Chapter 5: Test your own website

There are lots of offers from specialized consulting firms where you can get your website evaluated from every possible perspective. Some tests of course are incredibly complicated, require a lot of knowledge or special tools, but others you can do yourself. What you are able to test yourself you should, to level-up your website.

This part consists of a long list of simple tests for an initial health check of a website!

You can probably run the majority of the following tests on your own. We will go through what needs to be tested. Why you do it and what to look for. I suspect that in some parts, you may need help from colleagues or to read a bit about the topics you do not yet know much about. Google is your friend and I will try to use terminology that makes it easy to google for more information if you need it.

How to document your test

The evaluation begins with the website’s homepage. Please sample other pages when the respective assessment requires it or you know of any page that you think is suspicious. A spreadsheet to document your tests on is downloadable at tba.nu/wetools

Since the tools to check websites sometimes disappear, and better ones show up, you can always look at the above URL to see what tools and tricks I’m using myself at any given time.

Whether you use the template I suggest for documentation or your own, it makes sense to:

  • Enter dates for the test so you can go back to old tests for future reference.
  • Give each item a rating, such as Point 1.3 failed or Point 1.3 passed, to rate point 1.3 of the length of the page title.
  • Add a comment to the check point if needed showing which page you checked, or other details worth mentioning.

Now to the evaluation points.

1. SEO

Good search engine optimization is all about a website being accessible to search engine robots. That information on the website is accurately described and nothing is missing.

1.1 Indexable for search engines

Figure 75: SEO Doctor gives an SEO score.
Figure 75: SEO Doctor gives an SEO score.

If a webpage is not accessible to search engines, it becomes difficult for users of search engines to find. This means that you do not get good returns on the time you spend working on website content since it is not found in searches.

Checks to run:

  • Browser extensions, such as SEO Doctor, will protest if the page has settings that do not allow a search engine to index.
  • The website’s HTML code has metadata tags that say noindex, or canonical tags refers to another URL which disqualify the current page from being indexed.
  • Check if there is a file called robots.txt in the root of the website. In such a file, there can sometimes be instructions to search engines that stop the website in whole or in part from being indexed. Exclusions are written with the prefix disallow and then an address follows that search engines should not crawl.

1.2 Duplicate content

Information and texts should be as unique as possible. For example, it is not good to have multiple pages with identical titles, headers or content. As far as possible, each page should be unique.

Checks to run:

  • Google Search Console has a view for pages with similar or duplicate content.

1.3 Page title’s length is under 60 characters

The title of a page should be short for it to be easy to read. There is a limited length in a search engine’s results page; it is not to be taken for granted that the full title is shown.

Checks to run:

  • Browser extensions, such as SEO Doctor, will protest if the page title is considered too long.
  • Count the characters in the title tag in the HTML code.

1.4 Page title is readable and understandable in the search engine results page

Early on in the page title, it is important to specify the uniqueness and what is important on the page, because search engines do not have infinite space for your page titles.

Checks to run:

  • Try searching for your pages with the major search engines, via both a desktop computer and smaller devices. Check if the titles manage to communicate the general content of each page.

1.5 Page title contains relevant keywords that describe the page

We cannot assume that an Internet user reads everything you write. Therefore, your title needs to be short, pithy and confirm the users’ search word as early as possible in the page title.

Checks to run:

  • Try searching for your pages with the major search engines, via both desktop and a mobile device. Check that titles contain potential keywords that your target audience might use.

1.6 Correct headings are used

Only one main header element and at least one sub-header are preferable. Sub-headers should form a proper document structure without any skipped levels. If the website uses HTML5, you should try to have as few h1-headers as possible to make it clear what is most important.

Checks to run:

  • Browser extensions, such as SEO Doctor, will warn you if there is only one main header and no sub-headers.
  • View the HTML of your pages and see if both an <h1> and <h2> tag are present.
  • Check that header levels are not skipped, going from an <h1> to an <h3>, for instance.

1.7 Search engine friendly URLs

A website address is important as it can be used outside the website owner’s control. Among other things, it should not be too long to memorize or read. Also an address sometimes becomes a link when it is posted online, and then the wording of the address is supposedly an understandable hypertext link.

