Why Your Fake Chat Video Looks Fake (And How to Fix It)
There is a specific moment when a fake chat video stops working. The bubbles are the right colour, the font is right, the phone frame is right — and the viewer still knows, without being able to say why, that they are not looking at someone's screen. Nothing in the frame is wrong. What is wrong is everything between the frames.
Chat interfaces are made almost entirely of intermediate states. A message does not appear; it is typed, sent, delivered, read, and sometimes reacted to, and each of those steps has its own visual and its own delay. Strip the intermediate states out and you are left with a slideshow of correctly-coloured rectangles. That is the thing viewers detect.
This is a cross-template checklist for the details that decide whether a clip reads as a screen recording. It applies across the MockClip conversation templates — iMessage, WhatsApp, Instagram DM, Tinder, Reddit, and ChatGPT — and each section points to the deeper guide for the template you are actually using.
The tell is timing, not fidelity
Most people trying to fix a chat video that "looks off" reach for fidelity — a better font, a more accurate bubble radius, a more convincing status bar. Fidelity is rarely the problem, because the template already handles it. The problem is almost always rhythm.
Consider what actually happens when two people text. One person types for two seconds and sends a short message. The other person sees it, takes four seconds to respond because they are doing something else, types for six seconds because the reply is long, then sends. Then a fast one-word answer comes back in under a second. The pattern is irregular, and irregularity is what your eye has been trained on by thousands of hours of real messaging.
Now consider the default failure mode: every message appears one second after the last one. Same gap, same speed, no typing, no receipts. Within about three messages the viewer's pattern recognition flags it. They will not consciously think "the inter-message delay is suspiciously uniform" — they will just feel that it is fake.
Every MockClip conversation template exposes a delay before value on each individual message for exactly this reason. It is the single highest-leverage setting in the entire editor, and it is the one most often left at its default.
The fix. Before adjusting anything else, give every message its own delay and make the values uneven. A practical starting pattern for a five-message exchange: a short opening beat, a longer pause before the first reply, a fast follow-up, a noticeably long pause before the message that carries the punchline, then a quick closer. The long pause immediately before the payoff message is doing double duty — it reads as realistic hesitation and it builds anticipation.
Typing indicators: use them deliberately, not everywhere
The typing indicator is the most recognisable intermediate state in messaging, and it is the first thing people add when told their video looks flat. It is also frequently overused.
The iMessage, WhatsApp, and Instagram DM templates each let you turn the typing bubble on per message, control how long it runs, and control how fast the dots animate. The ChatGPT template has its own equivalent for the assistant's response.
Two rules cover most cases.
Match duration to message length. A typing indicator that runs for half a second and produces a long paragraph is a contradiction the viewer registers immediately. So is four seconds of typing that yields the word "ok". The indicator is a promise about how much text is coming; keep the promise.
Do not put one on every message. Real conversations do not show a typing bubble before every turn — short replies often land before the indicator would even render. Using it everywhere flattens the rhythm in exactly the way you were trying to avoid, just with more motion. Reserve it for the turns where the wait is part of the story.
The iMessage template also supports a cancelled typing state: the indicator appears, runs, and then disappears without a message arriving. It is a genuinely distinctive beat — the other person started writing something and thought better of it — and it does more storytelling work than almost any other single setting. The iMessage typing indicator guide covers the timing patterns in detail.
Delivery and read states are story beats
Sent, delivered, read. These are the states most often either ignored completely or switched on indiscriminately, and both mistakes cost realism.
The iMessage template exposes delivered and read states with a configurable read time. WhatsApp models the full three-step progression — the single sent check, the double delivered check, and the blue read check — each with its own independent delay, which is what lets a message sit on one grey tick for an uncomfortably long time before turning blue. Instagram DM has a "seen" state with its own delay.
The principle: a receipt is a beat, not a decoration. Turn it on when the story uses it.
- If the video is about being left on read, the read receipt is the content. Let the message sit unread, then flip it, then hold on the silence.
- If the video is a fast comedic exchange, receipts on every bubble are visual noise competing with the text.
- If the message is meant to feel unanswered, the gap between the read state and the next message is the whole joke, and it needs to be longer than feels comfortable while editing.
That last point deserves emphasis, because it is counterintuitive in practice. When you are staring at the editor, a four-second silence feels interminable. To a viewer who is reading the message for the first time, it is about right. Edit pauses slightly longer than your instinct suggests.
