Landing page builders a media buyer would pick

Landing page builders get reviewed as if the tool decides the result. It does not. The offer decides most of it, the load time decides a good chunk of what is left, and the builder decides how much of your week you spend fighting it.

I have run paid traffic to pages built on most of these. The differences that showed up in cost per acquisition were rarely the ones the feature comparisons talk about.

Which landing page builders should a media buyer use?

Judge landing page builders on three things only: how fast a page renders on a mid range Android phone, whether real purchase events can reach Meta and Google from your server, and how much of the builder’s own JavaScript ships to the visitor. A tool that scores well on all three is enough. Template count is irrelevant.

Those three questions map directly onto money. Speed decides how many arrivals become sessions. Server side events decide how much of the conversion data the bidding algorithms actually see. Builder weight is just speed again, from the other direction.

Everything else, the drag handles, the AI copy assistants, the popup libraries, changes how pleasant the afternoon is. It does not change the number on the invoice.

How much does page speed change your cost per acquisition?

Page speed changes cost per acquisition twice, and most people only count the first effect. Slower pages lose visitors before anything renders, so fewer arrivals become conversions. They also fire fewer events, which starves automated bidding of the data it optimises against. The second effect compounds over weeks.

Google publishes a library of documented cases with real business outcomes. Vodafone Italy improved largest contentful paint by 31 percent and reported 8 percent more sales. Cdiscount reported a 6 percent revenue uplift. Lazada reported a 16.9 percent increase in mobile conversions. Those are in the Core Web Vitals business impact collection, with each company named.

Note what those numbers are and are not. They are self reported by the companies involved, they cover different metrics, and none is a controlled experiment. They still point the same direction, consistently, across dozens of cases, which is more than most claims in this category can say.

The thresholds worth holding in your head are largest contentful paint under 2.5 seconds and interaction to next paint under 200 milliseconds, measured on a mid range Android device. That is the phone your ad traffic actually arrives on. It is not the machine the page was designed on, which is why so many pages pass in the office and fail in the field.

Which builders break your conversion tracking?

The ones that will not let you edit the head, add a custom endpoint, or point events at your own domain. Without those three permissions you are stuck with browser side conversion tracking, which loses events on Safari and inside iOS apps. That loss is invisible in the builder’s dashboard and very visible in your cost per result.

The fix is sending purchase and lead events from a server rather than only from the browser. Meta documents this in the Conversions API, including how to deduplicate against the browser pixel so you do not double count. Google has the equivalent through enhanced conversions and its own server side container.

Three practical checks before you commit to a tool:

1. Can you add arbitrary script to the head and body? If not, most measurement setups are off the table.
2. Can the form post to your own endpoint? You need the lead to hit your server, not only the builder’s database.
3. Can you serve tags from a subdomain you control? This is what keeps first party cookies alive for longer than seven days on Safari.

One more thing worth saying about landing page builders and consent. If you sell to European customers, the tool has to let you gate tags behind a consent choice. Several of these do not, and bolting a banner on top of a closed builder usually produces something that blocks nothing.

A builder that fails all three is not unusable, it is just permanently blind. That is the tradeoff nobody puts in the comparison table, and it is the part covered properly in advertisement tracking.

The ten, and what is wrong with each one

Ten tools, with the weakness stated rather than buried. Speed ratings are my own experience across client accounts, not a benchmark, and every one of them can be made slow by a careless build. The server side column is about whether the tool lets you do it at all, not whether it is easy.

Tool	Speed out of the box	Server side events	The real weakness
Framer	Fast	Via your own container	Young ecosystem, you host on their terms
Carrd	Fastest	Barely	No testing, no real plumbing, will not scale
Webflow	Fast if disciplined	Via your own container	Embeds and interactions add weight quickly
WordPress, block theme	Depends entirely on you	Full control	Only as fast as whoever maintains it
Swipe Pages	Fast	Limited	Small vendor, fewer integrations
Leadpages	Medium	Limited	Dated output, weak custom code support
Shopify	Medium	Native to Meta and Google	Theme and app bloat compounds fast
Instapage	Medium	Partial	Enterprise pricing for features most accounts never use
Unbounce	Slow	Partial	Heavy builder runtime, positioned layouts hurt mobile
ClickFunnels	Slow	Locked down	Closed ecosystem, poor speed scores, hard to leave

Two rows deserve a comment. WordPress is genuinely the best answer for control and the worst answer for neglect, and the deciding factor is whether a person owns it. Unbounce has the strongest template and testing story of the group, which is exactly why people forgive its page weight for longer than they should.

Price is the part most comparisons of landing page builders get wrong as well. The monthly fee is small next to the cost of the traffic you send through it. A tool at $99 a month that quietly costs you a tenth of your conversions is more expensive than one at $299 that does not. Work that difference out in real money at your own spend level before choosing on price.

Why do landing page builders matter less than the offer?

Because a fast page selling the wrong thing still loses. Across the accounts I run, changing the offer moves cost per acquisition by a multiple. Changing the builder moves it by a fraction, mostly through load time. Ranking landing page builders is therefore the fourth most useful thing you can do this month, not the first.

The order that actually works is boring. Fix the offer, then the proof, then the load time, then the tool. Most teams do it backwards because choosing software feels like progress and rewriting an offer feels like admitting the last one was weak.

Here is the awkward version. If your page converts at 1 percent and your competitor’s converts at 3 percent, the gap is almost never the builder. It is price, guarantee, delivery time, payment options, or the fact that they answer an objection you have not written down. No amount of A/B testing on button colour reaches that.

I would rather spend a week on the offer and ship the page on Carrd than spend a week comparing landing page builders. The traffic side of this is the same argument made in media buying: the buying is not where most of the money is won.

What we do about it

We do not sell page builds, so we have no reason to push you toward any of these landing page builders. What we do is refuse to run traffic to a page that takes six seconds on a phone, because it wastes your budget and makes our numbers look worse than they are.

In practice the first two weeks of an account are usually load time, server side events, and one uncomfortable conversation about the offer. Only after that does creative testing tell you anything trustworthy.

You can see what we charge before speaking to anyone, including the exact percentage at your spend level. If you want fifteen minutes to work out whether your page or your offer is the problem, book it and bring the URL.