Skills / Content-creation

Over the Shoulder Test

If someone looked over your shoulder while you were watching a video and asked, 'What are you watching?' — that's intrinsic appeal. That's a video the algorithm can take anywhere.

The Over the Shoulder Test is Colin & Samir's pre-production filter for YouTube concepts: if someone glanced over your shoulder while you were watching a video, would they stop and ask 'What are you watching?' If yes, your content has intrinsic appeal that drives algorithmic growth. If no, it only works for people who already follow you. This framework helps creators move beyond subscriber-dependent content and into videos that earn clicks from complete strangers.

By Colin Rosenblum & Samir Bhavnani  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisOver the Shoulder Test

Input

“I'm thinking: 'My Honest Review of 6 Months of Daily Journaling.' I've been covering personal development for a while and my subscribers seem to…”

Diagnosis

Let's run the Over the Shoulder Test.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Make videos strangers stop for, not just subscribers click

The Over the Shoulder Test is built on a core distinction: intrinsic appeal versus extrinsic appeal. Intrinsic appeal means a concept grabs any stranger's attention through visual novelty or a compelling curiosity gap — no knowledge of the creator required. Extrinsic appeal means the content only resonates because the viewer already follows and trusts the creator. The test is applied to the thumbnail and title combination — the 'package' of a video — before production begins. A concept passes if it would make a stranger, with zero channel awareness, stop and ask what they're watching. Failing concepts are reworked to lead with counterintuitive findings, stronger curiosity gaps, or visually surprising thumbnails rather than creator-centric framing. Colin & Samir use top performers like MrBeast and Mark Rober as consistent examples: their videos pass the test because the concept alone creates irresistible curiosity for any viewer.

The problem

Most YouTube creators unknowingly build a ceiling into their content. They make videos their existing subscribers will love — but those videos don't travel. Strangers scrolling through YouTube have no reason to click a concept that only makes sense if you already know and trust the creator. This is extrinsic appeal, and it keeps channels stuck below algorithmic escape velocity no matter how good the actual content is.

The solution

Run your video ideas through the Over the Shoulder Test before you film a single frame — so you only invest in concepts that earn clicks from strangers, not just loyalty from fans.

You bring

A video concept, working title, or thumbnail idea you want to evaluate
Context about your current niche and subscriber base
A backlog of competing video ideas to compare and rank
Honest reflection on whether you're making the video for fans or for strangers

You get

A clear pass/fail verdict on whether your concept has intrinsic appeal
Rewritten title options that open a stronger curiosity gap for unfamiliar viewers
Specific thumbnail notes identifying what makes your concept creator-dependent vs. universally intriguing
A ranked comparison of your video ideas by stranger-appeal potential

In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“You have a backlog of eight video ideas, all framed around your journey and process: 'My Morning Routine After 90 Days,' 'Why I Switched to This Productivity System,' 'What I Learned from 50 Books This Year.' Your subscribers engage reliably, but your channel has been stuck at 12K for six months. Your videos aren't being picked up in Browse Features or Suggested because the concepts only make sense if you already follow you.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With Over the Shoulder Test

“After running each concept through the Over the Shoulder Test, two ideas pass outright and three fail with specific, actionable notes. You rewrite the failing titles to lead with counterintuitive findings instead of your personal process. Your next video — repackaged around a stranger-facing curiosity gap and a tension-forward thumbnail — gets picked up in Browse Features for the first time in months, earning more than half its views from non-subscribers in the first week.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

Over the Shoulder TestClaude

Over the Shoulder Test

Let's run the Over the Shoulder Test. Picture this: someone glances over your shoulder while you're watching YouTube. They've never heard of your channel, they don't follow you, they have zero context about who you are. What would make them stop and ask 'What are you watching?' That's the bar your next video has to clear — on thumbnail and title alone, before they know anything about you. What concept are you working with?

I'm thinking: 'My Honest Review of 6 Months of Daily Journaling.' I've been covering personal development for a while and my subscribers seem to really like this topic.