Figure 76: Temporary things, such as 'phpsession', are not suitable to be in a URL.
Figure 76: Temporary things, such as ‘phpsession’, are not suitable to be in a URL.

Checks to run:

  • Surf around on your website in search of addresses that look hard to memorize even for a short space of time, addresses that do not describe the contents, or cannot easily be read aloud. A website’s internal system-structure should not be visible.

1.8 Descriptive text on all important pages

A text that describes the page, a so-called meta-description, is a short summary of a web page’s content and purpose. Although the text is not always visible to visitors, it is an opportunity to associate the page with qualitative keywords and synonyms. Sometimes the text appears as body text below the link in the search engine’s results pages.

Checks to run:

  • Browser extensions, such as SEO Doctor, usually look at length or the presence of a description text. It is also possible to check the HTML code manually, for something similar to below:
    <meta name=”description” content=”Description of the page.” />

1.9 Reasonable number of links

Too many links on a page indicates several things, and to a search engine indicates not only suspicious behavior, but also a messy structure and the fact that a visitor may find it difficult to get an overview of the page – especially on a small screen. Fewer than 100 links is preferable on all pages regardless of whether the links point to pages within the website or to other sites.

Checks to run:

  • Look around your website for pages that have an excessive number of links. The browser extension Accessibility Evaluation Checker has a navigation report tool that counts links on a page.

1.10 Pictures have alternative texts

Images that have content that complements the page text need to be described for visitors who cannot see them. Among those who cannot see images are search engines, the blind and those who for some reason do not load images when surfing – perhaps because of a shaky internet connection. An image that conveys important content should have an alternative descriptive text, or an empty alternative text if nothing is suitable to describe. This is particularly important for linked images since the image’s alternative text will be the hypertext link in some cases.

Checks to run:

  • Browser extensions, such as SEO Doctor, will point out if an image has no alt text. Otherwise, you can manually browse around and check the HTML code to see if the <img> tags have the alt attribute set to something meaningful. Decorating pictures should not have a descriptive alternative text, but should appear as follows:
    <img src = “image.jpg” alt = “” />

1.11 Structured description of the information

Is there content that in part or in whole describes itself? This is used by search engines to present a company’s geographical location, list its events and contact information, among other things.

Checks to run:

  • Browser extensions, such as Operator, will show in your browser if a page has been found to offer structured data.
  • Check pages manually one by one with Google’s Structured Data Testing Tool.
  • Verify in the Google Search Console which pages on your website they have discovered to contain structured data.

2. Web analytics

To draw conclusions from an analysis of a website, you need to save information in advance so that there is something to analyze. It is therefore important that you keep up with what is possible in web analytics.

2.1 Current visitor tracking scripts

To do a good analysis afterwards of how a website is used, it is important to first collect relevant data. There is variation in how tracking code can be written and sometimes things that need to be tracked will require additions to the code you use. Therefore, it is important to keep your tracking tool up-to-date.

Checks to run:

  • Manually compare the proposed tracking code from the service provider with the code you are using and check if there is any interesting information to take advantage of.
  • Browser extensions, such as SEO Doctor, will tell you if tracking code does not seem to exist on a web page.

2.2 Tracks the use of website search

If you collect statistics on the use of the website’s own search engine, you can draw conclusions from this and often improve user experience. If nothing else, it is useful to collect data on what visitors are looking for because you might want to do search analytics in the future.

Checks to run:

  • Check that your analytics tool, such as Google Analytics, is collecting data from the website’s internal search engine. Usually a relatively simple setting in the web analytics tool which does not require you to change the website architecture itself.

3. Performance

To optimize for performance is to give visitors as good an experience as possible. Visitors have different technical set ups and equipment in terms of connection speed, the type of connection and features in their browsers.

3.1 Reasonable time for loading the page

How long it takes before the first byte is sent, as well as a balance between the necessary amount of data and the total page load-time. In other words, an assessment whether the experience can be made better or faster. What you think is a reasonable time is up to you, but a regular page should be loaded in less than 2 seconds on a desktop computer in any case. On a mobile, you can have 5 seconds as your upper limit.

