The Homepage Is Not Your Slowest Page

Why the home page misleads

It is the page whoever built the site spent most time on.

It is the page anybody checking the site opens first, so problems on it get noticed and fixed.

And on most small business sites it receives a minority of arrivals, because people land on service pages from searches.

So testing it produces a figure that is both better than typical and describes a page fewer people see than you think.

Where the slow pages actually are

• Galleries and photograph pages.
• Long service pages with many images.
• Pages with an embedded map.
• Anything with a form from a third party.
• Pages built outside the template.
• The blog index loading every image.

The fifth is the reliable one. A page built for a campaign, or by a different person, or from a page builder rather than the site's template, frequently loads a set of resources nothing else on the site uses.

Pick the sample by arrivals

Rather than by which pages feel important.

Open your analytics and list the ten pages people most commonly arrive on, which is a different list from the ten most visited.

Those are the pages forming first impressions, and their speed is the speed most visitors experience.

Test those ten rather than the home page, and note the figures individually.

Frequently the home page is fourth or fifth on that list rather than first, which is itself worth knowing.

That is fifteen minutes and it replaces a single misleading number with a picture.

A worked example

A business tested their home page regularly and it loaded in about two seconds.

Their landing page report showed that the home page accounted for under a fifth of arrivals.

The most common landing page was a service page with a gallery of nineteen photographs, which took nine seconds.

The second was a page built for an advertising campaign the previous year, which loaded a separate set of fonts and a script nothing else used.

Fixing those two took an afternoon and improved the experience for roughly half of all arrivals.

The home page, which had been tested monthly for two years, needed nothing.

Template problems and page problems

Worth distinguishing, since they need different responses.

Where every page is slow by a similar amount, the cause is in the template or in something loading everywhere, and fixing it once fixes all of them.

Where one page is much worse than the rest, the cause is on that page: its images, an embedded element, or something added to it specifically.

Testing several pages is what lets you tell the difference, which a single test cannot.

Deal with the template problem first, since it applies to everything, then the individual pages in order of arrivals.

Weight by what the page is for

Since not all arrivals are equal.

A slow page that produces enquiries costs more than a slow page nobody acts on.

Order the list by arrivals multiplied by commercial importance rather than by arrivals alone.

That usually promotes one or two service pages above a popular blog post, which is the right priority.

And it prevents the common outcome where a great deal of effort goes into a page with traffic and no purpose.

The pages behind a form are worth a look

An easily missed category, since they do not appear in landing page reports at all.

A booking flow, a basket, or a multi-step enquiry has pages nobody arrives on directly and everybody who converts passes through.

Those carry the most third-party code on most sites, because payment and booking tools load their own scripts, and they are the pages where a delay costs an actual sale rather than a visit.

Walk through your own booking or checkout on a phone once, timing it, which is the only way to see them.

A slow home page costs you an impression and a slow checkout costs you the order.

Test after anything is added

Because a single page can be spoiled without touching the rest of the site.

Adding a map, a form, a video, or a booking widget to one page affects that page only, and nobody retests it.

Make a quick check part of adding anything: test the page before and after, and note the difference.

That takes two minutes and catches the case where a small addition costs several seconds.

Where the cost is large, that is a decision to make rather than a surprise to discover a year later.

Check the pages nobody looks at

Since they are the ones most likely to have drifted.

An old campaign page, a page for a service you rarely sell, or something built years ago outside the current template.

Those receive no attention and are exactly where an unused font or an abandoned script survives.

They may also still be receiving arrivals from search, which is the case worth finding.

Test three of them once a year alongside the main list, which is enough to find the outliers.

Speed matters most where the intent is lowest

Worth understanding, since it changes which slow page to worry about.

Somebody who typed your name and reached the home page will wait, because they came looking for you specifically.

Somebody who tapped a search result for a general question has three other results open and no attachment to any of them, so a slow page loses them at almost no cost to themselves.

Which means the pages arriving from search are both the most numerous and the least forgiving, and they are exactly the pages nobody tests.

That is the argument for the landing page list over the home page, put in terms of what it costs rather than what it measures.

The counter-case

The home page still matters.

It is the page people reach by typing your name, which is a high-intent visitor, and it is what somebody sees after hearing about you.

Testing it regularly is not wrong, it is just insufficient on its own.

And on some sites, particularly where most traffic is direct or from advertising to the front page, it genuinely is the representative page.

Test the ten pages people actually arrive on, tell template problems from page problems, and retest whenever something is added.

The fifteen minutes

1. Open the landing page report.
2. List the top ten arrivals.
3. Test each with the same settings.
4. Note whether all are slow or just one.
5. Fix the template problem first.
6. Order the rest by importance.
7. Retest anything you add to.

Step one is where the surprise usually is, since the home page is frequently well down the list of pages people actually arrive on.

Finding the single worst offender is covered in the slowest page on your site.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.