Okay — let's apply the test. A stranger scrolling YouTube sees 'My Honest Review of 6 Months of Daily Journaling.' They don't know you. They don't follow your journey. Does that concept make them stop and ask what you're watching? Probably not. This is extrinsic appeal: your subscribers click because they care about YOUR experience. A stranger has no stake in your 6-month experiment. The concept leads with you — your review, your months, your journey — rather than with something that creates universal curiosity. What would change that? Usually it's a counterintuitive finding, a result that contradicts expectations, or a visual that doesn't match what people assume. What did you actually discover after 6 months of journaling?

Honestly? It mostly didn't do what everyone says it does. It didn't reduce my anxiety or make me more productive. It just changed how I argue with myself — which is weird and hard to explain.

That's your video — and now you have something that can pass the test. A stranger sees: 'I Journaled Every Day for 6 Months. It Didn't Work.' That opens a curiosity gap. They think: wait, everyone says journaling is transformative. What happened? That question exists for any viewer, subscriber or not. Even better if you sharpen the specific finding: 'I Journaled Every Day for 6 Months. Here's What Nobody Tells You.' Now the title implies insider knowledge a stranger doesn't have, and clicking becomes necessary. For the thumbnail: skip the posed-with-journal photo. Show the tension — your expression should communicate surprise or conflict, not satisfaction. That visual should ask a question before the title does. Together, that combination passes the test.

That's a big reframe. So the rule is: lead with the finding, not the process?

Exactly. Subscribers will watch almost anything you make because they're already invested in you. The algorithm sends your video to strangers who have zero context — and they make their decision in under a second. A 'review' format is creator-centric: it asks viewers to be curious about your experience. The reframe leads with a result that stops anyone mid-scroll. Colin & Samir call this the difference between a video that needs you to already have an audience and one the algorithm can take anywhere. Want me to generate five title variations that open that curiosity gap, plus thumbnail notes on what the image should communicate to pass the test?

Yes, let's see the titles.

Here are five title options, each designed to pass the Over the Shoulder Test:

'I Journaled Every Day for 6 Months. It Didn't Work.' — Direct contradiction of conventional wisdom. Forces the click to resolve the tension.2. 'Everyone Says Journal Daily. They Left Out This Part.' — Implies knowledge the viewer is missing. Makes clicking feel necessary regardless of who you are.3. '6 Months of Daily Journaling: What Actually Changed (It's Not What You Think)' — 'What actually changed' signals a real finding; the parenthetical adds intrigue without revealing the answer.4. 'Daily Journaling Changed Me — Just Not the Way Anyone Expected' — Emotionally specific. Works if your thumbnail matches an unexpected reaction.5. 'The Hidden Side Effect of Daily Journaling (I Found It After 6 Months)' — 'Hidden side effect' triggers curiosity for anyone who journals or has considered it.

For the thumbnail: avoid a satisfied pose with a journal. Split the frame — journal on one side, your reaction on the other — and make the expression read as genuine surprise or conflict. The visual should raise a question the title then answers. That combination gives a stranger a reason to stop before they've read a single word.

Reply to Over the Shoulder Test…

What the skill is doing

01Intrinsic vs. Extrinsic Appeal Diagnosis

02Stranger Perspective Simulation

03Curiosity Gap Title Rewrite

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

OTOver the Shoulder Testtyping…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

I'm thinking: 'My Honest Review of 6 Months of Daily Journaling.' I've been covering personal development for a while and my subscribers seem to reall…9:14✓✓

Let's run the Over the Shoulder Test.9:14

Picture this: someone glances over your shoulder while you're watching YouTube.9:15

They've never heard of your channel, they don't follow you, they have zero context about who you are.9:15

Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

◦Reads the situation, names the pattern, returns one concrete next move.

◦Delivered in seconds, inside a conversation that already exists.

Specimen · WhatsApp Business API · live

Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Intrinsic vs. Extrinsic Appeal Diagnosis

Evaluates whether your video concept stands on its own for a stranger or relies on the viewer's pre-existing relationship with you. Content built on extrinsic appeal grows linearly with your subscriber list; content with intrinsic appeal gets pushed by the algorithm to new audiences who've never heard of you.

