retti.aiTeardowns › I Opened A FAKE McDonalds Drive Thru
Predicted Retention Teardown

I Opened A FAKE McDonalds Drive Thru

By Niko Omilana · Business · 7.9M views · 24:28

I Opened A FAKE McDonalds Drive Thru

The teardown in brief

What's working

What's costing attention

The first 30 seconds

This is my fake McDonald's that's way busier than an actual McDonald's down the road. And this is the story of how it happened. I was sat on my laptop one day when I received a random email from a stranger that said, "Nikico, I need your help. I've opened a drive-through restaurant and was struggling to get customers.

Strong Tier 1 delivery. Hook fires at 4 seconds with the cold open (fake McDonald's busier than real one), immediately reaffirming the thumbnail promise. Confusion is eliminated instantly — the viewer knows this is about helping a struggling restaurant beat McDonald's. The concept is crystal clear by 0:16 when the email is explained. Predicted 76% retention at 30 seconds (high end for challenge content).

Where viewers drop

10:00 — Mid-Video Repetition (critical)

From 10:00 to 14:00, you're cycling through the same beat 15+ times: customer arrives → orders food → eats → says it's good → leaves. Each cycle hits the same emotional notes with the same structure. By the 5th iteration, the viewer knows exactly what's coming. The novelty vanishes and it starts feeling like filler — they're waiting for something NEW to happen.

Why it matters — Repetition is the #1 retention killer on YouTube (219 flags in our 200-video dataset). When viewers can predict the next 90 seconds, they click away to find novelty elsewhere. You'll see a gradual acceleration in drop-off through this section as the pattern becomes obvious.

3:23 — McDonald's Heist Tangent (moderate)

At 3:23 to 6:30, you spend 3+ minutes on a McDonald's parking lot heist — booth falls apart, fake orders, hiding from managers. It's funny, but it's NOT the video's core promise. Viewers clicked for 'fake McDonald's beats real McDonald's' — not 'sneaking around real McDonald's parking lot.' This tangent delays the main event by 3 full minutes.

Why it matters — In a LONG video, you have more tolerance for tangents AFTER the commitment audition (first 3 minutes). But this tangent happens AT THE END of your setup, right when viewers are expecting 'okay, now the main thing starts.' It's well-executed comedy, but it's borrowed time from the core concept.

11:30 — Sponsor Breaks Momentum (moderate)

At 11:30, right as your queue is starting to build and momentum is rising, you hard-stop for a 60-second Shopify ad. The viewer was leaning forward (are they going to get overwhelmed? will demand exceed capacity?) and you pull them out of the story completely. When you return at 12:30, you have to rebuild that tension from scratch.

Why it matters — Sponsor reads cause an average 3-8% retention dip. At 11:30 (mid-video), this is worse than at the 2/3 mark. The viewer just watched you serve your first few customers — they're deciding whether the rest of the video will be interesting. A sponsor break here communicates 'the interesting part is over, now we're stalling.'

22:30 — Slow Resolution / Emotional Drag (mild)

From 22:30 to the end (24:28), you spend 2 full minutes on slow emotional wrap-up: thank you speeches, team reflections, Niko walking around empty parking lot contemplating, slow music, email reveal. For viewers who came for high-energy challenge content, this feels like the video ended at 22:00 but keeps going. The pacing drops from action to meditation.

Why it matters — In LONG videos, you MUST stick the landing. The outro is where you hemorrhage viewers fastest — even great videos drop 10-20% in the final minute. Your average retention score will heavily depend on how many people you hold through this. Right now, you're testing their patience.

How the video is built

What any creator can steal

Want this on your own video?

Paste any YouTube URL and Retti maps every drop, spike and plateau to the moment that caused it.

Analyse a video free