Playable ad best practices: hooks, tutorials and end cards

A playable has three jobs and very little time: stop the thumb, teach the action, hand the player to the store. When a playable underperforms, the cause is almost always one of those three, and the in-ad events usually tell you which. If players never touch the screen, it is the hook. If they touch it and stall, it is the tutorial. If they finish and do not tap, it is the end card.

These are the practices we apply on every build, with examples from our own concept playables, which you can play in the browser.

The first three seconds: the hook

TikTok's ad network guidance asks that players understand the gameplay within two seconds. That is a good standard for every network. In practice it means:

• Open on a game in progress. No logo animation, no title screen, no "tap to start". The first frame should look like the middle of a level, with one thing obviously wrong or unfinished.
• Show the problem before asking for the solution. In Park Out the lot is packed and one car clearly has a way out. In Hand in Hand the jelly friends' arms are visibly crossed. The player knows what "done" looks like before they move.
• One focal point. If the eye has three places to go, it goes nowhere. Dim or simplify everything that is not the next action.
• Design for silence. Networks such as AppLovin, Google and Liftoff require sound to wait for the first interaction, so the opening must work with no audio at all.
• Load the first scene first. Put the assets for the opening frame at the front of the file and defer the rest. A slow first paint wastes the best second of attention you will get.

Tutorials that teach in one gesture

The tutorial is not a separate screen. It is the first move, shown and then handed over.

• Show the exact gesture. A hand that drags, swipes or holds in the real direction, on the real object. Text, if any, is two or three words: "Swipe to cut", "Hold to heat".
• One gesture per playable. If the game has two controls, teach the second only after the first succeeds, or leave it out.
• Make the first move impossible to fail. In Fair Share, any swipe through the food produces a cut, and the result is scored rather than rejected. The player is rewarded for trying before they are judged for precision.
• Replay the hint when nothing happens. If there is no input for a couple of seconds, show the hand again. TikTok's ad network suggests moving the playable on by itself after ten seconds without input, which is a useful backstop everywhere.
• Accept sloppy input. Generous hit areas, snapping, forgiving physics. The player is holding a phone on a bus, not a controller.

Gameplay: difficulty, length and the near miss

Easy, then interesting

The first step should feel effortless. The second should ask for a little skill or carry a small risk. In Pop Pop Pan, holding the heat pops the kernels into the bucket, but hold it too long and they burn. That single risk turns a toy into a game, and it takes no extra explanation.

Keep it short

A frequently quoted rule from Unity's ironSource team: if users need more than 30 seconds to complete your playable, cut it down. The same article puts hyper-casual playables at around 12 to 15 seconds and more complex IAP games at around 22 to 28. Shorter is not automatically better; the right length is the shortest one that delivers the satisfying moment and a reason to want more.

Decide what failure looks like

Should the player be able to lose? Often yes. A near miss, like a slingshot shot in Counting Sheep that lands just short, creates a clear "let me try that again" moment that the end card can answer. A playable where nothing can go wrong feels like a demo. Test a version with a fail state against one without; the answer differs by genre.

Stay honest to the game

The playable should show a mechanic that exists in the first session of the real product. Players who install for something the game does not have leave quickly, and networks and stores have policies against misleading ads. If the real core loop is too deep for 30 seconds, show its most readable slice.

Feedback the player can feel

• Every touch answers immediately. A squash, a sound after the first tap, a small burst. A touch with no visible response reads as a broken ad.
• Show progress. A goal counter, a progress bar or an obvious target. In Tile Harbor, matched tiles fill a visible goal, so every match feels like movement towards something.
• Celebrate the right moment. Save the biggest reaction for the moment you want remembered, usually just before the end card.
• Protect the frame rate. Smoothness is part of feel. Test on a mid-range Android phone, not only on a new iPhone, and cut effects before they cost frames.

Words, sound and languages

• Fewer words, bigger type. Each screen should carry one short instruction or one short reaction. If a line needs a second reading, it is too long for an ad.
• Keep text live, not baked into images. Live text can be translated, tested and corrected without re-exporting art, which is what makes language variants cheap.
• Leave room for longer languages. German, French and Portuguese lines often run longer than English. Design buttons and banners with spare width so translations do not shrink to unreadable sizes.
• Watch font weight in the budget. Scripts such as Japanese or Cyrillic need extra glyphs. Subset fonts per language build so one market does not push every file towards the size cap.
• Sound supports, never explains. Once the player has tapped, short sounds tied to actions add feel. Nothing essential should depend on hearing it.

