Your website is fine on your laptop. That is not where you are losing work.
You look at your website on a big monitor, on office wifi, with every image already cached from the last ten times you loaded it. It looks great.
Your customer looks at it on a phone with a cracked screen, outdoors, one-handed, on one bar of signal, while something else is happening. That is the only version of your website that matters, and almost nobody tests it.
The background video is the most expensive decision on the page
An autoplaying video behind the hero section looks superb in a design review. It is also usually the heaviest thing on the page and it loads before the content.
On good wifi it is cinematic. On a weak connection it is a white rectangle where your headline should be, for several seconds. People do not wait out a white rectangle. They go back.
If you want motion up top, a still image that loads instantly beats a video that arrives late every time.
Thumbs are not cursors
A mouse pointer is one pixel wide and perfectly accurate. A thumb is not. Buttons sized for a cursor end up too small and too close together on a phone, and people tap the wrong one or give up.
The specific things worth checking: is your phone number an actual tap-to-call link, or just text someone has to copy out by hand? Can you hit the menu button without a second attempt? Are two links sitting so close that your thumb covers both?
Your form is asking for too much
Every field you add loses people, and the ones you lose are not evenly distributed. The ones in a hurry go first, which tends to be the ones with an urgent job and a budget.
Name, phone number and one line about what they need is usually enough to start a conversation. Asking for a full address, a budget range, a preferred contact method and how they heard about you, before you have earned anything, is asking a stranger to do paperwork.
You can get the rest when you call them back. That is what the call is for.
The furniture is eating the screen
Count what is permanently stuck to the screen on your phone: a cookie banner at the bottom, a sticky header at the top, a chat bubble in the corner, maybe a promotional bar above the header.
On a laptop these are a minor tax. On a phone they can take up half the screen, leaving a narrow strip of your actual website visible between them. Look at your own site and work out what percentage of the screen is yours and what percentage belongs to the widgets.
How to test it properly
Not in a simulator. On a real phone, like a real customer:
- Turn wifi off. Cellular only. That is the connection most first visits arrive on.
- Arrive the way they arrive. Search for your business, tap the result. Do not type the address in. The journey is part of the test.
- Stand up and use one hand. Sitting at a desk with two hands is not the situation.
- Try to contact yourself, and finish it. All the way through to the confirmation. Then check whether the enquiry actually arrived anywhere, and how long it took anyone to notice.
- Hand it to two people. Someone over sixty and someone under twenty. They will fail at different places and both failures are real.
The question is not whether your website works. It is whether it works in a parking lot, on one bar, with a toddler in the other arm. That is the bar, and it is much lower than most sites clear.