retti.ai › Teardowns › I built my DREAM New York City apartment from SCRATCH in 93
Predicted Retention Teardown

I built my DREAM New York City apartment from SCRATCH in 93 days

By Charles Liu · Diy · 1.7M views · 49:41

I built my DREAM New York City apartment from SCRATCH in 93 days

The teardown in brief

What's working

What's costing attention

The first 30 seconds

Welcome to my apartment. This apartment is a one-bedroom, one-bathroom apartment roughly 630 sq ft and some change. Notable details include a kitchen built into the wall, a relatively small bedroom, but considerably large for New York City standards. We have very tall ceilings, approximately 11 ft high. And lastly, we

The concept is clear within 90 seconds but the video never shows the finished apartment in the opening — for a transformation video titled 'dream apartment,' transformation viewers expect to see the destination in the first 10 seconds, and right now the hook delivers 94 seconds of 'before' with zero 'after.'

Where viewers drop

0:00 — No Final Result Teased in Hook (critical)

The video opens on a bare, empty apartment with a calm tour — you spend 94 seconds describing specs, the four phases, and the aesthetic goal, but you never flash-forward to show a single frame of the finished apartment. For a transformation video, viewers clicked to see the before-and-after. Right now the 'after' is entirely absent from the opening.

Why it matters — Transformation viewers need to see the destination in the first 15 seconds to commit. Without a glimpse of the finished room, a significant portion of viewers have no reason to stay through 50 minutes of construction.

0:00 — Zero Explicit Stakes Throughout (critical)

You mention the budget is nearly depleted around the 26-minute mark ('we're going to use up pretty much 100% of the rest of the budget') but you never tell viewers what the budget is, what happens if you fail, or what's actually on the line. The kitchen island is framed as a big challenge but there's no consequence stated for it not working — the viewer can leave at any point with no FOMO about the outcome.

Why it matters — For a 50-minute video, you need viewers to fear something going wrong. Right now everything is exciting but consequence-free. Adding one concrete stake early — budget, deadline, or a specific consequence — turns every setback into a genuine tension moment instead of just an entertaining moment.

37:13 — Personal Philosophy Interlude Pauses Build (moderate)

At 37:12, mid-woodworking-shop setup, the video pivots to a 30-second personal monologue about being 27, wanting to make movies, and giving creative life a shot. It's genuinely charming but it comes at the exact moment the kitchen island build is about to start — momentum stalls right before the video's most complex challenge.

Why it matters — Viewers who have stayed 37 minutes are here for the build, not a life-philosophy interlude. The interlude works better placed earlier (during the opening context) or right at the end as an emotional payoff — not mid-shop as a speed bump before the hardest sequence.

46:09 — Final Decoration Montage — Thin Narration (moderate)

The last three minutes of the main content — the decoration phase, your most visually rewarding section — has almost no spoken narration in the transcript. The 'home stretch' setup at 46:10 gives way to about 3 minutes where the transcript coverage is nearly empty before the closing line.

Why it matters — For a 25-44 audience who has invested 46 minutes, this is the payoff they've been waiting for. Letting it run as a silent montage means the emotional peak of the video is also its quietest and most disengaged section. A few well-placed voiceover lines here would anchor the viewer to the reveal.

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