inzpyre.me

Content & Marketing

You built it. Now someone has to see it.

The best product loses to the more visible one. Here is the craft behind it: how short videos earn attention and turn viewers into customers. Not viral luck, but a system you can learn.

On this page

01The 5 gamesWhich content game fits your goal.02Anatomy of a videoThe five building blocks every video is made of.03Finding ideasWhere good ideas come from and how to find them.04The hookThe first seconds that decide everything.05From view to customerHow a view becomes an email and a customer.06Growth channelsSeven tactics real apps grew with.07AI toolkitSkills and MCPs that speed up your content work.08Getting betterHow to get a little better with every video.09My experienceWhat 3 weeks of an internship on a World Cup tour delivered.

Strategy

The 5 content games

Before you worry about hooks and editing, answer one question: which game are you playing? The five differ in their goal and in how money flows. Confuse them, and you optimize for the wrong number.

Reach

Maximum fame

Goal
As many views as possible, broad across all topics.
Money comes from
Ad revenue and big brand deals. You are the product.
Typical for
Large creators without a fixed niche.
Niche

Known in one topic

Goal
High reach, but within a clear niche.
Money comes from
Brand deals and sponsorships that fit the niche. Here too, you are the product.
Typical for
Creators with a fixed topic and a loyal community.
Product feel

A world around the product

Goal
Build a world and a lifestyle around the product.
Money comes from
Sales, carried by mood and identification.
Typical for
Fashion labels, lifestyle and consumer brands.
Product education

Explain the purchase

Goal
Show the buyer why they need the product.
Money comes from
Sales through education, plus a bit of lifestyle.
Typical for
Products that need explaining, new tech, health.
Authority

Trust and expertise

Goal
Build deep trust and visible competence.
Money comes from
Viewers turn into leads and clients. Selling mostly happens off-platform.
Typical for
Consultants, coaches, agencies, info products.

For most people who build a product with AI

Games 3 to 5 are the relevant ones. You sell something concrete, so you either build a feeling around the product, explain its value, or build authority. Pure reach is its own business with its own rules, and rarely your goal.

Craft

The 5 building blocks of a video

Every short video is made of the same five parts. Name them individually, and you can tune one on purpose instead of blindly remaking everything.

  • Format

    the canvas

    The shape you tell in: breakdown, scenario, hero's journey, a list. The format frames how the information lands with the viewer.

  • Idea

    the subject

    What it is about. Three parts: topic, seed (the one-line premise) and substance (the facts, takes and examples that drive emotion).

  • Hook

    the first impression

    The first seconds. Three layers: text on screen, the visual shown, the spoken word. Important in exactly that order.

  • Script

    the paint

    The story structure, the through-line, the call to action and the small tricks that keep people watching.

  • Edit

    the brush

    Look, layout, pacing, captions, recurring elements. The edit sets the feel and whether the video is easy to follow.

The remix principle

You rarely invent everything. You take a video that worked, hold most of the blocks constant, and change only one or two. The more you keep constant, the higher the chance it works again.

Example: someone explains the three levels of a thing, from beginner to pro. You are a dental surgeon. You keep the format, hook words and layout and swap only the topic for the three levels of veneers. Format stays, topic changes.

Ideas

Where good ideas come from

An idea is not a flash of genius, but three parts that have to fit together. And it comes from predictable sources, not from nowhere.

Topic

The sub-category the video is about. For example: hooks.

Seed

The one-line premise. Ideally a take that contradicts the common belief. For example: the four most common hook mistakes.

Substance

The facts, takes and examples you use to fill the seed and drive emotion.

Whether a video outperforms comes down to two questions: is the topic relevant to your ideal viewer? And are the seed and substance genuinely interesting or useful to them?

The 6 idea sources

Ranked by hit rate. Early on, spend most of your time on source 1 and 2.

  • Your own outliers

    What already outperformed on your channel. The best source. Remake your winners with a slightly different seed, and they usually crush again.

  • Competitor outliers

    What is working in your niche right now. The easiest entry point. This is where you should spend most of your research time.

  • Other formats

    What is working in your niche elsewhere: long YouTube videos, podcasts, newsletters, posts. Translate it back into short video.