Checks to run:

  • Browser extensions, such as app.telemetry Page Speed Monitor, display this information regarding your own desktop computer performance.
  • Look for the appropriate view in your web analytics tool to see quantitative data from real user monitoring (RUM) at your website.

3.2 Compression of text files

Compression is about making sure that what is sent to the visitor is as small as possible. A small amount of information is faster to transfer.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will show if the website you are evaluating sends files in a compressed format.

3.3 Usage of the browser cache

You do not want to send unchanged files every time to visitors who have already downloaded them; it takes unnecessary time and degrades the experience. This is particularly important for files that are rarely or never updated. For example, logos and icons.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will tell you if the website you are checking instructs the browser cache on file lifespan.

3.4 Scripts and style sheets are sent in a compact format

For a web page not to be unnecessarily slow to display for the visitor, you want to remove as many spaces, line breaks, etc. as possible from the code.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will show in your browser if the website you are evaluating sends files in a minified format.

3.5 Images are optimized for fast transfer

A picture posted on the Web needs to have the right balance between image quality and reasonable file size. Often, image file sizes can be reduced even further without a visible deterioration of image quality.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will tell you if the images on the page can be further optimized.

3.6 Reasonable number of background images, scripts and stylesheets

If the website is dependent on many files to display properly, there is a prolonged waiting time for the visitor. Each file takes a while before it starts to download and that is why you might want a balance in the number of files required. In many cases, you can combine many smaller files into a single larger file, meaning fewer files in the queue for download.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, tell you if they feel that there are too many files to download and which ones ought to be combined into fewer files, or placed on a CDN.

3.7 Requesting files and pages that do not exist

Neither visitors nor search engines like error pages or that files are missing that they depend on for correct presentation. Although the page might look correct on the screen, behind the scenes requests to non-existing files might be made, contributing to a slower experience.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, usually report if a file that is requested is not available. It can sometimes be difficult to detect missing images or custom fonts.
  • Google Search Console has a view for addresses that give errors 404’s , or system failures in the 500 series.
  • Some content management systems, such as Episerver CMS, have reports and tools to check links.
  • Check that all important links from other websites to yours point to functional pages.

3.8 Minimal amount of scripts and CSS in page code

All frequently used functions and instructions on appearance need to be stored in a way that makes them reusable between pages. In other words, the individual page is supposed to have as little information as possible when this information is common with other pages. You put the common code / instructions in an external file for joint reuse. This does not contradict the performance practice to include critical, or structural, CSS on every page.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, may complain when finding too much inline code.
  • You can manually check if there are a lot of scripts or style tags embedded in HTML code. In addition, frequent use of style attributes on HTML elements is something to watch out for.

3.9 Images are not scaled down using CSS or HTML

A large image is generally slower than a smaller one to download for a visitor. Therefore, it is good practice that the images used on the website are shown in their actual size, in pixels. Think about whether you regard this as responsive image enhancement, or if you think it is a performance issue to send images with a higher resolution than needed for the available space.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will tell you if there is potential to reduce load-time.

3.10 Identical files are not referenced

Files that do, or contain, the same thing should not be downloaded multiple times. An example of this is using multiple versions of script libraries at the same time for one page, or retrieving them from different addresses. Sometimes it happens that you have accidentally, for example, put the logo file in several locations so that multiple files with identical content are sent unnecessarily to users.

Checks to run:

  • Web services, like Pagespeed, can point out, somewhat cryptically, that consistent addressing is needed. This suggests that identical files are downloaded from different addresses.

3.11 Reasonable amount of scripts in the page head

In many cases, it is recommended to have tracking scripts and other external dependencies at the end of the page’s source code. Because these may be interdependent, it is something you should plan for already in the design phase or in a major revision. It is especially important for the sake of mobile users to plan when scripts run, or if they are loaded only when needed.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, will indicate if there is an unusually large amount of Javascript blocking the page’s earliest opportunity to display, its rendering in other words.

3.12 Content networks are used when necessary

Many sites use common features, including collecting statistics and providing a richer experience, by way of, for example, image carousels. Instead of having their own copies of files that are required, you can use a Content Delivery Network (CDN) to improve performance on a website. This is particularly relevant for those who have not yet worked on minimizing the number of files. Do you have many self-created static files, such as images and other media? Then a personal subdomain such as media.mywebsite.com may be of help.

