What a YouTube Growth Agency Should Do in the First 90 Days

You have just signed with a YouTube growth agency, or the contract is sitting in your inbox, and the question in your head is probably a simple one, which is what should actually happen between now and the end of month three so you can tell whether the money is doing anything. A YouTube growth agency is the team that takes over the work around your videos, the editing, the titles and thumbnails, the Shorts and the reading of your numbers, and changes those things on purpose so the channel reaches more of the people it was made for. I run one of these, Pixel Samy Studio, which is a YouTube, podcast and Instagram growth team of close to fifty people sitting in-house in Dubai, and the group also runs seven YouTube channels of its own, each one past 100,000 subscribers, so what follows is the plan I would want if I were sitting on your side of the table.

Nobody can promise you a subscriber number on day ninety, and an agency that does is guessing out loud, so what I am going to describe is the work, the order it should happen in and what you should be able to see at each checkpoint, right, because those are the only parts anyone actually controls.

What should a YouTube growth agency do in the first 90 days?

A YouTube growth agency should spend the first 90 days in four overlapping stages: an audit of your YouTube Studio data in the first two weeks, packaging tests on titles and thumbnails from week three, retention fixes inside the edit once the clicks are understood, and Shorts plus a first series as the discovery layer by weeks nine to twelve. Each stage changes one thing at a time.

The order matters more than people expect. If you change the thumbnail style, the video length, the upload day and the intro all inside the same fortnight and the numbers move, nobody in the room can tell you which change did it, so you have learned nothing you can repeat. We review the numbers once a month and change one variable at a time, which feels painfully slow in week two and feels very sensible by week ten, because by then every decision on the channel has a reason behind it that you can point at.

I would rather know exactly why a video lost than get lucky on one I cannot explain.

What happens in the first two weeks of the channel audit?

The first two weeks of a YouTube channel audit should take your last twenty or thirty uploads one at a time in YouTube Studio and compare how often each was shown, how often it was clicked, how long people stayed and where its retention graph drops, so the agency can tell whether the channel loses people before the click or after it. That split decides the next ten weeks.

Google explains how impressions and click-through rate are counted, and it is worth reading yourself, because half the arguments I hear between creators and their agencies come from two people meaning different things by the same three letters. What I want out of an audit is not a forty slide deck, it is a short written list, something like these six uploads were shown a lot and clicked rarely, these four were clicked well and lost half the room in the first minute, and these two did both well and here is what they share. Basically, the audit should hand you a hypothesis you can test, and if all you receive is a folder of screenshots, ask for the list.

It also helps to open the channel logged out, the way a stranger would, with the last ten thumbnails side by side, and ask whether someone new could say in one sentence what it is for.

How should packaging tests run from week three to week six?

Packaging tests from week three to week six should put several title and thumbnail options against each other on real uploads, keep the winner, and then challenge it again. At ClickTheory, our thumbnail studio, the method is three thumbnails for seven days, then two new challengers against whichever one won, so the channel keeps learning instead of settling on its first guess.

Packaging comes before the edit in my order for a plain reason, right, which is that a video nobody clicks never gets the chance to prove its edit was good, and on a lot of channels I look at, the gap between the best and worst upload is explained more by the title and the thumbnail than by anything after the first ten seconds. Every thumbnail ClickTheory designs is built for that single video, never poured into a template, and the styles are tried on our own channels before they go near a client's. The case for giving packaging this much weight is made in titles and thumbnails are half the work, and our thumbnail design service shows what is made for each video.

By week four what you should be holding is evidence rather than a jump in the graph, meaning two or three tests finished, a note of which option won on click-through rate, and an agreed direction for the next batch.

Why do retention fixes come after the packaging work?

Retention fixes come after packaging because the audience retention graph only says something useful once enough of the right people are clicking. When the clicks are steady, the retention graph in YouTube Studio shows where viewers leave, often in the first thirty seconds, during a slow setup, or just before the payoff, and the edit can then be changed at those exact points.

Google's explanation of audience retention is worth five minutes, because once you know the difference between a gentle slope and a cliff you see your own videos differently. A cliff at second twenty usually means the intro promised one thing and then wandered off, so the fix lives in the edit, cutting the setup, moving the strongest line forward, getting whatever the thumbnail showed onto the screen early. A slow slide across the middle tends to be pacing, right, and retention editing for long-form YouTube goes through that kind of work in detail.

