Good work isn't enough — it has to be seen
For five chapters you built: you got to know me, learned the request, set up flows, opened the workshop, made your site and your system. But if nobody sees what you built, it may as well not exist. This chapter is the showcase: what to film, how to cut it, what the algorithm actually looks at, how to turn an incoming message into work. The camera isn't mine — it's yours. I'm the editor, the writer and the analyst.
By the end of this chapter
- You'll tell reach, engagement and conversion apart and know which to chase when
- You'll understand honestly what the algorithm is — and isn't
- You'll build the first three seconds deliberately
- You'll write your own content pillars
- You'll cut a video with me from raw footage to publish
- You'll set up a smart DM that turns messages into work
- You'll know which number matters and build the weekly content machine
Contents
01The invisibility problem
There are more people doing good work that nobody knows about than people doing bad work that everybody knows about. That's unfair, but it isn't a secret: visibility is a separate skill and it can be learned. It doesn't replace your ability — it sits on top of it.
I'm not teaching advertising in this chapter. What I'm teaching is narrower and more useful: a system for putting the work you do in front of the people who need to see it, regularly. I'm underlining the word system, because sharing when inspiration strikes is not a strategy.
A warning, right at the start: producing content doesn't replace the system you built. An account that's visible with nothing behind it gets looked at once and dropped. If you did the previous five chapters, you have something real to say. A showcase only works in front of a full shop. A concrete example: this site’s own /icerik-uret skill turns a chapter into five posts, a reel script, a carousel and a DM template; the 7 campaigns it has made so far sit in the site’s studio/ folder. The repository isn't public; in the FYOS demo on the home page, the Studio card counts that folder.
Do this now: Complete this sentence: «I do ___ for ___.» Fill in the blanks. That sentence is the starting point for this entire chapter; if it isn't clear, no piece of content will be.
02Three words: reach, engagement, conversion
In every chapter three words opened the whole field. Same here — and confusing these three is why most of the time spent on social media is wasted.
The rule that follows: boast with the top number, decide with the bottom one. View counts give you a lift; client counts pay the rent. When you judge your content the question isn't «how many saw it» but «how many wrote to me».
Do this now: How many people messaged or called you in the last thirty days? Write the number down. At the end of this chapter we'll measure that number — not views.
03What the algorithm actually does
Most of what's said about the algorithm is made up. Let me be honest: nobody outside knows the exact rules, they change often, and they differ from platform to platform. But what it's trying to do is something everyone knows, and that doesn't change.
sent to someone?
came back?
Three practical consequences: completion rate (do people stay to the end), shares (do they send it to someone) and consistency (how often you publish). All three are in your hands. Tactics like «post at this hour», «use this many tags» are seasonal; don't chase them.
One more thing: the algorithm isn't punishing you. A bad week is a measurement, not a penalty. The «I'm a prediction machine» idea from the chapter about me applies here too — the system looks at the past and predicts; it doesn't pass judgement.
Do this now: Look at the average watch time of your last ten posts. Compare the best three with the worst three: what's different in the first three seconds? Write the answer down — that's the next section.
04The first three seconds
A video's fate is decided in the first three seconds. People sit there with a thumb ready to scroll; you have to give them a reason to stop. That's called the hook, and it isn't an accident — it's built.
- 0.0s Visual stopper An unusual frame, a movement, a face. The thing that breaks the scroll reflex.
- 0.5s The promise One sentence saying what they'll get: «in three minutes you'll build this».
- 1.5s Tension A question whose answer comes at the end, or a contradiction. A reason to stay.
- 3.0s The start The actual content. A viewer who gets here usually stays to the end.
I can write the hook, but not well on my own: I need to know what you're going to say. Give me the video's transcript and I'll pull out five different hooks; you pick the honest one. I'm underlining honest: a hook that promises something the content doesn't deliver wins the first three seconds and loses the next thirty, along with the viewer's trust.
Do this now: Describe one of your best pieces of work and tell me: «Write five hooks for this; eight words maximum each; no exaggeration.» Read all five, pick one, throw away the rest.
05Your content pillars
The question «what should I post today», asked every day, never gets answered well. The fix is deciding on three or four pillars: every post falls into one of them, so you don't start from scratch each day.
Once the pillars are set, producing ideas is my job. Describe your pillars and your work and I'll pull out ten ideas per pillar — forty ideas, a month of programming. Pick the ones you can actually make; an idea you can't make isn't an idea.
Do this now: Write your own four pillars. The ones above are examples; yours may differ with your work. Then hand them to me and ask for ten ideas each.
06From idea to video: editing with me
Editing is the most time-consuming part of making content — and the part that can be shortened the most. I can't see your video, but I can read its transcript; and most editing decisions are made in the words, not the picture. The animation below shows how raw footage becomes ready to publish.