  • Niche news

    Current topics and breaking stories. No proven video data, but depending on the field a strong source for new ideas.

  • Good ideas that flopped

    Interesting competitor ideas that did not land. Riskier, but often only the execution was weak, not the idea.

  • Your own takes

    What you find interesting right now. Highest risk, highest reward, because no proof exists. Save these for later, once your eye is trained.

How to find validated ideas

A simple research loop. An analytics tool for social reach helps, but is not required.

  1. Build a watchlist

    10 to 15 accounts in your niche that serve the same audience. Sort their last 30 days by outlier score, meaning views relative to the channel average.

  2. Validate topics

    Scan the top videos. Which topics keep showing up on top? These five to seven become your focus, because there is proof.

  3. Find seeds

    Watch the winning videos. For each one ask: what is the non-obvious takeaway? Do you have your own, better take on it? If yes, that is your new seed.

  4. Keep a shortlist

    Topic plus seed is one validated idea. Collect 10 to 20 per round, with a link to the original, so you can compare while remaking.

Remix, do not copy

Hold most of the blocks constant and change one or two. You build on something that already worked instead of starting from zero.

The hook

The first seconds

The hook decides whether someone stays or scrolls on. Most people think it is about the spoken words. Those matter least.

The 3 layers

Important in this order. Most videos are watched without sound, so text and visuals carry the load.

Text on screen

The first thing processed, because the eye darts to the text. Counts the most.

The visual shown

Reinforces text and sound, or creates confusion. Stops the thumb, or it does not.

The spoken word

Matters least, because many watch without sound. Spoken words with weak visuals and text rarely work.

The 2 musts

Immediate clarity

In the first sentence it has to be clear what the video is about. Enough context for the viewer to decide: is this for me? Speak simply, short sentences, straight to the point.

Rule of thumb: if your first sentence does not open the topic right away, you have not compressed it enough yet.

A curiosity loop

Curiosity comes from contrast, between what the viewer believes and what you are about to explain. The bigger the gap between expectation and reality, the stronger the hook.

Instead of 'how to save on taxes', try: 'stop putting money into the obvious thing, and do something you have never heard of instead.' The contradiction creates curiosity.

Strong versus weak

Same content, two hooks. One only announces, the other breaks an expectation. That difference decides whether someone stays.

Topic: saving on taxes

Weak

In this video I explain how you can save on taxes.

Strong

Stop paying into the obvious thing. There is a way nobody tells you about.

Why: The weak hook only announces. The strong one breaks an expectation and creates curiosity.

Topic: productivity

Weak

5 tips for more productivity.

Strong

Most to-do lists make you slower. One thing changes that.

Why: List hooks wear out. A contradiction to the status quo pulls.

Build your own collection: save videos whose hook stopped you, and note why.

Creating contrast

Contrast creates curiosity, curiosity holds attention, and that is exactly what the algorithm rewards. Two ways, alone or stacked.

Against the common belief

What does your audience hold to be true? Find the angle that contradicts it most. If everyone says 'post daily' and you show that three times a week works better, that is contrast.

Against the visual habit

Break the pattern your niche is used to. If it is usually calm and tidy, try loud and fast. If it is high-gloss and polished, try raw and unedited.

Trust through visible proof

Trust is built through pattern recognition, often before you have said a word. Set up your set, clothes and location so the viewer's brain does the math on its own.

  • Doctor: film in scrubs or in a clinical setting.
  • Mechanic: in the shop, with tools around you.
  • Lawyer: in an office that looks like a law office.

The dopamine ladder

Creator Kallaway (Kane Kallaway) thinks of a video as six stages that grab the viewer one after another. The hook is only the first. Serve all six and you keep viewers in the video and turn them into followers.

  1. 1

    Stimulation

    Message

    The first second or two. Motion, color, contrast, brightness: the visual has to stop the thumb before any content lands. Build your own visual signature.

  2. 2

    Captivation

    Message

    Open a question in the viewer's mind. That's the curiosity loop from above: contrast between the familiar and the new, or a surprising premise. If many people skip, the question wasn't relevant enough.