◦ MethodBased on Colin & Samir's foundational framework distinguishing intrinsic appeal (visual or conceptual novelty that stops any viewer) from extrinsic appeal (creator-dependent interest), developed through their analysis of top-performing YouTube channels in Project 300.

CapabilityC-02

Stranger Perspective Simulation

Applies the test's central question — 'Would a stranger glance over your shoulder and ask what you're watching?' — to your actual concept from the perspective of someone with zero prior knowledge of your channel. This forces creators out of their own point of view and into the mindset of a cold audience member who owes them nothing.

◦ MethodDirectly implements the Over the Shoulder Test heuristic from Project 300, which asks creators to simulate a stranger's reaction to evaluate whether a concept has earned its click on the concept alone.

CapabilityC-03

Curiosity Gap Title Rewrite

Rewrites your working title to open an irresistible unanswered question — one that can only be resolved by watching. Weak titles describe what happens; strong titles imply something surprising, unresolved, or counterintuitive that makes clicking feel necessary even for a viewer with no prior context.

◦ MethodApplies the curiosity gap principle central to Colin & Samir's packaging framework, where the title's job is to trigger a question that compels even unfamiliar viewers to investigate — not to describe the creator's experience.

CapabilityC-04

Thumbnail Novelty Audit

Reviews your thumbnail concept for visual novelty and intrigue — the elements that lock a stranger's eyes onto your frame while scrolling past dozens of others. Flags thumbnail designs that rely on recognizing your face or channel versus those that communicate something visually surprising on their own merits.

◦ MethodBased on Colin & Samir's repeated analysis of top-creator thumbnails (MrBeast, Mark Rober, etc.), which consistently use visual novelty and emotionally specific imagery to create stranger appeal independent of creator familiarity.

CapabilityC-05

Concept Ranking by Over the Shoulder Potential

Compares multiple video ideas from your backlog and ranks them by their estimated Over the Shoulder Test performance — helping you prioritize concepts with the highest stranger-appeal potential for your next upload, rather than defaulting to whatever feels most comfortable.

◦ MethodExtends the Over the Shoulder Test into a comparative evaluation tool, consistent with Colin & Samir's approach of treating the test as a pre-production filter that guides which ideas to invest in rather than a post-production retrospective.

Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces

OutputD-01

Over the Shoulder Scorecard

A structured pass/fail evaluation of your video concept across the core test criteria: thumbnail novelty, title curiosity gap, and stranger appeal. Includes specific notes on what's working, what's creator-dependent, and exactly what needs to change to earn clicks from people who've never heard of you.

OutputD-02

Repackaged Title & Thumbnail Brief

A set of 3-5 rewritten title options and thumbnail direction notes designed to maximize intrinsic appeal — ready-to-test alternatives that lead with curiosity gaps and visual novelty rather than creator-centric framing.

OutputD-03

Concept Priority Ranking

A comparative ranking of your video backlog by Over the Shoulder Test potential, with a brief rationale for each position so you know exactly which concept to prioritize for maximum algorithmic reach on your next upload.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Colin Rosenblum & Samir Bhavnani

Colin Rosenblum and Samir Bhavnani are YouTube strategists and creators behind The Colin and Samir Show, one of YouTube's leading channels on the business and strategy of content creation. Through their Project 300 course and public channel, they've analyzed the growth mechanics behind top creators including MrBeast, Mark Rober, and others to distill repeatable frameworks for breaking through on YouTube.

Status · Inspired by Colin Rosenblum & Samir Bhavnani’s work — not yet claimed. Are you Colin Rosenblum & Samir Bhavnani?

Primary sourceS-01

Project 300

by Colin Rosenblum & Samir Bhavnani

Hosts of The Colin and Samir Show; creators of the Project 300 course; recognized YouTube strategy analysts whose channel and podcast reach millions of creators

Read the original ↗

Citationyoutube.com

In the build queue

Be first to run it.

Over the Shoulder Test is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail

At launchI want to run my next video concept through the Over the Shoulder Test. Here's my working title and thumbnail idea — tell me if a stranger would stop and ask what I'm watching, and what I'd need to change if they wouldn't.