3 client slots open —     
hello@native-advertising.net
native-advertising.net

7 min readBy Marcel Sattler

Behind the Scenes: Our Native Ad Creative Process for 4 Networks

The exact internal framework we use to build native ads for Taboola, Outbrain, RevContent, and MGID — audience research, 3-5 angles, advertorials, and data-driven testing.

From the post

They failed before they ever launched, because they skipped the creative process that separates a profitable Taboola campaign from a money pit.

— Marcel Sattler

↓ read on

Most advertisers who fail on native didn't fail at media buying. They failed before they ever launched, because they skipped the creative process that separates a profitable Taboola campaign from a money pit. The teams that crush it on YouTube and Meta walk into Taboola, Outbrain, RevContent, and MGID and get destroyed — because native has no smart algorithm to bail you out.

This is the actual production line we run inside the agency, end to end, before a single dollar hits a platform. Same framework for ecommerce, lead-gen, and affiliate. The order matters, and so does the volume of variations — we test 3 to 5 angles per offer and 3 to 5 advertorials per angle, every time.

I'm Marcel Sattler, founder of native-advertising.net, and since 2015 I've deployed more than $100M across Taboola, Outbrain, Newsbreak, MGID, Yahoo Native, Mediago, and RevContent — almost all of it spent learning that the creative comes before the click, not after. The whole reason this framework exists is to make results replicable, duplicatable, and predictable. None of it is rocket science. It's just homework most people refuse to do.

Why native ad creative is different from Meta and YouTube

On Meta and YouTube, the algorithm finds your buyer. You feed it a few creatives, the machine learns, and it pushes spend toward the people most likely to convert. That safety net does not exist on Taboola, Outbrain, RevContent, or MGID.

Native is interruption traffic. Someone is reading an article and your ad sits at the bottom of the page. They are not in a buying mindset. They are not searching for your product. The prospect should not even feel advertised to — the moment they sense a hard sell, they bounce.

That changes everything about how we build. The creative has to do the targeting work the algorithm does elsewhere. So before we touch a campaign, we spend the bulk of our effort on understanding the prospect and building a cold-audience-friendly path from the ad to the offer. If you're running ecommerce or affiliate campaigns and wondering why your Meta playbook collapsed on native, this is why.

Step 1: Audience research before a single ad

The foundation of every native ad we run is the audience. Not the offer — we assume you already have a proven, converting offer. The variable we control is how cold the audience can be and still buy.

We deliberately target completely cold traffic, not people already in the decision-making stage. Cold scales easier and cheaper on native, which is the entire point of the channel. To reach those people, our copywriting team reads the whole web on the prospect's behalf:

  • Amazon reviews of books in the niche — in a weight-loss play, we read reviews of diet and fitness books to find the exact pain in the customer's words
  • Reddit threads where the audience complains, asks questions, and uses its own slang
  • Blogs and YouTube videos in the vertical to capture tone of voice
  • Forums and comment sections for the raw, unfiltered language

The goal is two things: the real problem the audience is itching about, and the exact words they use to describe it. Our copywriters run this, not a media buyer, because the tone of voice is the whole game on native. This is the part lead-gen advertisers skip most often, and it's why their advertorials read like ad copy instead of editorial.

Step 2: One concept board, multiple angles

Once the research is in, we dump everything onto a single board — a Miro-style canvas with the audience notes, the reviews, the tone of voice, all in one view. From that mess we pull the distinct angles.

When you dig into an audience properly, you almost always find several viable ways into the sale. The same product can be sold on three completely different promises, and you cannot guess which one wins from the armchair.

The diet-pill case: same product, three angles

We had a weight-loss client essentially selling diet pills. The lazy approach is "take these pills, lose weight." We didn't run that. The audience was male, so we built three different angles around the same product:

  1. Look better — the straightforward appearance angle
  2. Attract more girls and get more dates — the social/status angle
  3. More success in the gym — the performance angle

We didn't know which would win, so we tested all three. The winner, by a wide margin, was the dating angle — the internal headline ran along the lines of "be attractive to more girls and get the girls you want." Simple, almost crude, and it beat the obvious weight-loss pitch. That outcome is invisible until you test it, which is exactly why we build 3 to 5 angles per offer instead of betting on one.

Step 3: Advertorials, not direct-to-offer

This is the piece most newcomers miss. On native, you do not send the click straight from the ad to the offer. You put an advertorial in between.

An advertorial is a landing page that reads like editorial content — a news magazine piece or a third-party article — with legit copywriting about the product underneath. It bridges the gap between a cold reader and a sales page, warming them up so the offer doesn't feel like a cold pitch.

We write 3 to 5 advertorials per angle, on top of the 3 to 5 angles, so a single offer can easily produce 15 or more distinct landing pages. And we diversify the format on purpose:

  • Newspaper-style editorials that mimic a magazine article
  • Blog-style articles written as if a real audience member or micro-influencer wrote them
  • Storytelling pieces that give an "under the hood" personal account

If every advertorial uses the same format, you've only tested one idea wearing different hats. Diversity at this layer is where the breakthroughs come from on Taboola and Outbrain alike.

Step 4: Build the landing pages on a proven template

We build the advertorials in Webflow because it lets us test and iterate fast, and we work from templates that have already absorbed real native traffic. This is not the place to improvise.

Two warnings I give every advertiser. First, do not build your first advertorial from scratch — you have no reference for what a converting native landing page actually looks like, and you'll burn budget proving it the hard way. Second, do not go to Fiverr for a cheap, quick-and-dirty advertorial. The danger isn't the price; it's that the page has never seen native traffic and almost certainly won't convert.

Use a template that has already received tons of native advertising spend behind it. A proven layout is the difference between a Taboola or MGID campaign that has a fighting chance and one that's dead on arrival. The CMS doesn't matter much — we keep templates for Webflow, WordPress, and ClickFunnels — but the proven part is non-negotiable.

Step 5: Assemble the ad and diversify the creative

Only now do we build the ad itself. On Taboola, Outbrain, RevContent, or MGID, an ad is a creative — an image or a video — plus a headline, and in some cases a description. That's the whole unit.

Our campaign structure follows the angles: roughly one campaign per angle, then split by device so we can test mobile, desktop, and tablet traffic separately. Inside each, we push the variations hard.

On headlines, we test with numbers and without — a "review"-style headline against a plain one. On creatives, we deliberately go in different directions:

  • Merging a person's head shot with a blurred background
  • A hand holding a product or an ID
  • AI-generated images
  • Vector graphics

We go wide on purpose. The point is to plant a lot of seeds, then let the data tell us where the biggest tree grows. Once a direction shows signal, we pour more into it and cut the rest. That's the entire logic of the framework: launch broad, read the data, double down. The same discipline runs through our case studies across every network.

Watch the full breakdown

Is your account a fit for the same play?

If you have a proven offer and you're either stalled on native or haven't launched yet, the gap is almost always upstream — in the audience research, the angles, and the advertorials, not the bidding. Get those right and the platform side gets a lot simpler. We run this exact framework on Taboola, Outbrain, MGID, and RevContent for ecommerce, lead-gen, and affiliate clients.

If you want us to look at your offer and map the angles ourselves, book a strategy call and bring your numbers. You can also work through our free video and post library to see how the pieces fit before we talk.

▸ Keep reading

Three more on the same topic.