Checks to run:

  • Browser extensions and web services, like Google Pagespeed and YSlow, tell you if you need a CDN. A CDN can be a good idea for files that many websites use, such as Jquery, Modernizr and others.

4. Accessibility and Usability

Accessibility and usability are about caring for visitors and meeting their needs. Since we have no control over the user’s context, the website needs to be accessible to as many people as possible to achieve its purpose. It should be easy to understand, users are supposed to do the things they want to do, such as buying something, instead of trying to understand what is on offer and how to proceed.

4.1 Website validates the chosen code standard

Website validation is important for many reasons. One of them is that it reduces the risk of technical barriers arising for your visitors, regardless of whether the user has a disability or not. Validation reports are commonly divided into errors and warnings.

Errors should not be tolerated and most warnings are important to look into.

Figure 77: W3C offers a great markup validation service.
Figure 77: W3C offers a great markup validation service38.

Checks to run:

  • Browser extensions for validation or the W3C validation tests.

4.2 Using correct header structure

Pages with real headers (h1 – h6 in the HTML code) is a support for visitors who do not actually see the page. Headers can be used as shortcuts and table of contents for those with screen readers, and even search engines.

All too common mistakes are using bold text or otherwise making plain text look like a header, or having a large body of text that lacks sub-headers.

Figure 78: Browser add-ons, such as Accessibility Evaluation Toolbar, find shortcomings.
Figure 78: Browser add-ons, such as Accessibility Evaluation Toolbar, find shortcomings.

Checks to run:

  • Browser extensions, such as Accessibility Evaluation Toolbar, offer a range of reports on a page. You can manually check by going through the HTML code and hunting for header tags like <h2>, <h3> and so on.

4.3 Anchor-texts are descriptive

When a user skims a page and finds a link, the link text should explain where the link points to. You cannot assume that a user reads the text around the link.

It is, unfortunately, all too common that link texts are phrased as ‘read more here’ or ‘go here’ which does not suggest either the reason for the link, or what is behind it.

Checks to run:

  • Browser extensions, such as Accessibility Evaluation Toolbar, offer a navigation report summarizing a page’s links. Otherwise, check them manually link by link.

4.4 Link titles not used for non-essential information

The title of a link is the small notification that may pop up if you hold the mouse pointer over a link for a while. The function is to give additional information about the link itself, for instance the size of a PDF file to download. If the link’s text is repeated in the link title or if it does not add anything of value, it will just be an unnecessary annoyance to visitors. Keep in mind that link titles are rarely seen by those with a touchscreen.

Checks to run:

  • Browser extensions, like Accessibility Evaluation Toolbar, provide navigational reports that summarize a page’s links. Otherwise, you can manually check every link, by hovering the mouse pointer over them, or looking at the HTML code in the attribute alt or title used by search engine crawlers.

4.5 Favorite icon is present

A favorite icon (favicon) is a small image used in the address bar, browser tabs and bookmarks among other things. It helps users to navigate between tabs and recognize bookmarks among other things.

Checks to run:

  • Look in the address bar or tab in a browser to see if there is a unique icon for the website.

4.6 Possible to navigate with keyboard

Using the keyboard to navigate a website is necessary for some, while for others it is about good ergonomics. Keyboard navigation is when, for example, you use the tab key to move between links and form fields, or use special hotkeys to jump to the next header or list.

Checks to run:

  • Browser extensions, like Accessibility Evaluation Toolbar, provide navigational reports that summarize a page’s access keys. You can manually check how well you can use the Tab key to move between form fields, links etc.

4.7 Texts are written to be read by a human – not with exaggerated SEO

When text is written for the web, the first and foremost concern should be users as readers, all other aspects are secondary. Except for adapting the style to suit the Web, such as putting the most important thing first in a sentence or paragraph, we should be wary of focusing on keyword optimization too much if it causes a human reader difficulty.

It is all too common with peculiarly formulated page titles, headers and links, often due to eagerness to be the highest ranked on search engines. Visitors can see through this easier than many seem to think, and they will then have a less positive impression of the website.

Checks to run:

  • Go ahead and critically examine texts. Are they written primarily for a human reader or is it mainly a lure for visitors? Headers and other text crammed with potential keywords are a warning sign.

