Writing for the web: Introducing SEO

If your content is on the web, but the people who need it can’t find it, it might as well not exist.

Search engines work by analysing pages, working out what they are about, and matching them to what the searcher asks for. Whether you’re a writer or an editor, you want to make it as easy as possible for the search engine to do this accurately.

SEO is the art of helping potential readers find your content.

I will assume that your goal is to attract readers who actually want what your site has to offer – whether products, services or information. There are also techniques for drawing people to your site who aren’t looking for what you offer, but who might be persuaded to want it: these are sometimes described as ‘black hat’ SEO.

SEO advice from Google

Despite challenges from Bing and others, Google remains the dominant general search engine. At Google webmaster central  you can find links to a range of tools including Webmaster guidelines for content and design.

Bing and Yahoo also provide advice for webmasters, but it is more fragmented, and you have to dig for it.


The first step in writing effective web content is to identify what you are writing about, and what your potential readers call it. Large organisations seem particularly prone to using ‘clever’ marketing or business terms to describe their products and services: but if your organisation’s word for a product doesn’t match the word that people who want it type into a search engine, those people are unlikely to find you.

These are your keywords – the words that people use to describe what they want to know or do when they come to your site. There are a number of tools available to identify keywords – some paid, others free – or you can undertake market research. One important source of keywords is people who deal directly with the site’s customers: they are the ones who are most likely to know whether your widget is called a thingummy or a whatsit by the people who use it. Don’t ask the marketing department!

If you don’t have access to any of this, you can use your skills as an editor to understand the potential audience for each page and the language they are likely to use.

Writing search engine–friendly content

Once you’ve identified your keywords, what should you do with them?

Your page title, headings and opening paragraph tell readers what the page is about, so this is where you include the most important of your keywords. Repeat them, and variations of them, throughout the text, but don’t stuff them into every second sentence: any visitors you do attract that way will quickly become bored and leave!

In the body of your content, start with the information that’s most valuable to most people, before moving on to more obscure, technical information, or details of interest to a small segment of your audience – you can even put these on separate, linked, pages. Don’t forget to include your keywords (where relevant) in the alt text for images.


Some web writers swear by keyword metadata, but Google in particular gives relatively little weight to it. Consider that if a Wikipedia article matching your search term exists, it will usually appear in the first few search engine results, despite the fact that while Wikipedia articles have good titles, they have no keyword or description metadata.

The next article in this series will look at metadata in more detail.

Writing for the web: Hyperlinks

Hyperlinks create connections

Hyperlinks connect one image or piece of text with another part of the same web page, a page somewhere else on the same site, or a file somewhere else on the web. If it isn’t clear where a link is supposed to go, there’s something wrong with the link: on most sites, a visitor should not be surprised by the result of clicking on a link.

Links to other sites can be risky, as you have no way of knowing when the other site is rebuilt: the domain name could be sold and the content completely replaced, so site maintenance must include a regular program of checking all external links. It is good practice to let people know if they’re going to leave your site: sometimes a small icon is used for this.

Text hyperlinks should rarely be more than five or six words long – button text is generally only one or two words. Avoid ‘clever’ names for programs or products unless your audience is already familiar with them. Ginny Redish gives the example of a link ‘Children’s Compass’ on the British Museum’s site which led to online tours. (The site has been redesigned and this link no longer exists.)

Avoid ‘click here’ (and more …)! If it looks like a hyperlink, anyone who has been using the web for more than five minutes knows they will need to click to activate it. Ideally, every link should make sense out of context, and be unique on the page: this means that when someone using a screen reader chooses a list of the links on a page, each of them makes sense.

Linking to download files

When you link to something other than an HTML file, or a point within an HTML file, the file must be downloaded to the user’s computer. It may appear within the existing browser window, within a new browser window, or within its own application (PDF files may appear within the browser or within an Acrobat Reader window).

Before including a download file, consider what advantages it has over HTML. PDF is often used where a particular format needs to be preserved (e.g. for some forms). If you provide content in an HTML page and also offer a second copy in another format, you need to ensure that both are updated together.

Within the link to a download file, include the file type and size, so that users can be confident that they have the right software to read it and that it isn’t likely to crash their system (e.g. download A guide to writing hyperlinks [PDF, 3MB]).

