never allowed to fail does not know what failure looks like, cannot recognise it approaching, and often cannot tell whether they are doing well. Openings that permit small, low-consequence failures produce players who understand the game much faster — provided the cause is legible. Failure without a visible cause teaches nothing.

Why the First Five Minutes Specifically

Our desk keeps a separate note for the opening five minutes of every game we assess, and we are occasionally asked why that window rather than a first hour.

Two reasons. First, it is the only part of a game that essentially every player experiences, so it is the fairest single stretch to compare across titles. Second, it is where the design's priorities are least disguised. A game that puts an account prompt, a permission request and an advertisement in front of the first interaction has told us what it is optimising for, and it usually turns out to be consistent about that later.

We also record how the game behaves on a second launch. Openings that cannot be skipped the second time around are a small thing that becomes a large irritation on devices where the system closes a game regularly.

What This Means When You Are Choosing

If you are trying a new game, give it five minutes and pay attention to one question: at any point, did you do something before being told what it was. If the answer is yes and it felt natural, the design has a teaching instinct, and that instinct usually shows up in the rest of the game too.

This is one input among many when we set a position on the ranking, and that position is our own editorial judgement rather than a restatement of the Play Store star average — the two measure different things, as our methodology explains. More of our craft coverage is indexed on the blog.