  3. 3

    Anticipation

    Message

    The viewer starts guessing. The more they guess, the stronger the pull. Clear facts feed the guessing, and tension peaks right before the answer. Ask yourself: which answer should the viewer expect?

  4. 4

    Validation

    Message

    You resolve the question and close the loop. Only the resolution makes the video rewarding, as entertainment or as insight.

  5. 5

    Affection

    Messenger

    The viewer starts trusting you. Presence, demeanor, smiling, and most importantly: you solve their problem. Nearly impossible with faceless content.

  6. 6

    Revelation

    Messenger

    You become a reliable, consistent source. Easier in education than in entertainment, because you can solve real problems again and again.

The first four stages carry the message. The last two decide the messenger, meaning whether someone follows you for good.

Watch Kallaway's video

Funnel

From view to customer

Reach alone pays no bills. The path from viewer to customer is simpler than most think, but it needs a few steps.

One principle I will stand by: social media is not for selling. It is for building trust and acquiring email addresses. Selling happens later, over the list.

The activation flow

Each step has one job. You make a video, offer a fitting free resource inside it, and lead interested viewers step by step onto your list.

Video
Comment as CTA
Auto-DM with link
Lead magnet
Email address
Email sequence

CTAs that do not feel like CTAs

Build the call natively into the video or the bio. Instead of 'buy now', point to a free, fitting resource. Whoever wants it comments a keyword and automatically gets the link by DM.

The 7 lead magnets

A lead magnet is something valuable you give away for an email address. Like an impulse buy, only free. From simple to advanced:

Single download

A PDF, Google Doc or Notion page that solves one specific problem. Easiest to build. Checklist, template, framework.

Email course

An automated series that teaches something over several days. Exactly like the one you are reading right now.

Free community

Access to a free group or forum.

Waitlist

Early access or sign-up for something coming soon.

Product landing page

A free trial or account for a tool or piece of software.

Newsletter sign-up

Just the list sign-up. Least effective, but better than nothing.

Sales page

Straight to a purchase page for a product, course or service.

Start with the single download. A Google Doc behind an email form is enough. Do not make it bigger than it needs to be.

Growth

Seven channels apps grow through

Starter Story asked seven founders what made their apps take off. Seven tactics, each backed by a real app. Here's the essence, freely retold:

01

User generated content

Test lots of formats fast until one takes off. The founder in the cheat sheet stuck with user videos and grew on a single format on a single platform.

  • Test many content types in a short time.
  • When one works, double down on it.
  • One platform and one format are enough for early traction.
02

Influencer marketing

Small budgets work in niches if you talk to creators directly and structure deals properly.

  • Build credibility first (profile, references), then reach out.
  • Get on a call fast: deals close easier in direct conversation.
  • Protect deals with upfront payments plus view guarantees.
03

Build in public

Show openly what you're building. The trust and community support bring early users and first revenue.

  • Talk to users early and validate use cases.
  • Launch fast, a waitlist captures the interest.
  • Community hype accelerates adoption and sales.
04

Reddit as a distribution channel

Authentic build stories in the right subreddits bring thousands of users and honest feedback.

  • Technical transparency builds trust and engagement.
  • A single viral post can deliver your first users.
  • Feedback from the comments flows straight into the product.
05

App store optimization

Fitting keywords, localized screenshots and ratings at the right moment drive organic downloads.

  • Use location- and context-specific keywords, the way people actually search.
  • Adapt screenshots to local contexts.
  • Ask for ratings in golden moments, right after a success in the app.
06

Reviews as a growth lever

Ask friends, family and early users for reviews at the right moment. That lifts rankings and with them organic growth.

  • Ask for reviews when users are happy with the app.
  • Build requested features and follow up personally.
  • Positive reviews improve rankings, which brings downloads.
07

Turning trials into customers

Guide trial users with emails, videos and personal onboarding. That's how a test becomes a subscription.

  • Provide simple onboarding plus helpful resources.
  • Email touchpoints keep trials engaged.
  • Make a clear upgrade offer near the end of the trial.

The seven tactics are our short version of the Mobile App Growth Cheat Sheet by Starter Story (via HubSpot), with video lessons from the seven founders. You'll also find Starter Story in our recommendations.