Where to place your link

A link is an invitation to go somewhere else. Each time you add a hyperlink, consider whether it would be better placed within the flow of text or at the end of a document: on the one hand in-text links are immediate and their context is clear; on the other they may break up visitors’ ‘flow’.

The test is to put yourself in a reader’s shoes, thinking of the task they’ve come to your site to complete. Will the link move them forward or distract them? There is no one right answer.

Further reading

Redish, Janice (2007) Letting go of the words: writing web content that works. Morgan Kaufmann Publishers.

Writing for the web: Managing content

Why does content need managing?

In my introduction to writing for the web, I identified three challenges for web writers and editors:

  • impatient users
  • missing context
  • content management.

In this article, I’ll address the third of these: content management. If you Google ‘content management’ you will find plenty of advice about selecting content management software. A large site may indeed need this; most can get away with a spreadsheet or small database.

Even if a site doesn’t need content management software, it does need a content manager (whatever their title). Someone needs to take responsibility for ensuring that content standards, policies and procedures exist and are enforced. In an ideal world, every site would have a managing editor with the skills, knowledge and –perhaps most important – the clout to undertake this role.

We don’t live in that ideal world, but if you are editing content for a new or redeveloped site, you have an opportunity to establish the standards, policies and procedures that person will need. (Of course, a site owner who has hired an editor can already be counted among the enlightened few who value their content!)

Getting started

Before you can start writing or editing anything, you need to know what the site is for. Many websites exist because someone told the owner they needed one. Every piece of content on a website should help a visitor complete a task, or support a key business objective. So you need to know what the objectives of the business are, and what tasks a visitor is likely to want to complete.

Once you have documented that, you can start on the content inventory. This means identifying every piece of content on the site and recording information about it. At the very least, you need to know:

  • what’s on the site (or what should be)
  • why it’s there (what visitor task or business objective it supports)
  • when it was last reviewed
  • when it should be removed.

You can record additional details if you wish – file type, word count, location, file name, language, an assessment of quality, metadata and more – but these first four are essential.

Establishing style

Most editors would create a style sheet for virtually every editing job. Mention style sheets in the context of a website, and many people will assume that you’re referring to the visual design of the site, not the language.

Nevertheless, an editorial style sheet is an important tool for maintaining the quality of a website’s content. It is arguably more important in web writing than in other forms of publishing because content is likely to be developed and updated over time by a variety of writers and editors. Without a comprehensive style sheet, each person will make the decisions that seem best at the time, and they won’t all be the same.

It is also worthwhile including guidelines about the voice and tone to be used, the approach you will take to writing hyperlinks, and tips on web writing.

Finding content

When you write or edit content for a website, it’s important to know how visitors are going to find it. Menu structures and labels should be flexible enough to allow for the addition of new content.

Don’t be too concerned with optimal Google page ranks, but do think about the search terms your readers are likely to use. If your potential customers want to know about shoes, don’t refer to your products as footwear.

Establishing workflows

Every site needs a well-documented process by which content is developed, approved, published, reviewed, updated and eventually removed from the site.

This document must specify who’s responsible for making sure that the process is followed – ideally, this is part of the site’s managing editor’s role.

Maintaining content

On most websites, visitors are looking for current information: what’s the price of that fridge today? If a site has information about past events, that information must be separated from current and future events, and clearly labelled. That means that content management requires a review plan – every piece of content must be reviewed and updated regularly (how regularly depends on the site and the individual piece of content).

Where next?

For more on managing content, read:

  • Halvorsen, K (2010) Content strategy for the web
  • Sheffield, R (2008) The web content strategist’s bible: A complete guide to a new and lucrative career for writers of all kinds

Writing for the web: Introduction

What is writing for the web, and how does it differ from writing for other media?

Three of the key challenges for web writers and editors are:

  • impatient users
  • a lack of context
  • the need to manage your content.

Users are impatient

Website visitors are busy, impatient, multi-taskers. They may also be busy and impatient when they’re using other media, but the web seems to magnify these qualities.

