How much text markup is good, before it becomes bad?

Question

In 1997 Jakob Nielsen told us about how users read on the web. (Actually in a typically headline grabbing way, he says they don't and then tells us how they do!)

Nielsen is a big proponent of using bold text - often with several items marked up in this way per paragraph (despite also telling us to use one idea per paragraph).

There is always going to be an argument that it depends on context, so to put the question into a context we can all see, I'm specifically thinking about questions and answers here on stack exchange where there are frequently long bodies of text, but also bearing in mind that visitors are short on time and need to get the 'takeaways' clearly and succinctly.

Clearly, some markup helps with conveying ideas and making the points easily scannable, but on the other hand for those wanting to read an article in full, there can be too much markup that it makes the text harder to read - especially for the (ahem) older readers who find some aspects of readability increasingly more of a problem. Therefore, the effectiveness of markup depends on the reader as much as the writer's style.

There are, of course, a huge number of factors that affect the desire and ability to read or scan text, but is there any modern research (or examples) for the web that show, for most readers, how much text markup is just right?.

Answer 1

I don't have any newer research to point to in this answer, but I believe that the question of how much is too much is not objectively answerable. Following is my subjective opinion.

---

It depends on the situation; the audience; and the context. The rest of my answer is specific to reading on the web.

People will generally scan to see if an article looks interesting, and then, if it appears to be, they (might) read it all. Bolding your key points is a way to improve scanning.

Formatting text makes scanning better but makes straight reading more difficult (for some). So the question comes down to a balance between the two. Which you choose should depend on your relative value of each in the application that you're using.

My (subjective) take on it is readers are going to read a full article even with text formatting, But scanners are lost without formatting. So as you may have guessed, I prefer to format the text.