Technote

Performance optimization Practical

Series Speed and conversion Part 4 of 8

This is how a landing page gets heavy

Nobody decides to make a landing page heavy. It accumulates: one addition per campaign, and nothing ever removed when the campaign ends.

Open a slow landing page and nothing looks unreasonable. One hero image, a few pixels, a chat widget, a reviews widget, a couple of font weights. Every one of them was a sensible request at the time. The problem is not any single decision but the way decisions accumulate.

Why this is a marketing problem is straightforward: most of that list was added by marketing rather than engineering, which means marketing is also the team entitled to remove it.

The five places weight collects

Five places a landing page collects weight — mostly added by marketing

Media is both the largest item and the easiest to reduce. Serve a 3000px original into a slot displayed at 800px and visitors download pixels they will never see. Auto-playing hero video deserves particular scrutiny: it is frequently the most expensive element on the page.

With tags and widgets, the issue is accumulation rather than count. Each channel experiment adds a pixel, and the pixel outlives the experiment. If nobody on the team can state exactly what is attached to the page six months on, you are already in that position.

Build an inventory

The fix starts with a record rather than with technology. List what is attached to each landing page and note four things per item: what it is, who asked for it, what it was added to achieve, and when it can come off.

Inventory verdicts — anything not in the top three is a removal candidate

Make it a rule that a request for a new tag carries an end date, and the list maintains itself. Adding is one click and removing is nobody’s job, so the moment to decide on removal is the moment of addition.

Confirming the clean-up

Once something is gone, verify the effect under identical conditions: same URL, same device profile, median of three runs, once before and once after. Those two numbers are what let you hold a grounded conversation the next time someone asks for “just one more widget”.

Watch the conversion figures alongside. The widget you removed may well have been producing enquiries. Optimising on speed alone can leave you faster and converting worse. Further performance writing is in the performance archive, and tidying a site’s assets and configuration as one piece of work is part of the optimization program.

Next part

We have not yet explained why a tag is heavy. The next part covers what a tracking script actually does — and how to measure its cost yourself.

More on this topic

All technotes

Performance optimization Practical

Who pays for the milliseconds a tag adds?

A tracking script does not stop at downloading. While it is parsed and executed the main thread is busy, and a busy page ignores fingers. That cost is…

Marketers 5 min read

Performance optimization Practical

The object cache: what Redis actually removes

Repeated identical lookups disappear. One slow query does not. Miss that distinction and you will wire up Redis and then ask why nothing got faster.

Developers 7 min read

Performance optimization Practical

The N+1 problem: get_post_meta in a loop, and priming

WP_Query primes caches for its own results only. A hand-built list of IDs gets none of that for free — and that difference decides your query count.

Developers 7 min read

₩270,000 · Join the program