They’ve come to your website to find out something or to do something: they want to do it quickly and get on to the next thing on their schedule. There is a good chance that if you don’t have what they want, someone else does: if you do have it, you need to make sure they can find it, understand it, and act on it, quickly.

They will scan a page for their key words: the words they associate with the task, not necessarily the ones you (or your organisation) use. If the words are there, they’ll read a bit more – if not, they’ll try somewhere else.

Context is missing

When you read a book, you are dealing with a physical object. You can see how large it is, and how the sections relate to each other. You can (usually) identify its author, editor, publisher, publication date – all of which help you to assess how trustworthy it is. Thanks to the skill of the designer and other members of the publishing team, you can see how each piece of information you’re looking at relates to the rest of the book.

On the web, your readers may have bypassed your home page and come directly from Google, from Facebook or via a Twitter feed. You (and the site designer) need to make sure that they can answer the following questions:

  • Where am I (what site am I on, and where am I within this site)?
  • What can I do here (on this page/on this site)?
  • Can I trust this page and this site (to tell me the truth and not steal information or infect my computer)?

Content needs management

Because a website is not a physical object, it can be difficult to appreciate its size. It is also easy for published content to languish, unloved and unrevised.

A well-managed site has a clear content management strategy and a content inventory: most sites don’t. To manage a site, you need to know:

  • what’s on the site
  • why it’s there
  • when it was last reviewed
  • when it should be removed.

Content which doesn’t actively contribute to your visitors’ goals makes it harder for them to find the content which does.

What can I do about it?

In my next article, I’ll explain how you can start to address these challenges.

Printing pains

What’s the problem with print?

Have you ever wanted to print a web page? How often do you find that what you have printed has a multi-coloured banner and navigation buttons or worse, text that disappears off the edge of your page?

Why does it occur to so few website designers and developers that you might want to print a page? After all, you might want to:

  • read an article away from the computer
  • show it to a friend
  • file it away for future reference.

What are the options?

There are a number of different ways to provide printer-friendly web pages:

  • style sheets
  • scripting
  • portable document format (PDF) files.

Style sheets

One option is style sheets. These take a little time and thought to set up, but almost no maintenance. The key is to build your site around valid HTML/XHTML and CSS: it’s then quite easy to have one style sheet which controls the appearance of files on screen and a second which controls the appearance of files sent to a printer. (You can also set up style sheets for mobile devices and screen readers, but that’s an article for another day.)

If you don’t want to print a page, find the Print Preview option on your browser and notice the difference between what you see in your browser and what will print. You can find examples at A list apart and456 Berea St.

The main downside of this approach is that (unless they always preview before printing) visitors won’t find out what the print version looks like until they’ve printed it. Oh, and if you don’t check carefully, you can end up with pages that don’t quite fit on your page: on some sites, the printed pages lose 4 or 5 characters from the right margin. I suspect they might have been checked on US letter paper, which is a little wider than A4.


On many large sites, such as Wikipedia, content pages include a button or link labelled ‘Print’ or something similar. In general, these present the visitor with a new page (sometimes in a new window) with menus, ads and similar elements removed.

Like style sheets, these take some work to set up and test, but are relatively low-maintenance after that. They do require an extra click for the user, but also offer more control.

Wikipedia has the added feature of ‘books’: a reader can combine a number of articles into a single file, which can be downloaded as a PDF or ODF file.

PDF files

For many people who have been using the web for a while, PDF is the first format they think of when it comes to printable versions of web pages. PDFs have many advantages: most people have Adobe™ Acrobat® Reader, and if they don’t they can download it free of charge from Adobe.

This is the ideal format if it’s important that the document your visitors print looks exactly the same as the one you see on your computer. While Acrobat itself is not cheap, there is a range of free or inexpensive software available to create your own PDF files.

The downside is that every time you make a change to a document, you need to make it in two files (the HTML and the PDF) or end up with two different versions. Again, there is software available to create both HTML and PDF files from a single source, but this means extra work in the setup stages.

So what’s the answer?

There is no one ‘right’ answer: there are only options. Developers and site owners need to consider the possibility that some visitors will want to print their content, and provide an option. Fail to do so and you risk irritating – even alienating – a group of your site visitors: they may not return.

For advice on the best option for your site, contact me.