4.8 Language set in the source code

For both search engines and screen readers, it is easier to know the language if it is set. The language is a lang attribute on the html tag, such as lang=”sv-sv” for Swedish.

Checks to run:

  • Display the HTML code, look for the HTML element’s start tag and check if it indicates the correct language.

4.9 Not depending on browser features

To make a website appealing to as many people as possible, it is important to guarantee users a good experience even when they do not have certain technologies installed on their browsers, or have difficulty using certain technologies. Flash, Javascript, and to some extent CSS risk causing bad user experiences.

We must not refrain from using the technology simply because not everyone can use it, but by working based on the lowest common denominator and then refine the interface for those who support it. This practice is known as progressive enhancement.

Checks to run:

  • Use browser extensions, or change your browser’s settings manually, to turn off Javascript and CSS (style sheets). Is the content presented in a logical order? Can you still use the website?

4.10 Specifies image sizes in HTML

If the visitor’s browser receives instructions on image size, it does not need to redraw the page again when the image is fully downloaded, so reducing the perception of a flickering screen. We get a more robust impression of the page when content does not jump around while images are loaded.

Checks to run:

  • Browser extensions and web services, like Pagespeed and YSlow, report if images in the HTML code are not provided with their dimensions.
  • Browse around manually and carefully pay attention to what happens around images. Remember that it may vary depending on your device’s screen size, for example, responsive websites display fewer columns on mobile screens where this can have a greater impact.

4.11 Works with and without the www prefix

The website should work regardless of whether the visitor has entered the www prefix or not. Make sure to use only one of these variants by forwarding the visitor to the page at the address you have chosen. This is to ensure that things work no matter how one enters, but that it is clear what is considered the correct address.

Checks to run:

  • Try to add or remove the www. prefix in the browser’s address bar, not only for the home page. Do you end up on the correct page? Is only one variant used? Great.

4.12 Only one domain is used for the website

In order not to cause uncertainty among visitors about what your address is, it is preferable to use only a single domain for all of your web presence. A rather common example is that blogs are placed on subdomains such as blog.example.com, or that sub-pages with login requirements on subdomains like customers.example.com

Several domains are quite in order as long as the user will not see them or suffer from this. For example, it is alright for images to be retrieved from a custom domain to optimize performance.

When it comes to intranets, this can cause obvious problems when users are allowed to login even though they are not in the organization’s network. Then we can discover that certain links on the intranet point to internal domains that are not (yet?) available even though they are perceived to be a part of the intranet.

Checks to run:

  • Browse around manually on the website and check if subdomains show up or whether completely different domains appear.

4.13 RSS subscriptions can be detected

If a website offers news subscription through RSS-technology, it should be marked up in code for so-called auto-discovery. This means that any subscriptions are listed among the page’s other metadata and are easy to detect technically.

Checks to run:

  • View HTML code similar to below, or install a browser extension that detects RSS feeds:
    <link href=”/rss/” type=”application/rss+xml” rel=”alternate” title=”Subscribe” />

14.4 Useful error pages

A so-called 404 page should tell the user what happened and help the user to find what they requested. For your own sake, you should also log how people ended up on the error page, where they came from, and more.

Checks to run:

  • Type an incorrect address to the website and see what happens. Will it present a page that tells you that an error occurred? Does it show whose website you are on? Is there help to find what you were looking for? Are there obvious ways to reach the home page?

4.15 No surprises when scrolling

It is all too common that a page stops scrolling when the mouse pointer, or finger on a touch screen, ends up in a map or iframe. It is of course very frustrating and for most visitors very illogical.

Checks to run:

  • If you do have maps, place your mouse pointer over the map element and begin to scroll down. Did you zoom the map when the mouse pointer hovered over the map?
  • Try to scroll on a device with a touch screen and assess whether the behavior is logical. It is possibly a bigger problem for maps or iframes, since it is not within the visible area of the screen and the user triggers it during fast down-scrolling.
  • Also, check if you have iframes (a peephole on the page where you download content from another location) as these, among many other things, degrade usability.

