Web writing guidelines backed by research

In the past, I have been bothered by the lack of a coherent summary of research on web writing.

In November 2003 the problem was solved by the (US) National Cancer Institute, for the time being, at least.

Random bits of research about usable web content float across my screen like corner-of-the-eye flyspecks. They accumulate as a background to what I teach.

I rarely need state the source of my guidelines. In a one-day workshop, people just want to learn how to fix their web content -- tomorrow.

So I tell them. For example, "You must write short sentences," I say. "You must write short paragraphs. You must put the topic phrase or clause at the start of every paragraph."

I'm happy to learn from NCI that research backs up these guidelines.

Reading random research is risky

You and I may read an awful lot of research findings. But academic rigour is required for a literature survey and and objective analysis of the results.

We are right to be skeptical, because on the Web, the more popular you are, the more popular you become. Popularity can whoosh upwards in an uncontrollable spiral.

Only a tiny, random sample of research results float to the top of search results, and casual surfers simply cannot escape those same few articles from around 1998. Through constant repetition, handy rules of thumb (like "cut text by 50%") become commandments set in stone.

We can't see the big picture by reading random bits of research.

That's why I am very pleased to see the NCI book, based on a survey of the literature on web site usability. It was released on 27 October 2003.

"Research-based Web Design & Usability Guidelines"

This book goes beyond personal opinions about what works on the Web (which are never in short supply).

The authors present selected guidelines that are either strongly supported by research, or considered by a team of experts to be of high importance -- or both.

Frankly, I'm only interested in the findings with cumulative research-based evidence. The remainder are still just personal opinions, albeit* the opinions of experts.

The 128-page book includes a chapter on writing web content. It starts boldly,

"Content is the most important part of a website. If the content does not provide the information needed by users, the website will provide little value no matter how easy it is to use the site."

The NCI team's research-based web writing guidelines are listed below. I have omitted those with high perceived "relative importance" but weak research support.

Guidelines supported by research

Avoid jargon.
Include the primary theme of a paragraph, and the scope of what it covers, in the first sentence of each paragraph (i.e. frontload paragraphs).
Use verbs in the active voice.
A sentence should not contain more than 20 words. A paragraph should not contain more than six sentences.
Make action sequences clear (e.g. list instructions in chronological order).

Those five guidelines would vastly improve many a web page content.

Why the NCI?

Here's a puzzle. Why was it the US National Cancer Institute who initiated this project?

Originally the book was intended as a tool to improve federal sites on cancer. But the project grew, and the guidelines are useful for any web site whatsoever.

Nice of them to share, don't you think?

Link to "Research-based Web Design & Usability Guidelines":
http://usability.gov/guidelines

(*At last I got to use Princess Diana's favourite word.)

QWC home

contact

Contented.com
2012 Rachel McAlpine PO Box 19184 Wellington 6149 New Zealand