The end card and the CTA

What goes on it

The game or brand name, the icon, one line of value and one large button. The art should match the store page, so the player recognises what they land on. For brand playables, the line is the offer: our Brava Burger concept ends on free delivery on the first order, with an Order now button.

When it appears

After a win, after a loss and after a time limit. Every player should reach it, including the ones who wander off mid-level. If the end card only follows a win, you are only asking the people who were already convinced.

How the click works

• Use the network's store call, only on a deliberate tap. mraid.open(), FbPlayableAd.onCTAClick(), ExitApi.exit(), window.openAppStore() or window.install(), depending on the network. Our specs checklist lists which is which.
• Never redirect automatically. AppLovin and Unity both forbid opening the store automatically or on the first tap, and Mintegral blocks redirects fired on load or on a timer.
• Plan for an always-visible button where required. Mintegral asks for a CTA button that stays visible for the entire playable. Elsewhere, a small persistent install button during play is worth testing against an end-card-only version.
• Make the button big and unambiguous. One verb, high contrast, away from the corners where networks place their close controls.

Technical hygiene that protects performance

Some of the biggest performance losses are not creative at all:

• Self-contained file. No outside requests while the ad runs. A missing image on a slow connection is a broken ad.
• Size with headroom. Stay comfortably under each network's cap, which is around 5 MB on most of them.
• Pause when hidden. Stop the loop and the audio when the ad goes out of view; AppLovin explicitly requires audio to stop when the ad is hidden or closed.
• Both orientations, or the right one. Most in-app networks expect portrait and landscape; Meta expects portrait.
• No close button of your own. The network draws it. Keep your UI out of the corners.

Our step-by-step guide to making a playable ad covers the build side in more detail.

A one-minute checklist

1. Does the first frame look like a game in progress, with one obvious action?
2. Is the gesture shown, in two or three words at most?
3. Is the first move impossible to fail?
4. Does the hint replay when nothing happens?
5. Is there a moment of risk or a near miss?
6. Can a typical player finish in under 30 seconds?
7. Does the end card appear after a win, a loss and a time limit?
8. Does the store open only from a deliberate tap, through the network's own call?
9. Does it work silently, offline and in the orientations the network needs?

Then test it. The best practices above give you a strong first version; A/B testing tells you which of them matter most for your game.

Start a playable with us

Want a second pair of eyes on a playable that is underperforming? Send us your store link and the current build. We will tell you what we would change first, and storyboard a new version if it needs one. Write to info@zerind.com or use the form on our homepage.

Start a projectPlay our concepts

Related reading

• A/B testing playable ads: hooks, difficulty, end cards and variantsConcept tests before iteration tests, one primary metric, the order we vary things in, and how to build a playable so new variants are a config change rather than a rebuild.
• How to make a playable ad: from storyboard to network-ready HTML5The production path we follow on every build: brief, one-page storyboard, asset prep, a lean HTML5 engine, one adapter per network, and QA on real phones.
• Playable ads vs video ads: when each one winsThe formats do different jobs. A framework for deciding where a playable earns its production cost, where video should lead, and how to compare them without fooling yourself.

Sources

Checked on 8 October 2026. Networks update their specs; always confirm against the current page before you upload.

1. TikTok Ads Help, How to create TikTok Ad Network playable ads: https://ads.tiktok.com/help/article/how-to-create-tiktok-pangle-playable-ads
2. Unity (ironSource content team), 5 missed opportunities that can destroy your playable ad: https://unity.com/blog/5-missed-opportunities-that-can-destroy-your-playable-ad
3. AppLovin, Creative specs and guidelines: https://support.applovin.com/en/growth/promoting-your-apps/creatives/best-practices-and-guidelines
4. Unity Ads, Playable asset specifications: https://docs.unity.com/en-us/grow/acquire/creatives/playable/specifications
5. Mintegral Help Center, Playable test guide: https://helpcenter.mintegral.com/en/docs/playable-ad-guide
6. Google Ads Help, HTML5 and playable asset requirements for App campaigns: https://support.google.com/google-ads/answer/9981650
7. Liftoff Accelerate, Interactive ad integration: https://support.accelerate.liftoff.io/hc/en-us/articles/52724569196699
8. Meta Business Help Centre, Playable ad specifications: https://www.facebook.com/business/help/412951382532338