A visitor does not read your page, they scan it. They look at the heading at the top, jump to the subheadings that look like they answer their question, and only start reading once something matches. Google does much the same thing: the headings say what the page is about and how it is put together.
This article covers the text as it sits on the page: the H1, the subheadings, how much text a page needs, the keywords tag that has done nothing for years, and the few lines of source that decide whether a browser renders your markup the way you wrote it. They are the checks a report lists under “Content”.
The H1: the heading above your page
The H1 is the most important heading on the page and sits in the source as <h1>. It does two jobs at once. For the visitor who just arrived from Google it is confirmation they are in the right place: they searched for “roof repair Etten-Leur” and they see those words again in the heading. For Google it is a strong signal about the subject of the page.
With no H1 that confirmation is missing. The visitor lands in a block of text with no announcement and has to work out where they are. That costs seconds you usually do not have.
A single word is too little. “Services” or “Welcome” says nothing the title did not already say. Write a heading that carries the promise of the page: “Roof repair in Etten-Leur, on site within 48 hours”. Under twenty characters that rarely fits.
More than one H1 is not an error
Plenty of SEO tools turn a page red the moment it has two H1s. That is a stubborn myth. Google is clear about it: there is no ideal number of headings, and using them out of order is not a problem. Search Central says so in as many words in its documentation on heading tags.
Our report therefore counts the H1s but deducts nothing for them. What does deserve attention is the intent behind it: if your theme makes every section title an H1 for styling reasons, no heading stands out any more. That is a design problem, not a ranking problem.
Subheadings: the structure under the heading
Below the H1 your text should fall into sections, each with its own H2 and an H3 beneath where needed. Google asks for this explicitly: break long content into paragraphs and sections, and give those sections headings so people can navigate them.
A page without subheadings is a wall. Nobody reads a wall, not even when the text is good. Six paragraphs with four headings above them get read; the same six paragraphs in one run get closed.
Do not skip levels. Jumping from an H2 to an H4 will not hurt your position in Google, but screen readers use those levels to build a table of contents, and a gap makes that table useless. The same goes for headings that are only bold text: if it is a heading, make it one.
How much text does a page need?
There is no minimum. Google is unusually direct here: the length of the content alone says nothing about its quality, and there is no magical word count. A contact page of eighty words can be perfect, and two thousand words of filler will not move you up.
Our report counts the words, the paragraphs and the average sentence length, but scores none of it. It is there to give you a picture. If a service page comes out at fifty words, that was usually not a decision but a page that never got finished, and that is the signal worth acting on.
What does matter for reading: paragraphs of three or four sentences, the occasional list, and sentences that do not run across three lines. That reads on a phone, and a phone is where most of your visitors arrive.
The keywords tag: leave it or remove it
The <meta name="keywords"> tag is a leftover from the nineties, when you could tell search engines which words you wanted to be found on. It was abused so thoroughly that Google publicly stopped using it back in 2009. The tag does nothing today.
It does no harm either, so there is no need to make a project of it. What is a waste is the time spent on it: if somebody in your organisation fills in a row of keywords for every new page, they are doing work nobody reads.
What a browser needs
Two lines at the top of your source decide whether the rest behaves. The first is the character set: <meta charset="utf-8">. Leave it out and a browser is left guessing, and then “café” turns into “café” and half your accented words become unreadable.
The second is the doctype: <!DOCTYPE html>, and it has to be the very first line of the document. If it is missing, or an old variant is used, the browser switches to a compatibility mode with layout rules from the nineties. Your CSS then does something other than you expect and you go hunting for a bug that is not there.
Older pages also still carry tags that have left HTML, such as <center>, <font> and <marquee>. They mostly still work, but they are a sign that the page has not been touched in a long time, and that is rarely the only thing lagging behind.
What you can do today
- Open your most important page and check there is a clear heading at the top saying what you offer.
- Scroll through it and count the subheadings. If there is more than a screen of text without one, add one.
- Read the first paragraph out loud. If a sentence runs across three lines, cut it in half.
- Check whether anyone is still maintaining keywords in a keywords tag, and stop.
- View the source and confirm the first line is
<!DOCTYPE html>and that a charset is declared.
Want all of this checked in one go? The free SEO check runs through these points and shows you what was found on your page for each of them.
