Nobody opens Instagram to find your product. They were scrolling, something stopped their thumb, and they tapped out of curiosity rather than intent. That is the whole difference between paid social and paid search, and it changes what the page has to do. A page that works for a searcher who typed your category will usually fail a stranger who tapped a video. Here is what the page needs instead.
Meta traffic is interrupted traffic. The person was mid-feed, half-attentive, almost always on a phone, and their decision to tap was made in under a breath on the strength of a hook. They do not know your brand, they did not ask for your category, and they have no problem in mind that they are actively trying to solve right now. The ad created a flicker of interest; the page has to turn that flicker into a reason to stay.
That changes the page's opening job. On search, the page confirms. On social, the page has to re-earn attention, explain the problem it solves to someone who may not have named it yet, and do all of that before the feed pulls them back. The back-swipe is faster than any click, and the feed is right there waiting.
It also changes what counts as proof. A stranger has no reason to believe you, and claims that would be fine for a warm visitor read as marketing to a cold one. Everything on the page needs to answer the unspoken question: "Who are you, and why should I give you another ten seconds?"
The ad and the page have opposite jobs: the creative interrupts, the page reassures. Here is how to keep them connected without making either do the other's job.
Whatever line stopped the scroll, whether a question, a bold claim or a specific outcome, belongs in the page headline in recognisable form. The visitor is checking that they landed where the ad implied. If the ad asked "Still paying for software you don't use?" and the page says "Enterprise spend management," the thread is cut.
If the ad was a person demonstrating something, that person or that product should appear in the hero. If the creative had a distinctive colour or style, carry it. Visual continuity is what tells a half-attentive thumb that this is the same thing they just saw, not an unrelated website.
Do not try to make the page as loud as the creative. A page full of hooks reads as noise. The ad earned a tap; the page's tone should drop half a step to calm, specific and clear. The hero restates the promise; the next section explains it in plain terms; then proof; then the ask.
If you are testing three angles in ad sets, the pages should differ in the headline and first section to match. A problem-led angle, a result-led angle and a curiosity-led angle land on visitors in three different mental states. One generic page serves none of them well and makes every ad test harder to read.
Most Meta traffic arrives through the in-app browser on a phone. Design the page in that environment first and let desktop inherit from it.
Open your own ad on your own phone and tap through. The in-app browser has its own quirks: a different viewport height, a bottom bar that eats space, and occasionally slower script execution. A page that looks fine in a desktop browser narrowed to phone width can still lose the headline below the fold here.
Before any scroll, the visitor should see the headline that matches the ad and either the primary button or the first form field. No logo-only header, no full-bleed hero image pushing the words down, no cookie wall covering both. If the page needs a consent banner, keep it to a strip, not an overlay.
Cold visitors have a short patience budget and are often on cellular. Compress images, drop autoplay video unless it is the proof itself, defer every script that is not needed for the first paint, and avoid font files that block rendering. The feed loaded instantly; the page is being compared to that.
Large tap targets, one column, the right keyboard for each field, and as few fields as the next step genuinely needs. If a visitor has to zoom to tap a checkbox, you have lost them. Put the button directly under the last field, not after a paragraph of legal text.
A navigation bar is an exit menu on a phone. Remove or minimise it. The only prominent links should be the call to action and, if needed, an anchor to the proof or the FAQ.
Warm visitors already trust you a little; cold visitors do not. The proof on a paid social page has to work for someone who has never heard your name. That rules out most of what pages usually offer: a row of anonymous five-star icons, a quote with no name, a claim with no context.
What works is specific and human. A short testimonial with a real first name, a photo, and a concrete before-and-after is worth more than a wall of logos the visitor does not recognise. A screenshot of a real result, a short video of a real customer, or a plain description of exactly what the product did for one person all read as true in a way that adjectives do not. Place the strongest piece directly next to the call to action, where the doubt is highest, and not in a carousel further down.
If you do not have many customers yet, say what you can truthfully say: how many people have used it, what the first users reported, what you do if it does not fit. Honest, modest proof beats inflated proof with a cold audience, because the cold audience is already expecting to be sold to.
If you are not sure whether your page's proof and first screen would survive a stranger's ten seconds, a landing page audit scores both and shows you what to fix first.
Retargeting traffic is a different visitor again. They have seen the page once, did not act, and are now seeing a follow-up ad. Sending them back to the same cold-traffic page repeats what did not work. A retargeting page should assume familiarity and skip the introduction.
Lead with the objection that most likely stopped them. For most products that is one of three things: price, fit, or trust. Build the page around answering that directly: a plain pricing explanation, a clear statement of who it is and is not for, or a more detailed proof section. Shorten the path to the action, because they have already read the pitch, and consider a lower-commitment ask than the original if the original was a purchase.
The practical pattern is one page per cold angle plus one for retargeting, with shared structure. The retargeting page gets the deepest proof, the shortest form, and the most direct answer to "why not."
A Google Ads visitor searched for something and is checking that the page answers it, so the page's job is to confirm and deliver. A Facebook or Instagram visitor was interrupted mid-scroll, tapped on curiosity, and has no problem in mind yet. The page has to re-earn attention, explain the problem in plain terms, and prove itself to a stranger, all on a phone inside an in-app browser. That means a calmer tone than the ad, continuity with the creative, lighter pages, and proof with names and faces.
Yes, in the headline, the visual and the offer. The visitor is checking that they landed where the ad implied, and a different hook, a different look or a different promise reads as a wrong turn. The page should not try to be as loud as the ad, though. The creative interrupts; the page explains. Restate the hook in the headline, show the same person or product in the hero, and then move to specific, calm copy with proof next to the call to action.
Proof that is specific and human. A short testimonial with a real first name, a photo and a concrete before-and-after; a screenshot of a real result; a short video of a real customer. Anonymous star ratings and unattributed quotes do not work on people who have never heard of you. Place the strongest piece directly next to the call to action. If you have few customers, state what you can truthfully say rather than inflating it; cold audiences expect to be sold to and reward modesty.
In most cases, yes. Retargeting visitors have already seen your page and did not act, so sending them back to the same page repeats what failed. A retargeting page should skip the introduction, lead with the objection most likely to have stopped them, such as price, fit or trust, offer deeper proof, shorten the form, and sometimes propose a lower-commitment step than the original ask. Share the structure with your cold-traffic pages and vary only the opening and the depth of proof.
Run your page through the same diagnostic. The free preview takes about a minute — no signup.
Audit my page