What Should You Cut From a Homepage That Got Too Long?

What should you cut from a homepage that got too long?

Cut every section that exists because somebody internal asked for it rather than because a visitor needs it. Then cut every section that repeats a point already made above it. What survives is usually half the page, and it converts better because the argument finally has a shape.

Homepages do not get long on purpose. They accrete. A new feature launches and gets a band. Sales asks for a logo strip. Someone reads about social proof and adds testimonials in three different places. Two years later the page is nine screens deep and nobody remembers why.

Here is how I go about shortening one, and the specific things I look for first.

Why does a long homepage actually hurt?

Three ways. It buries the one thing you most need a stranger to understand. It slows the page down, which Google measures and buyers feel. And it lets you avoid the hard work of deciding what you are, because a page that says nine things never has to commit to one.

The burying problem is the worst of the three. Attention on a homepage is front-loaded. Whatever is in the first screen carries most of the weight, and every section you add below dilutes the chance that a visitor reaches the part that would have convinced them.

The avoidance problem is the one nobody admits to. Adding a section is easy and feels productive. Deciding that your product is primarily for one buyer, and that the other three audiences get a different page, is hard and creates arguments. Long homepages are frequently a symptom of an unresolved positioning question.

What does page length cost in performance terms?

Real, measurable amounts. Google's Core Web Vitals set thresholds you can check against: Largest Contentful Paint should occur within 2.5 seconds, Interaction to Next Paint should be 200 milliseconds or less, and Cumulative Layout Shift should stay at 0.1 or less.

Those thresholds are assessed at the 75th percentile of page loads, segmented across mobile and desktop devices. That percentile detail matters more than the numbers. You are not being judged on your average visitor. You are being judged near the slow end, which is where the person on a phone on a train lives.

Every extra section is more images, more fonts, more scripts, and more layout to compute. A long homepage is not just a reading problem, it is a loading problem, and the two compound. The slow page is also the one asking for more patience.

You can check your own numbers in PageSpeed Insights, in the Core Web Vitals report inside Search Console, or straight from the Chrome User Experience Report if you want the raw field data. Lab tools are useful for diagnosing, but the field data is what counts, because it reflects real devices on real connections rather than a simulated run on a fast machine.

This is the argument that usually wins internally, by the way. Nobody wants to remove their section because a consultant finds it wordy. People will remove a section when you can show it costs measurable speed at the percentile Google actually grades.

Which sections go first?

The ones that serve an internal audience. Award badges nobody has heard of, a mission statement written for a board deck, a founder's letter, a timeline of company milestones, and any section that exists because a department wanted visibility. None of these answer a question a buyer is asking.

Next go the duplicates. Most long homepages make the same point three times in different formats: once as a headline, once as a feature card, once as a testimonial. Keep the strongest version and delete the other two. Repetition feels like emphasis to the person who wrote it and feels like padding to the person reading it.

Then go the sections that belong on another page. Detailed feature specifications, integration lists, pricing tables, and security details are all legitimate content that mostly does not belong on a homepage. Link to them. A homepage's job is to make someone want the next page, not to be every page. If the sections are built as reusable components in Webflow, or as components in your Figma file, moving them is a drag rather than a rebuild, which removes most of the objection to moving them at all.

Trust elements are the exception people get wrong in both directions. Some proof belongs high up, and I have argued elsewhere about which trust badges actually build trust rather than just filling space. The test is whether a specific badge would change a specific doubt.

How do you decide what survives?

Give every remaining section one job, written as a sentence, and check that no two sections have the same job. If a section cannot be described in one sentence, it is doing too much. If two sections have the same sentence, one of them goes.

I write these as the visitor's question rather than our message. Not "communicate our security posture" but "is this safe enough for my data". Framing it as their question immediately exposes the sections that exist to satisfy us, because you cannot write a plausible visitor question for them.

Then order the surviving sections by when the question occurs. Nobody asks about integrations before they understand what the product does. Nobody asks about pricing before they believe it might work. Getting the order right frequently fixes a page that felt long, because a page in the wrong order feels longer than it is.

This ordering problem is close to what I described in how much a page should explain before it asks. Length and sequence are the same problem viewed from two angles.

What if different buyers need different things?

Then you probably need more pages, not a longer one. A homepage trying to serve three distinct buyers ends up serving none of them well, because every paragraph is a compromise. Route early instead: say who you are for, and give each audience its own path.

The routing pattern is simple and underused. A short section near the top that names the audiences and links onward beats four parallel sets of feature blocks. It also gives you clean pages you can point campaigns at, which the homepage never does well.

The version of this that does work on one page is a clear primary audience with secondary ones acknowledged briefly. That is a different thing from treating all three equally, and I went through the trade-offs in designing one page for two different buyers.

How do you handle the political part?

By making the criteria public before you touch anything. Agree the rule first: every section states a visitor question, no two sections answer the same question, and anything that fails moves to another page rather than disappearing. Then apply the rule mechanically.

The word that saves you is moved rather than deleted. Almost nobody fights to keep a section on the homepage when the alternative is a dedicated page where their content gets more room and better attention. The fight is about erasure, not location, and you can usually remove the fight entirely.

Write the decisions down as you go, with the reason next to each one. Six weeks later someone will ask why the awards section went, and having an answer that is not a matter of taste is the difference between a decision that holds and one that quietly gets reversed.

How do you know the shorter version is better?

Measure the things a shorter page should improve and be willing to be wrong. Scroll depth to your primary call to action. Click-through to the pages you routed people to. Your Core Web Vitals field numbers. And the conversion rate, given enough traffic to mean anything at all.

Scroll depth and click maps come from a session tool such as Microsoft Clarity or Hotjar, and the destination traffic from Google Analytics or whatever you use. None of these individually prove anything. Together they tell you whether people are getting further and doing more, which is the actual question.

Give it a full cycle before judging, and watch the routed pages as well as the homepage. A successful homepage cut often moves conversions rather than creating them, and if you only look at the homepage you will conclude you made things worse.

Be honest if it does not improve. Shortening is usually right and it is not a law. A page can be cut too far, particularly for a complex or expensive product where the buyer genuinely needs more before they will act. What I have not seen is a page that was too short because somebody removed the awards section.

What should you do next?

Open your homepage and write the visitor question each section answers, in one sentence, without being generous to yourself. Sections you cannot write a question for are your first cut. Duplicated questions are your second. That exercise takes an hour and usually settles the argument.

Then check your field performance data before and after, because the speed improvement gives you the one piece of evidence nobody can argue with. A shorter page that loads faster at the 75th percentile is a better page by a standard you did not invent.

If you want an outside read on which sections are earning their place, reach out. It is easier for someone with no history with the page to say what everyone internally already suspects. Let's chat.