Landing pages for native advertising: what changes with Taboola and Outbrain

You finish reading a piece on a news site, and below the last paragraph a grid of thumbnails appears: "These 5 mistakes are costing homeowners a fortune," "The new scheme retirees don't know about." Those blocks are sold by content recommendation platforms — Taboola, Outbrain and a handful of others — which buy inventory from publishers and resell it to advertisers on a cost-per-click basis. The entry price is very low compared with search, and that's what draws people in. The trouble starts immediately after: the landing page receiving that click has almost nothing in common with the one that works behind a Google Ads listing.

A click nobody asked for

It all comes down to intent. On Google Ads, the visitor typed a query: they articulated a need, and your ad answers it. On Meta Ads, they're not looking for anything but they're in discovery mode, flagged by targeting as plausibly interested. On a recommendation widget, intent is weaker still: the visitor came to read an article, they've finished it, their attention is drifting, and they tap a thumbnail because the headline sparked half a second of curiosity. They often don't realise they're leaving the site they were on — or that they've just clicked an ad.

That last point isn't a practitioner's hunch. The study by Bartosz W. Wojdynski and Nathaniel J. Evans, "Going Native," published in the Journal of Advertising in 2016, tested different wordings and positions for the sponsorship disclosure: whether readers recognise content as advertising depends heavily on where the disclosure sits and which words it uses, with "advertising" or "sponsored" working better than softer phrasings such as "brand voice." A study by Michelle A. Amazeen and Bartosz W. Wojdynski, published in Journalism in 2020 using a representative sample of 800 US adults, goes further: fewer than one participant in ten identified the sponsored content as advertising, even though the disclosure was present.

Why a standard promotional landing page falls flat here

A conventional campaign landing page speaks to a visitor who knows where they're going: full-screen hero, bold commercial promise, "Request a quote" button above the fold, benefits, testimonials, form. That format assumes an implicit contract has already been accepted — "I'm on a commercial page, this is what I expected." A visitor arriving from a widget never signed that contract: they thought they were opening an article. The abrupt shift from an editorial context to a sales context triggers an almost instant back-button reflex, before any assessment of the offer. It's banner blindness, applied to the destination page.

• Tonal break: you go from journalistic writing to brand copy in a fraction of a second. The message match discipline is compounded here by a need for consistency of register.
• Commitment demanded too early: a form visible on load asks for effort from someone who had never heard of you thirty seconds ago.
• Degraded mobile context: native traffic is overwhelmingly mobile, at the tail end of a reading session. Every second of load time is paid for in abandonment — see our article on mobile-first design.
• Headline promise not kept: thumbnails run on curiosity. If the page doesn't immediately pick up the advertised topic, the visitor feels baited and leaves.

The advertorial: a page that continues the reading

The channel's standard answer is the advertorial: a hybrid page, somewhere between an article and a landing page. It opens like editorial content — a headline close to the thumbnail's, a standfirst, running text, subheadings — and gradually introduces the product as an answer to the problem it has laid out. The first call to action only appears after several screens. This isn't trickery: the visitor came to read, and needs something to read before being asked for anything. That's why the long page versus short page debate almost always resolves the same way here.

1. Pick up the headline: the first words must echo the exact angle of the thumbnail that was clicked, not the brand's commercial promise.
2. Set the context: two to four paragraphs that genuinely address the announced topic, useful even to someone who will never buy.
3. Introduce the problem: the passage where the editorial subject meets what your offer solves.
4. Present the solution: the product, described plainly, with its supporting evidence.
5. First call to action, then one or two spaced-out reminders further down the page.
6. Reassurance: terms, shipping, refunds, company identity.

Transparency: the "sponsored content" label

The advertorial walks a narrow line: it works precisely because it resembles an article, and that resemblance is exactly what creates an ethical and legal problem. The second major study on the topic, "The Deceptiveness of Sponsored News Articles" by Bartosz W. Wojdynski, published in American Behavioral Scientist in 2016, shows a two-stage effect: the presence of a logo and the prominence of the disclosure increase recognition of the content as advertising, and that recognition in turn lowers perceived article quality and attitude towards the sponsor, through activated persuasion knowledge and a sense of having been deceived. That last variable is the key one: it isn't understanding that something is an ad that penalises the brand, it's finding out late. In the EU and the UK, hiding commercial intent also falls under misleading commercial practices rules. A clear "Sponsored content by [brand]" at the top of the page is therefore a legal requirement as much as the cheaper position; the alternative belongs to the family of dark patterns.

Traffic quality and real profitability

The channel's second pitfall is the quality of the traffic you're buying, where three phenomena stack up. Accidental clicks: on mobile, widgets sit inside the scroll path, and a share of clicks is unintentional — those visitors leave within seconds. Non-human traffic: like any inventory bought per click at scale, native attracts fraud patterns, even though platforms filter part of it out. And the wildly uneven quality of the network: alongside recognisable news brands, the inventory includes sites built for the sole purpose of hosting ads. The practical consequence: a native campaign is managed at source-site level, not at campaign level.

• Tag URLs with UTM parameters that include the source site identifier, so each conversion maps to a specific publisher.
• Build aggressive exclusion lists from the first days: a site sending clicks and no conversions should be cut quickly.
• Watch time on page and scroll depth rather than bounce rate alone, which can't tell an accidental tap from an attentive read.
• Protect the form against automated submissions and disposable email addresses, or your lead volume will be misleading.
• Set a minimum click threshold per site before making any decision, so you don't cut a decent publisher on statistical noise.

Which leaves the central question: is the channel profitable? A low cost per click is a classic reading trap. A CPC far below search rates is mechanically offset by a conversion rate that is also far lower, and the only figure that counts is cost per acquisition — or margin per acquisition. The method is the one in our guide to calculating a landing page's ROI, with two precautions specific to native: widen the attribution window, because cold traffic converts later, and track lead quality through to closed business. A channel producing three times as many leads at a third of the unit cost is still less profitable if those leads close ten times less often.

Native, Meta, Google: which channel for what

Native makes sense when your offer targets a broad audience and needs explaining before it can be sold: comfort products, home equipment, consumer financial services, courses, subscriptions. It also makes sense when search volume in your category is too thin to feed a search campaign. Conversely, it's rarely the right pick for long-cycle B2B, a local and urgent service, or a low average order value. Compared with a Meta Ads campaign, it offers a longer reading context but coarser targeting: you're buying slow attention, not precision. Either way, treat a native click for what it is — attention borrowed from an article, not purchase intent. If your offer is a physical product that needs explaining before it sells, LanderKit's ecommerce-produit template provides that structure: a long page, context sections ahead of the purchase, with proof and reassurance built in. The 10 LanderKit templates are available at €89 each, or €229 for the full pack.