Get the original cheat sheet

AI toolkit

Skills and MCPs for your content

Five skills and one MCP that speed up content work in Claude. Each entry is one command, installed in seconds.

Remotion

Build videos programmatically in React: animated explainers, caption styles, series formats.

bash
npx skills add remotion-dev/skills

Competitive Ads Extractor

Pulls your competitors' active ads and analyzes hooks, CTAs and offers.

bash
npx skills add composiohq/awesome-claude-skills --skill competitive-ads-extractor

Deep Research

Multi-step market and topic research in the background, backed by sources.

bash
npx skills add composiohq/awesome-claude-skills --skill deep-research

Content Research Writer

Researches a topic, writes the draft and delivers the sources with it.

bash
npx skills add composiohq/awesome-claude-skills --skill content-research-writer

Voice DNA Creator

Learns your writing style from real texts and applies it to new drafts.

bash
npx skills add az9713/ai-co-writing-claude-skills --skill voice-dna-creator

Plus one MCP for audio

ElevenLabs

The official MCP server for voiceovers and text-to-speech straight from Claude. The free tier includes 10,000 credits per month.

bash
claude mcp add elevenlabs --env ELEVENLABS_API_KEY=YOUR_KEY -- uvx elevenlabs-mcp

For images and videos we use Higgsfield. Its MCP card lives with all the others in the building chapter.

See the full MCP list

System

Get better with every video

Good creators do not guess. They check, measure, and change one thing at a time. Three tools for that.

The 11-point check

Run it before posting. Each point is an honest question to your video.

  • Format

    Does this format frame the information best?

  • Topic

    Is this relevant to your ideal viewer, not just to you?

  • Seed

    Is the take non-obvious? Does it challenge the common view?

  • Substance

    Can the viewer act on it afterwards, or is it just entertainment?

  • Visual hook

    Does the visual stop the thumb? Is there contrast and movement?

  • Text hook

    Does the on-screen text create curiosity? Does it match visual and sound?

  • Spoken hook

    Do the first seconds create clarity and curiosity?

  • Story structure

    Does the story flow logically? Are you building tension or just listing?

  • Call to action

    Is there a clear next action for interested viewers?

  • Layout

    Does the execution fit your strengths, for example talking head despite camera shyness?

  • Elements

    Do pacing, cuts, captions and music match the content?

Analyze in batches of 10

Do not judge video by video, but in blocks of ten. That way you spot patterns instead of chance.

  1. Pull the data

    Put views, new followers and conversions of all 10 videos side by side and rank them best to worst.

  2. Find outliers

    Was there a clear outlier on top? Run the 11-point check and figure out what you did differently.

  3. Form a hypothesis

    What did the winners have that the losers did not? For example stronger text hooks or more relevant topics.

  4. Test one thing

    In the next block of 10, change only one block, not everything at once. Then measure again.

The honest timeline

At 4 to 5 videos per week. No shortcut, but a realistic path.

Phase 1: Build the skill

Month 1–3, ~50 videos

The hardest phase. Every video feels awkward, the hooks are not great yet. That is normal, you are building the muscle.

Goal: enough reps for clean data.

Phase 2: Build systems

Month 4–6, ~50 videos

Patterns emerge. You find outliers and double down. You test in batches instead of guessing. The machine starts to run.

Goal: lock down two or three working formulas.

Phase 3: Compounding

Month 7–12, ~100 videos

Content stacks up, trust compounds, the system runs. You no longer wonder whether it works, you just optimize.

Goal: scale output and refine the funnel.

Explain twice for teaching content

Explain a thing twice. Once normally for the experts, then again very simply, as if explaining it to a child. Both stay, because both find something that fits them.

My experience

3 weeks of an internship. 10.6M views.

I used the World Cup as a reason to become a creator myself. Not in theory, but for real. These are the numbers after 3 weeks of a social media internship on a World Cup tour.

10.6M

Views

700,000+

Interactions

8,600+

Followers

The full story

How an 11-second dog video goes viral, why the internship suddenly turned into press work, and what I learned about being a creator.

Read the full story