The WhatsApp read receipt guide, the iMessage delivered and read guide, and the Instagram DM seen guide each go deeper on their template's specific behaviour.
iOS adds two quieter states worth knowing about in the iMessage template: messages can be marked as delivered quietly, and a contact can be shown as having notifications silenced. Both are small, specific details that a viewer who has seen them on their own phone will recognise instantly — and both are the kind of thing nobody adds to a fake, which is precisely why they work.
Reactions have to land late
A reaction that appears at the same instant as the message it is attached to is physically impossible. Somebody has to read the message first, then react. When the two appear together, the viewer reads it as a graphic rather than an interaction.
Every template that supports reactions exposes a delay for exactly this. The iMessage template calls them tapbacks and offers six: heart, thumbs up, thumbs down, haha, exclamation, and question. WhatsApp offers like, heart, laugh, surprised, sad, and pray. Instagram DM offers heart, laugh, surprised, sad, angry, and thumbs up. In each case you also choose which side reacted, which is what determines whether the reaction attaches to your bubble or theirs.
A short beat between the message and the reaction is usually enough — long enough to imply reading, short enough to keep the clip moving. The reaction sets differ per platform, so a mismatched one is a quiet tell: a "pray" reaction is a WhatsApp thing, and an "angry" reaction is an Instagram thing. Using the wrong set is the sort of detail that only some of your audience will catch, but the ones who catch it are the ones who use the app daily.
Fill the fields real apps fill
The second-largest category of tell, after timing, is empty interface. Real app screens are dense with small pieces of information, most of which nobody consciously reads. A mockup that leaves those fields blank looks subtly like a demo, because a demo is the only place you ever see an app that empty.
This is most visible on the Tinder template. A profile card with just a name and an age is not what anyone's app looks like. A real card carries an occupation, a distance, interest chips, a bio line, and often a verification badge — and around the card sits the app's own chrome, including the likes badge on the tab bar and the notification dot on the messages tab. None of that is what the video is about, and all of it is what makes the frame read as a real screen.
The same applies everywhere else:
- WhatsApp and Instagram DM put a status line directly under the contact name. "Online", "Active now", "last seen recently" — an empty header is unusual.
- iMessage can show a timestamp header and date separators between messages. Conversations that span time show that they do.
- Reddit posts carry a vote count, a comment count, a subreddit, an author, and an age, and the Reddit template lets you set whether the post appears already upvoted or downvoted — a small, specific detail that implies the viewer's own account state.
- ChatGPT shows a model label in the header and can display the welcome screen, quick actions, or a header banner. A ChatGPT clip with a model label that does not match the era of the conversation is a tell to exactly the audience most likely to share it.
The general rule: fill in the fields that a real user would have filled in, even the ones the camera never lingers on.
Match the platform to the audience
A convincing mockup is partly a casting decision. The template your audience uses every day is the one they will accept without thinking and the one where they will notice a single wrong detail.
This cuts both ways, and it is worth being deliberate about. If your audience lives in iMessage, an iMessage clip gets instant recognition and zero friction — but your bubble colours, receipt behaviour, and tapback set all have to be right, because they know. If you pick a platform your audience uses less, you get more tolerance for small inaccuracies and less instant recognition.
The chat format guide compares what each template actually renders, which is the fastest way to pick. For the per-platform pillar guides, start with the iMessage conversation guide or the Reddit thread guide.
One consistency note that is easy to miss: every template offers a light and a dark theme, and the theme should match the rest of your footage. A dark-theme chat clip dropped into bright daytime footage reads as an inserted graphic rather than as someone's phone, no matter how accurate the interface is.
Write shorter messages than you want to
This is a writing problem rather than a settings problem, and it defeats more chat videos than any interface detail.
Messages written for a video tend to be too long, too well-punctuated, and too explanatory, because the writer is trying to convey a story. Real messages are short, fragmentary, and frequently split across several bubbles. People send "wait", then "what", then "when did that happen" as three separate messages rather than one composed sentence.
Three practical corrections:
Split long messages. One long paragraph in a single bubble is a strong tell. The same text across three bubbles with uneven delays reads naturally and gives you three timing beats instead of one.
Cut the punctuation down. Full stops at the end of short messages read as formal, and in some contexts as hostile — which is a real effect you can use deliberately, but not one to trigger by accident.
Let the reader infer. If a message explains the situation for the viewer's benefit, it is narration wearing a chat bubble. Real participants already know their own context. Exposition belongs in your caption, your voiceover, or your on-screen text.