The second step is the heart of it. You give me the transcript and say: «Cut this down to ninety seconds. Which sentences stay, which go — list them with timestamps. Don't break the meaning, drop the repeats and the pauses.» I produce a cut list; you cut at those times in your editing program and finish in half an hour.
Let me also say what I can't do: I can't watch your video, judge the picture quality, choose music or grade colour. Wherever there are words I'm fast; wherever there's picture I trust you. The honesty from the chapter about me applies here too.
Do this now: Take a raw clip sitting on your phone and get its transcript (your phone or your editing program can do this). Hand it to me and ask for a ninety-second cut list. Publish a video tonight.
07Captions, cover and title
Once the video itself is done, three small jobs remain, and all three affect whether it gets watched more than the content does. In order:
Captions
Most people watch with the sound off. A video without captions is a silent film for all of them. Editing programs generate them automatically; you only fix proper names and terms. Give me the transcript and I'll catch the spelling mistakes too.
Cover
The square that shows in your profile grid. It shouldn't be the video's best frame but its clearest frame — with three words written on it. Six months later, when someone opens your profile, the covers are their menu.
Title
The first line matters as much as the hook, because in some places it's read before the video. Keep it short, repeat the promise, end with a single question — a question brings comments, and comments are engagement. Give me the transcript and I'll write ten titles.
All three are five-minute jobs and all three get skipped. The way to make them a habit is to count them as part of editing: a video isn't finished until its captions and title are written.
Do this now: Give me the transcript of the video you published in 06 and ask for ten titles. Pick the most honest one — not the most tempting.
08Rhythm: little, but unbroken
The most-asked question is «how many posts a day» and the answer is nearly always the same: as many as you can sustain. Two videos a week, unbroken for six months; seven videos a week, abandoned after three. The second doesn't come close to the first.
Two posts a day for a week, then a month of silence. It exhausts you and it never teaches your audience when to expect you. And every return feels like starting over.
Twice a week, always the same days. The audience waits, you plan, I prepare ahead. Even if you're ill for a week there are two videos in stock.
What makes rhythm possible is batch filming: shoot four videos in one sitting, publish over four weeks. Filming is a day, editing with me is a few hours, publishing is automatic. The schedule trigger from the agent chapter works here too; but publishing by hand is fine as well, as long as the date is fixed.
Do this now: Pick two days of the week in your calendar and mark the next eight weeks. Then set aside a day for the first batch shoot. Rhythm doesn't happen without dates.
09Making DMs smart
If the content works, messages arrive. And this is exactly where most people lose: a message comes, it's seen three days later, answered, forgotten. The flow you built in the agent chapter was for exactly this gap; now we connect it to a real use.
- 1A new message arrives, the flow wakes
- 2Me: an order, a question or a collaboration?
- 3If it's a common one, I draft a reply
- 4Notification to you: draft + approve / fix
- 5Into the CRM: person, source, subject, date
The limit of automatic replies is this: greeting and routing can be automatic, a relationship can't. A machine writing «Hello, your message reached me, I'll get back to you today» is fine. A machine writing «I'll do that for you at this price» is not. The irreversible-tool rule from the agent chapter here means making promises.
Do this now: Look at your last fifty messages and write down the three most common questions. Write the answers to those three with me and put them in your flow. From tomorrow those three answer themselves.
10Comments: the cheapest growth
The comment section is a continuation of the content, and most people leave it empty. Yet someone who wrote a comment has already shown their interest; a good reply there brings work faster than a new post.
The first hour
Reply to comments in the first hour after posting. It keeps the conversation alive and content gets shown more in that window. This isn't a trick; systems like live conversation because people do.
Turn the question into content
If the same question came three times, that's a video. Comments are a free survey telling you what your audience wants taught. Keep a list; «what should I post today» never gets asked again.
The hard comment
When criticism comes, don't get defensive. If it's right, accept it and fix it; if it's wrong, answer briefly and kindly or not at all. Most onlookers remember not the comment but how you behaved.
You can have me write comment replies, but don't have me write all of them. Your audience knows your tone and after a while senses which reply didn't come from you. Where I'm good is drafting twenty replies; your reading and changing them is essential.
Do this now: Reply to every comment on your last post today. Then note the most repeated question — that's this week's video.
11Where does the click go
Someone liked your content and tapped the link in your profile. Now the critical ten seconds: what does the page say? On most accounts it's a home page and the visitor gets lost. What it should be is a page with one job.
What matters isn't the page's length but that it has one job. «About», «blog», «references» links don't belong on this page; they scatter attention. The visitor either presses the button or leaves — and both are information.
Do this now: Describe these four blocks to me and say «build a page with one job, connect the form to this flow». Then replace the link in your profile with it.
