Creator Marketing Playbooks

Should you work with the same creator again?

What a finished job tells you about the creator, what it doesn't, and what a second job decides from scratch.

What carries is the file you just wrote: how the job ran.

The campaign is done. There's another one starting, and when somebody asks who should make the video, the first name anybody says is the creator from last time.

The question sounds easy. Did it go well? But that's two questions, not one. One of them is about a video that already exists, and only the other one is about the person you'd be booking.

What did working with them once tell you?

Before the first job, almost everything you knew about them you could read off their account, or ask them for. Their recent videos, the comments under them, whatever branded content was already there. That's what you picked them against, and it was the right thing to pick against, because those checks answer whether somebody fits the campaign you were about to run.

A finished job leaves you something else, and it's the kind of thing nobody can read off an account. Call it the file about them.

Whether the things your brief fixed came back intact. Your brief fixed three things — the sentence a viewer should walk away with, the content format, and the element that had to stay intact. The draft either carried them or it didn't, and you watched that happen.

What one round of changes cost. Going in, you priced rounds against your own turnaround and left room for what you couldn't list yet. Now you have one round behind you: what you sent, what came back, how long it sat.

What they asked before shooting. Questions that arrive before the camera does tell you two things — where your brief was thin, and what they wanted settled before they'd shoot this one.

A lot of what the campaign left you isn't about them at all. It goes in a second file, the one about the campaign. The window was too short. The launch moved. The plan had no line for putting money behind anything. Those read true no matter whose name goes in the next brief, which is exactly why we've said before that what you file about a campaign and what you file about a creator don't belong in the same place.

For the bike light: the spring campaign sold the front light, and the sentence was that it clips on and off without tools, so it comes inside with you. The brief fixed three things: that sentence, one continuous take at the bike, and the clip reading as a clip rather than a strap. All three came back in the draft. One note went out — a glove was covering the clip in the shot that mattered — and it came back answered the next day. Before filming, the creator asked whether the bike had to be their own. That's the file about them. That the launch slid a week and ate the front of the window is the file about the campaign.

Why isn't how the video did part of this?

Because whatever that question answers, it isn't about the person you'd be booking.

Judging a posted video checks it against the sentence you wrote and the list you built from that sentence. It's a judgment about a video against the sentence it was made to land, and the same guide names the case that can void it: the sentence existed in your head and never made it into the brief. Then the brief never carried it, the video was made from the brief, and nothing on the video tells you which case you're in.

So a video that reads like it missed isn't a fact about the creator at all. Before you can even say it missed, you have to go back to the brief, and the miss may turn out to be a fact about you.

The other direction fails too, and it's the one that costs you, because nobody argues with it. The video landed, so book them again. What landed was a sentence, in a content format, in a campaign. It isn't a reason attached to a person.

For the bike light: the list said people would ask how it mounts and would repeat the no-tools line back. The mounting question came back; the no-tools line didn't. That split isn't in either file — it's what the next brief has to answer. The window the launch ate is the campaign file.

What carries into the second job, and what doesn't?

What carries is the file you just wrote: how the job ran. That's the whole of it.

Three things do not carry, and each one is a decision the second job makes again from the start.

Fit. The next campaign has its own sentence, and the test is the one you ran the first time: has the moment that sentence describes ever appeared in their work, and are the people watching there for it? Having worked together isn't an answer to that — it just means you're running the test on a name you already know.

Permission. What you bought last time was bought against last time's plan, and an expired term doesn't stretch to cover a new one — new reuse is a new ask. If the new plan wants paid ads, it buys its own, from your handle or theirs, whichever the plan asks for.

Price. The number you remember is attached to the job you ran, and the pay guide has already said what to do with it: put the two jobs side by side before you let the earlier number do any work.

For the bike light: the second campaign is the rear light, in the fall, and its sentence is that it's the light you don't have to remember, because it stays on the bike. Fit gets checked against that sentence, not against the spring. The spring reuse term ended with the season, so the fall plan's paid ads are a new ask. And the spring number priced a job that had no paid ads behind it at all.

Working out whether the name you know is the right one for the next campaign? Bring the file and the campaign and we'll run it with you. Book a demo →

So when do you go back, and when don't you?

Go back when two things are true at the same time: the new campaign's own test picks them, and your file says the job ran the way you'd agreed it would.

Don't go back when the test doesn't pick them. That's the ordinary case and it isn't about them — a sentence somebody isn't right for is just a sentence they're not right for. And don't go back when the file says the agreement didn't hold. That one is narrower than it sounds. The things your brief fixed didn't come back, and the note asking for them went unanswered. A date you'd both written down went past without a message. Those are events, and they're in the file because they happened, not because of how the campaign felt.

What doesn't go in the file is what the events say about the person. You have one job's worth of evidence, and it's evidence about that job. Write down what happened and stop there. It's the same instinct behind our position that an empty branded content history is something you don't know rather than a mark against anyone — don't let a file answer more than it was keeping track of.

For the bike light: nothing in the file about them is a missed date or a note left unanswered, and the entries that record how the job ran say it ran as agreed. That's one half. Run the test on the rear light sentence, and if it picks them too, you have both.

What's different the second time you brief and price it?

Two things get smaller.

The first is the facts in the brief. They know the product now, so you're not writing that part twice — but what the brief has to fix doesn't get shorter for it. The new campaign writes its own sentence, and that's where what the last video taught you lands, whoever ends up filming.

The second is the room you priced on purpose. Going into a first job, some of what could come back wrong isn't listable yet, so the room you leave for it is a judgment. Now one of the listable fixes has been priced once — you've seen a note go out and come back, and you know what that one took. The floor is still set by the most expensive fix you can name, and the room above it is still a judgment. You're just making it against something you watched instead of against nothing, and that's what lets the room be smaller.

For the bike light: the mounting question came back on the front light video, and the rear light mounts too. So the fall sentence answers it, and the brief still fixes three things — that sentence, a content format, and the mount reading as a mount.

Frequently asked questions

They were good to work with. Can we just keep them on for the next few months?

You can set up an arrangement that covers several videos.

What it can't settle ahead is anything belonging to a campaign nobody has scoped. Each video still needs a sentence, and permission is still bought against a plan that exists. Write the arrangement so those stay open, or you're agreeing terms for a job you haven't described yet.

The video was fine, but everything took longer than we expected. Is that a reason not to go back?

It's a fact for the calendar, not a verdict.

If nothing you agreed on fixed the timing, then nothing was broken and what you have is a turnaround you've now seen once. Size the next campaign's dates against it, and say out loud that it's one observation. If those dates only work when nothing slips, that's a fact about the dates.

See what this looks like for your brand

Got a name you'd go back to and a campaign you haven't tested it against yet? We'll go through both with you.

Or email us: business@popow.ai

Ready to run this with creators?

See how Popow briefs, matches, and ships TikTok-native content for your campaign.

Book a demo →business@popow.ai