Back
Commentary Guide
Dashboard →
⬡ The Tester Handbook

How to Give Great Commentary

The goal of a SquidVox playtest is not to create a walkthrough. It is to give the developer an honest window into your mind as you experience their game for the first time. That requires a specific kind of narration.

The Core Principle

Tell us the why and the feeling — not the what. A developer can watch a recording and see what you did. What they cannot see is why you did it or how it made you feel. That gap is what your voice fills.

Good vs. Bad Commentary

Read these pairs carefully. The bad examples are not lazy — they are a common mistake that even thoughtful people make.

Navigation & Exploration

Bad
I'm walking forward. Now I turned left. There's a door. I'm going through the door.

This is pure transcription. You're describing what the developer can already see in the video. It adds zero value.

Good
I'm heading left because that area looks brighter — I assume that's where I'm supposed to go. Actually, I feel like the game hasn't told me what my goal is yet. I'm just wandering and hoping something will make sense.

This reveals intent, assumption, and emotional state. The developer learns: their lighting guided the player, but the lack of a clear objective created anxiety.

Stuck Moments

Bad
I'm stuck. I don't know what to do.

This tells the developer you're stuck, but not why or what you've tried. It's not actionable.

Good
I'm stuck. I see a lever on that platform but I can't figure out how to get up there. I've tried jumping from this box three times. The jump feels a bit floaty so I keep overshooting. Maybe I'm supposed to come from a different angle?

The developer now knows: the lever is visible (good), the path to it is unclear (bad), the jump feel is undermining precision, and the player is self-directing their problem-solving (good sign).

Combat & Mechanics

Bad
I'm fighting the enemy. I hit it. It died.

A timestamp would be more useful than this. No information is conveyed about the feel of the combat.

Good
The sword swing feels satisfying — there's a nice screen shake when it connects. But I have no idea what my health is right now. I think I'm taking damage but the UI is so small I keep losing track of it. I feel anxious but not in a fun way.

The developer learns: hit feedback is working, the health indicator has a readability problem, and the anxiety it creates is not the intended tension.

UI & Menus

Bad
I opened the menu. There are some options. I closed it.

Nothing useful communicated.

Good
I opened the inventory but I honestly don't understand it. There are three columns and I'm not sure if items go into slots or if I just have a bag. I spent about 30 seconds just hovering over things hoping a tooltip would appear. Nothing did.

The developer learns exactly where the mental model breaks down: column layout is unclear, slot vs. bag mechanic is not communicated, and the absence of tooltips is a critical gap.

Practical Tips

01

Think out loud before you act

Before clicking or pressing a button, say what you are about to do and why. This single habit transforms a playthrough into a usable data source.

02

Name your emotions

Words like “confused”, “surprised”, “bored”, “delighted”, “frustrated”, and “tense” are enormously useful. Developers designing for a specific emotional arc need to know when they're off-track.

03

Don't self-censor your confusion

If something doesn't make sense, say so — even if you think it's your fault. “This is probably obvious but I have no idea how to...” is some of the most valuable feedback a developer can receive.

04

Keep the camera audible

If you pause to read something on-screen, read it out loud. Developers want to know which text is actually being absorbed and which is being skipped.

05

End with a summary

Spend the last 1–2 minutes of your video giving your overall impression: what worked, what confused you most, and whether you'd keep playing. This gives the developer a high-level synthesis alongside the detailed narration.

Recording Setup Tips

!

Test your setup before you claim

Do a 2-minute test recording before starting any playtest — verify that your screen capture is working and your microphone is audible. If you complete a 60-minute playtest and discover you forgot to hit record, that is on you. We recommend OBS Studio — it is free, open-source, and the industry standard for screen recording.

06

Play in one sitting

We strongly recommend completing your playtest in a single uninterrupted session. This keeps your commentary natural and avoids the need to edit together multiple clips.

If life interrupts you, pause your recording, handle what you need to, and resume as soon as possible. A small gap in the footage is fine — just note it briefly when you resume (“quick break, picking back up now”).

07

Pick a session length that fits your day

If you only have 15 minutes free, claim a 15-minute playtest — not a 60-minute one. The browse page lets you filter by duration. A focused 15-minute session is far more valuable than a rushed hour.

08

We understand life isn't perfect

We don't expect a perfectly polished production. Long sessions will naturally have small stumbles, pauses, or moments where commentary dips. That is fine. What we do expect is a genuine, good-faith effort to play and narrate for the agreed duration. Do your best — that is all we ask.