What I would watch, as the creator, is whether the agency is fixing your videos or quietly turning them into somebody else's. More zooms and faster cuts are not automatically better, and if a change makes you sound like a different person, your regular viewers will notice it before the graph does.

Where do Shorts fit when the goal is long-form growth?

Shorts fit into long-form growth as the discovery layer, meaning short vertical clips that reach people who have never seen the channel and give them a reason to try the full videos. Shorts should be cut from the long-form uploads, point clearly back at them, and be judged on whether they bring viewers into the channel, not only on their own view counts.

Most creators who come to us either ignore Shorts completely or run them like a second channel with a different personality, and both leave something on the table. What works better in my experience is that every long video produces a handful of Shorts built around one idea each, the strongest moment first, captioned for people scrolling with the sound off, with the long video it came from set as the related video. Our YouTube Shorts editing is built around that idea, and on the OS plan at 5,900 dollars a month it means thirty or more short-form pieces next to four long-form videos, all produced from one shoot day.

Weeks five to eight are where I would expect Shorts to be going out on a steady rhythm, right, because a few weeks of them is the minimum before anyone can say whether they are pulling people toward the long-form.

When should a new channel start a series?

A new channel should start a series once packaging tests and retention data have shown which topic and format people actually click and finish, which on most 90-day plans lands around weeks nine to twelve. A series turns that one proven idea into several connected videos grouped in a playlist, so a viewer who enjoyed one has an obvious next video waiting.

Starting a series in week one is a guess wearing a strategy costume, so I would rather spend the first two months finding the one topic where your click-through rate and your retention both sit above your own channel average, and then build four or five videos around it. Your own average is the only benchmark that means anything at this stage, right, because comparing yourself with a channel in a different niche tells you about their audience and nothing about yours.

What should you be able to see at week 4, week 8 and week 12?

At week 4 you should have the audit findings in writing and the first packaging tests finished; at week 8 a steady upload and Shorts rhythm, with retention changes visible in the edits; at week 12 a first series under way and a plain account of what changed, what it did and what gets tested next. None of these checkpoints is a promised number.

If you reach week twelve and nobody can tell you in plain words what was tested and what was learned, that is the conversation to have, whatever the subscriber graph is doing. The reverse holds as well, right, a channel can have a quiet first quarter on the numbers and still be in a far stronger position, because the tests have told you which direction to push. At the end of the day the first 90 days buy you clarity about your own channel, and growth, when it shows up, gets built on top of that clarity.

What is the creator's part of the first 90 days?

The creator's part of the first 90 days is turning up for the shoot or recording days, giving quick and honest feedback on drafts, and staying with the testing plan when a single week looks bad. An agency can carry the editing, packaging, posting and reporting, but the ideas, the face on camera and the final say on what sounds like you stay with the creator.

Let me ask you something directly, because it is the part people skip: when a test loses, are you going to want to throw out the whole plan, or will you let it finish? I have watched more channels stall because the owner changed direction every ten days than because of one weak thumbnail, and that pattern shows up in very different niches.

With us the practical side is one focused shoot day, or one recording, that turns into the month's content, and the editing, the packaging, the posting and the monthly review sit on our side, while the footage and every finished file stay yours because they belong to you. Our YouTube channel management explains the weekly rhythm once that is set up.

When is a growth agency the wrong hire for your channel?

A growth agency is the wrong hire when you cannot commit to recording regularly, when the channel has too few uploads to learn from, or when what you actually need is clean editing and nothing else. In those cases a solo freelance editor, or doing your own packaging for a few months, will teach you more for less money than any retainer.

I would rather say that on a call than take a retainer that cannot work, since a growth process needs material to test, and four long videos a month with Shorts alongside is roughly the volume where tests start to mean something. So for a channel built on long videos I usually point people at the OS plan when they ask Pixel Samy Studio about YouTube growth agency work, and our plans and prices show the rest, every plan billed month by month, so if the first quarter does not convince you, nobody is holding you to the next nine.

So the standard I would hold any agency to, ours included, is that by day ninety you understand your own channel better than you did on day one, with every test written down and the next one already chosen, because that is what lets every decision after it be made on evidence instead of mood.

So yeah. That's my way of saying it.