12Campaigns: paying to go faster
Advertising speeds up something that works; it doesn't rescue something that doesn't. That sentence is this whole section. Putting money behind content that has never brought work organically is like hanging more signs on an empty shop.
- 1Wait for proof first: have a piece of content that has brought work at least once organically. That's what you put money behind, not something new.
- 2Start small: your daily budget shouldn't exceed a tenth of what one client earns you. It's a test budget, not an investment.
- 3Change one variable: if you change the image, the copy and the targeting at once, you'll never know which one worked.
- 4Count conversions, not clicks: messages received and work won. A click is an indicator; a client is a result.
- 5Wait seven days, then decide. Shutting it down based on day one's numbers is quitting before measuring.
Honest advice: most people finishing this course won't need advertising at all. Two good videos a week, answered messages and a page with one job is usually enough to fill a small agency's capacity. Think about ads when your capacity sits empty.
Do this now: Don't run any ads yet. Instead write down a number: what do you earn from an average client? A tenth of that will be your daily test budget later.
13Which number matters
Platforms show you dozens of numbers and most of them seem designed to keep you busy. An agency needs to look at four numbers once a week; the rest is curiosity.
Write these four numbers once a week, same day, same place. Adding one more sheet to the spreadsheet from the lab chapter is enough. After a month you'll see a curve; after two you'll stop guessing what works and start knowing.
Do this now: Add a sheet called «week» to your CRM spreadsheet: a date and these four columns. Fill in this week's row now — measuring starts today.
14The weekly content machine
Everything in this chapter comes together in one weekly loop. Once you set it up it turns the same way every week and takes about five hours in total.
- Collectquestions from comments and messages
- Choosetwo ideas by pillar, have me write hooks
- Filmone sitting, raw, not perfect
- Edittranscript → cut list → captions → title
- Measurewrite the four numbers, adjust next week
Do this now: Spread these five steps across the days of your week. Collect and choose on Monday; film Tuesday; edit Wednesday; measure Friday. Put it in your calendar as a repeating event.
15Brand voice: what makes you you
Hundreds of accounts do everything in this chapter and most look like each other. The only thing that separates them is voice: how you speak, what annoys you, what you never do. A voice isn't a slogan, it's a repeated attitude.
«I'm not going to tell you I'll finish this in three days. It takes two weeks, because I don't start without measuring.»
There's an attitude, a boundary and a reason. Nobody else could write that sentence.
«Quality service, fair prices, customer satisfaction is our priority.»
The sentence everyone says and nobody believes. Delete it and nobody notices.
I can't find your voice — but I can capture it. Give me ten pieces you wrote yourself and I'll extract your style: your sentence length, the words you like, the ones you avoid. We put that description in a file as in the workshop chapter, and from then on every draft I write sounds like you. But you'll write the first ten; there has to be an original to copy.
Do this now: Collect ten pieces you wrote yourself (posts, messages, emails) in one file and give them to me: «extract my style, write it as rules». Read what comes back — you'll recognise yourself.
16The honesty line
There's a lot you can do with these tools and not all of it should be done. Four boundaries; two legal, two simply because they're right.
Don't show work you didn't do
Putting work you found online into your own portfolio is called fraud and it gets caught sooner or later. If you have little work, show little; one real job beats ten stolen ones.
Don't hide what's artificial
Don't present AI-generated images or audio as real. Getting help with editing and copy is normal; publishing a sentence a person never said in their voice is not. In some countries that's also a crime.
Don't buy comments or followers
Purchased engagement lowers the rate at which real people are shown your work — so you're paying for invisibility. And when it gets cleaned up one day, nothing is left behind.
Don't share client work without permission
The best content is real work, but you have to ask first. Add one line to your contract: «may I use this work in my marketing?» Most clients say yes; sharing without asking ends the relationship.
These aren't a morality lecture, they're business advice. Visibility is a long game and your only capital in it is trust. Once it's shaken, the cost of winning it back is higher than everything you gained by the shortcut.
Do this now: Look at your profile and portfolio: is everything you're showing genuinely yours and permitted? If not, take it down today. Then add that one line to your contract.
17Five common mistakes
1 · Not publishing until it's perfect
Ten good videos sitting in a folder are worth less than one average video published. Publish, measure, fix. Your first videos will be bad anyway; so get past them quickly.
2 · Starting with a greeting
By the time you've said «hi everyone, today I'm going to…» the viewer is gone. Let the first word belong to the subject; put the introduction at the end or leave it out.
3 · Deciding by view count
Content that gets lots of views and brings no work is worse than content with few views that brings two clients. Decide with the bottom number.
4 · Leaving messages waiting
Doing the content work and getting back to a message three days later throws all that effort away. Automate the greeting, answer yourself — but the same day.
5 · Having me write everything
The draft is mine, the voice is yours. Text published without a single touch starts sounding the same, and your audience notices before you do.