A checklist before you export
Run through this before rendering. Most fixes take seconds in the editor and are far cheaper than re-cutting the video later.
- Are the delays uneven? If two consecutive gaps are identical, change one.
- Is there at least one deliberately long pause? Preferably right before the payoff.
- Do typing indicators appear only where the wait matters, and does each duration match the length of the message that follows?
- Are delivery and read states carrying story, rather than switched on everywhere by default?
- Do reactions land after their message, and are they from the set that platform actually uses?
- Are the interface fields filled in — status line, timestamps, profile details, vote counts, model label?
- Does the theme match the footage you are cutting this into?
- Are the messages shorter than your first draft? They usually should be.
- Does the clip end on the right frame? A chat clip that runs two seconds past its last beat loses the ending.
Where to go next
The fastest way to internalise this is to build one short exchange and deliberately get it wrong — uniform delays, no receipts, no typing, long composed messages — then fix one item at a time and watch each change land. Three messages is enough to feel the difference.
Pick the template that matches your audience and start there: iMessage, WhatsApp, Instagram DM, Tinder, Reddit, ChatGPT, or the phone call and notification templates for hook-length clips. Everything described here runs in the browser, and the export settings and tiers are listed on the pricing page.
The interface accuracy is already handled for you. What is left — the timing, the receipts, the empty fields, the length of the messages — is the part that decides whether anyone believes it.
Frequently Asked Questions
Why does my fake text message video look fake?
Almost always because every message arrives at the same moment with no typing indicator, no delivery state, and no gap between turns. Real conversations are uneven — someone types for two seconds, sends, the other person reads it, waits, then replies. A chat video that skips those intermediate states reads as a slideshow of bubbles rather than a screen recording.
What is the single biggest realism mistake?
Uniform timing. If every message appears exactly one second after the previous one, the eye registers the rhythm as mechanical within about three messages. Varying the delay before each message — a fast reply here, a four-second pause there — fixes more perceived realism than any other single change.
Do I need typing indicators on every message?
No, and using them everywhere is its own tell. Real people do not trigger a visible typing bubble before every single message. Use typing indicators on the messages where the wait matters — the reply to a difficult question, the long message, the one the viewer is anticipating.
Should I show read receipts in a fake chat video?
Only when the story uses them. A read receipt is a story beat: it proves the message was seen. If your video is about being left on read, the receipt is the whole point. If it is a simple back-and-forth, receipts on every message add clutter without adding meaning.
Which chat template is the most convincing?
The one your audience uses daily. Familiarity does the work — a viewer who uses WhatsApp every day will spot a wrong detail in a WhatsApp mockup instantly, and will also accept a correct one without a second thought. Match the template to where your audience actually talks.
Does the light or dark theme matter for realism?
It matters for consistency. Every MockClip template offers a light and a dark theme. Pick the one that matches the rest of your footage and stay with it — a dark chat clip cut into bright daytime footage reads as an insert rather than as someone's screen.
How long should a fake chat conversation video be?
Long enough for the exchange to land and no longer. Most chat clips used as hooks run a handful of seconds and contain three to five messages. If the conversation needs more than that, the clip is usually carrying story that belongs in narration or on-screen text instead.
Is it legal to make fake chat videos?
Creating clearly-fictional chat conversations for entertainment, teaching, or marketing is legal in most jurisdictions. Do not use the format to impersonate real people, to defraud anyone, or to harass. Label the content as dramatized when there is any chance a viewer would read it as a genuine screenshot.
Related Articles
Which Chat Video Format Should You Use? Creator Guide
iMessage, WhatsApp, Instagram DM, Reddit or ChatGPT? Compare every MockClip template by what it actually renders, and pick the right one for your video.
iMessage Typing Indicator Video: How to Animate the Pause
Animate the iMessage typing bubble as an MP4 — including the cancelled-typing beat where the dots appear, hold, and vanish without a reply arriving.
iMessage Read Receipt Video: Delivered, Read, and Quiet
Animate iMessage delivery status as an MP4 — Delivered, Read with a timestamp, and the Delivered Quietly state when notifications are silenced.
WhatsApp Blue Tick Video: How to Animate Read Receipts
Animate WhatsApp read receipts as an MP4. Chain the sent, delivered, and blue read ticks with per-step timing and export the clip for Shorts and Reels.
Instagram DM Video: Animate Reactions, Typing, and Seen
Animate Instagram DM reactions, typing bubbles, and the Seen indicator as an MP4. Time each beat, land a heart with no reply, and export vertical.