4.16 Enough distance between links, buttons, etc.

A common usability problem that we all can suffer from is that interactive elements on a website are placed too close to one another. Most often, this becomes a problem with lists of links where they are stacked vertically. Clicking on the wrong link causes irritation, and together with slow loading, this could be enough reason for a user to give up.

Checks to run:

  • Appears in Pagespeed Insight (called tap targets) but can also be checked manually on any touch screen where you use your unpracticed thumb to see if you can hit the right link – every time. Is it good enough?

4:17 Acceptable text size

Text size has become more important than ever since many people use the Web outdoors in daylight. It provides challenges with size, contrast between text and background, among other things.

Checks to run:

  • Appears in Pagespeed Insight but can also be controlled manually by taking your mobile device outdoors. If you have a tablet, it is worth testing too because they tend to have more problems with reflections on the screen because of the larger surface. Is it easy to use?

4.18 Zoomable, also on mobile

There are many websites where the creators seem to have thought that now that they have made it so suitable for mobile users, no in- and out-zooming is needed. Unfortunately, disabling the ability to zoom is a stupid thing to do. For example, those with motor disabilities may need to zoom in a little extra on links to be certain of hitting the one they want.

Checks to run:

  • Enter the website with your mobile or tablet. Is it possible to zoom in the same way you normally zoom with the device?
  • Use your browser’s feature to zoom and see if it works.

4:19 Icons for the website

A user can add a shortcut to a website on their home screen on many kinds of devices. You have probably seen invitations to do this while browsing the Web with a mobile. Just like an app icon, this icon needs to be highly recognizable. This is a common requirement with logos or other familiar visual marks.

Figure 79: When designing icon's for home screens, recognition is important.
Figure 79: When designing icon’s for home screens, recognition is important.

Make sure that you have enough image resolution so that it does not appear blurred or ugly on some high-end screens.

Checks to run:

  • In addition to adding the website to the home screen on your device, you can check the HTML code for something similar to the examples below.
    Example for Iphone / Ipad, Android and others
    <link href=”/apple-touch-icon.png” rel=”apple-touch-icon” /> <link href=”/apple-touch-icon-precomposed.png” rel=”apple-touch-icon-precomposed” />
    Example for Windows 8
    <meta content=”#123456″ name=”msapplication-TileColor” />
    <meta content=”/win8-tile-bild.png” name=”msapplication-TileImage” />

4:20 Useable printouts

Figure 80: A printout of my website discloses the URLs behind the links.
Figure 80: A printout of my website discloses the URLs behind the links.

Not that you should ask people to do printouts on paper, but there is still reason to keep an eye on how printing looks and functions. Partly because some people read PDF print versions on their e-readers (which sometimes lack Internet connectivity, and this messes up the links), and also for archiving of your website by you or others.

Checks to run:

  • Print a few different types of pages, either on paper but rather as a PDF draft. Is all the information properly presented? Can you see where the links point to?

5. Others

5.1 Forms and other sensitive information is sent through a secure channel

When a form on the Web is sent back to the web server, sometimes it is transmitted in a way that makes all the content readable, in plain text, if someone is monitoring what is sent over the network. If a login form, your password can be read by every network administrator who oversees the network’s traffic. However, your visitors can also be sitting on an unsecured wireless network in a hotel or be vulnerable to other threats, such as man-in-the-middle attacks, which means that unwelcome people can monitor everything that is sent. This also applies to the content of web pages that users receive, not only forms that are sent; sensitive information in the page’s content can theoretically be read in transit between the server and the user’s computer.

What every website should do is to protect sensitive traffic, secure it by passing it through the HTTPS protocol. This means that the information is transmitted in an encrypted form via SSL/TLS.

It is important that visitors can verify that traffic is protected so that they feel safe to use the website. Therefore, it is a good practice to use HTTPS on the entire website even if the need exists only on parts of the website. To verify that information is sent securely through your organization’s own APIs to user devices, you probably need to contact a developer.

Checks to run:

  • There are different ways to see if a website communicates via HTTPS, often displayed as https:// or a padlock in the address bar. You can also look at the HTML code and verify that https:// is used, or files called with just // are safe on a page requested with https://

Continue to the book’s outro, references and reading suggestions ›

Web Strategy for Everyone

Table of Contents