Do this now: Check these five against your last ten posts. Especially 2 and 4 — both are the easiest to fix and have the biggest effect.
18What everyone asks
Do I have to be on camera?
No, but there's a cost. A face is the fastest way to build trust; faceless accounts grow too, but more slowly and with more content. Alternatives: hands and the work on screen with your voice; screen recordings; before–after shots. Camera fear usually passes by the third video — everyone who has done it says the same thing.
If the algorithm changes, is it all wasted?
Nothing you learned in this chapter depends on the algorithm: the hook, the pillars, the rhythm, the editing, answering messages, the one-job page, the four numbers. All of it stays the same across platforms. I wrote it that way deliberately — I left out «post at this hour» tactics because they go stale in six months.
How many followers before work starts coming?
The link between follower count and work is weaker than people think. An account with five hundred followers showing the right work gets steady enquiries; one with fifty thousand posting generic content may get none. What matters is that a viewer understands what you do in ten seconds and knows where to write.
Can you edit my videos?
Partly. I can't watch the video; but I can read the transcript and write the cut list, fix captions, write the hook, the title and the description — that is, most of the decision-making in editing. You'll do the cutting itself, or have an editing program do it. In practice this turns a three-hour edit into forty minutes.
Should I post the same thing on every platform?
The same idea yes, the same file no. Every platform has its own dimensions, its own length and its own habits; a video with another platform's logo on it gets shown less everywhere. Give me the transcript and I'll write a version for each platform — that's a five-minute job.
When will I start seeing results?
Honest numbers: first messages within a few weeks, a steady flow of work in three to six months. That time can be shortened but not skipped, because people trust you after seeing you several times, not once. That's why rhythm matters even more than the quality of the content — that's what section eight was about.
Do this now: If a piece didn't go the way you expected, give me its transcript and its numbers and ask «why didn't this land?» We'll look at the hook, the length and whether the promise matched.
19Exercises
1 · One sentence and four pillars
Write the sentence from 01 and your four pillars from 05. Then hand them to me and ask for ten ideas per pillar.
Goal: to never ask «what should I post today» again.
2 · Edit your first video with me
Shoot something raw, get the transcript, apply the five steps from 06 and publish it this week.
Goal: to experience once that editing isn't frightening and doesn't take hours.
3 · Connect the page and the DMs
Build the one-job page, change your profile link, turn on the DM flow and send yourself a message.
Goal: to see that interest from content reaches the CRM without getting lost.
4 · Four weeks of rhythm
Prepare four videos in one batch shoot, publish two a week, and write the four numbers every Friday.
Goal: to be looking at a curve instead of guessing, four weeks from now.
Do this now: The first today, the second this week. The third next week. The fourth already takes a month — start it today.
20Test yourself
Answer it yourself first, then open it.
Reach, engagement, conversion — in one sentence?
Reach is how many saw it, engagement is how many stopped, conversion is how many wrote. You boast with the top number and decide with the bottom one.
What is the algorithm trying to do?
Not to reward you, but to keep people on the screen. It shows your content to a small group; if it's watched to the end and shared it spreads, otherwise it stops. The details change, this logic doesn't.
The four parts of the first three seconds?
Visual stopper, the promise, tension, the start. None of them is a greeting; by the time «hi everyone» ends, the viewer has gone.
What are content pillars for?
They give every post somewhere to land, so you don't hunt for ideas from scratch each day. Roughly three teaching, one selling. An account that only sells turns into a billboard.
What can and can't I do in editing?
I can: a cut list from the transcript, caption fixes, the hook, the title, per-platform versions. I can't: watch the video, choose the footage or the music, grade colour. Words are mine, pictures are yours.
What's the right number for rhythm?
The number you can sustain. Twice a week for six months unbroken; seven a week abandoned after three. The second doesn't come close. Batch filming is what makes rhythm possible.
What do you automate in DMs, and never?
Greeting, classifying, drafting common answers and the CRM record can be automatic. Quoting a price, making a promise, committing to a date never. Irreversible things stay with your approval.
The four blocks of a one-job page?
One-sentence promise, three proofs, what happens next, and one button. No menu, no blog, no distracting links. Put two options and neither gets chosen.
The four numbers to check weekly?
Enquiries, saves and shares, watch time, posts published. Followers aren't on the list — followers are a result, not a target.
When do you advertise?
When you have content that has brought work at least once organically. Ads speed up what works and don't rescue what doesn't. Budget a tenth of one client's value, one variable, seven days, then decide.
This chapter in one sentence
Say what you do in one sentence, split it into four pillars, build the first three seconds deliberately, leave the decision-making part of editing to me, answer messages the same day, send the click to a page with one job — then look at four numbers once a week and choose the next week accordingly.
