How to move from gut-based creator selection to performance-driven systems

Hey everyone,

Welcome back for another bite to chew on.

Creator marketing isn’t a game of luck, but that’s how most brands play it.

They hire a few creators, post a few videos, and wait for one to magically take off.

And when it doesn't perform, they blame the creator. Or the platform. Or just chalk it up to bad luck and try again.

The problem? You're treating creator marketing like a creative gamble instead of a measurable channel.

We sat down with Donatas Smailys, CEO & Co-Founder of Billo, a platform that helps brands find and manage creators. His team works directly with Meta and TikTok and has tracked over half a billion dollars in creator ad spend.

After seeing thousands of brands up close since 2020, his takeaway is simple: the ones winning with creators don’t get lucky. They built feedback loops that show what’s working and what to do next.

On the Menu:

Let’s get into it.

You can watch the whole conversation with Donatas here:

Listen on Spotify 🎧 and Apple 🍎 as well.

About 70–80% of what people see on social media now comes from accounts they don’t follow. That shift, powered by AI-driven feeds, created what Donatas calls a “black hole of content fatigue.”

The algorithm doesn’t reward consistency. It rewards volume. You need a steady flow of fresh creative just to stay visible.

That’s why the brands winning with creators aren’t posting a few videos and praying one takes off. They’re testing at scale, tracking every piece, and turning what works into repeatable systems.

Here’s what that looks like in practice:

But don’t confuse activity for progress. Volume alone doesn’t win. The content still has to connect. Humor, storytelling, and genuine emotion still outperform polish.

Most brands fail because they create without learning. The smart ones treat creator marketing like paid media: they test fast, double down on winners, and cut what doesn’t move the needle.

Billo’s data proves it. The brands testing at scale are seeing the biggest wins in health and wellness (35–40% of their clients) and fashion and beauty (around 25%). Meta Reels drives the highest ROAS by far—proof that structured testing beats random posting every time.

Systemizing creator marketing isn’t optional anymore. You’d never run Meta ads without tracking performance. Don’t run creator campaigns that way either.

Right now, you're probably picking creators based on their portfolio, hoping they perform, and finding out after you've spent the budget whether they can actually drive results.

That's the expensive part. You're paying to find out.

Billo is a creator marketing platform that lets you see how creators perform before you commit to a budget. Instead of hoping, you're choosing based on real performance data from 326,000+ video ads and $505M in ad spend.

Here's what that looks like:

The results speak for themselves: Brands using the platform have seen 120% increases in sales and 15% higher CTR on TikTok compared to content from other sources. 

That's what happens when you stop guessing and start using data.

Try Billo and get a $50 bonus when you purchase a $500 pack → 

The difference between random creator marketing and scalable creator marketing starts with one question: how do you choose creators?

For years, the answer was gut-based. "I like how they look." "Good energy." "They'd fit our aesthetic."

That caps your ceiling. Data-based selection scales. 

The single most important metric? Hook rate.

It's the first three seconds of your video. The moment that stops the scroll. Creators with consistently high hook rates compared to industry benchmarks are the ones driving revenue.

So, what makes a good hook? A quick jolt of surprise. An unexpected contrast. A sharp point of view that earns attention before the thumb moves on.

It sounds subjective until you start tracking it. Then it becomes your clearest performance signal.

Billo tracks half a billion dollars in ad purchase data from brands running creator content on Meta and TikTok. They use that to rank creators by hook rate performance. 

So instead of guessing which creators might work, you're choosing based on who's already proven they can stop the scroll.

But launching content is just step one. The real leverage comes from what you do after.

You're running organic posts, Spark ads on TikTok, partnership ads on Meta. Track how everything performs. Then categorize your results:

Winner: This creator's content performs. Iterate on it. Make more with similar hooks and angles.

Almost winner: The concept works but the hook doesn't. Rewrite the script or test a different opening.

Loser: Replace and move on.

You're not trying to predict winners 100% of the time. You're building a feedback loop that tells you the next move. Being right more than 50% is enough if you iterate quickly.

The process compounds: 

This moves you from random creator experiments to a compounding content engine. The framework works whether you're using a platform, building your own system, or working with an agency. What matters is that you have one.

AI avatars are everywhere right now. Spin up 100 videos for five bucks each, no real creators required. Fast, cheap, scalable…and a disaster waiting to happen.

"AI UGC works today because people believe the content is real," Donatas said. "A grandma buying supplements thinks the grandpa in the ad is real. But what happens when people can't tell what's real anymore? They stop trusting everything."

That’s the risk. 

Once audiences lose the ability to spot authenticity, every piece of content (real or not) loses power. The short-term win becomes a long-term trust crisis.

Creators pay for it first. When they license or clone their faces, those avatars can end up anywhere: adult sites, scam products, or shady brands they’d never touch.

Then brands pay for it too. You’re borrowing a face with no history, attaching it to your reputation, and hoping customers never find out it’s fake. When they do, the backlash spreads faster than the content ever did.

Billo’s stance is clear: AI isn’t the villain, it just needs boundaries. 

They use it to write smarter briefs, analyze performance, and edit faster. The tech supports their system; it doesn’t replace the humans inside it.

Tech will keep moving faster. The scroll will too. But the moment customers stop believing what they see, the game’s over.

Keep your creators human, and your content worth trusting.