Practical SEO Guide

Inhalt

How to use this guide

No prior SEO knowledge is necessary to use this guide, however some previous skills in building and maintaining your own website will be helpful. All websites use different tools and frameworks (e.g. Wordpress), and it isn't in the scope of this guide to include information on all website technologies. If you ever get stuck, try searching for specific terms in relation to your Content Management System (CMS).

If you really want to get the most out of SEO it is important to have a solid understanding of how it works, not just go through checklists. Although it seems cumbersome, understanding the "Technical Background" section will be vitally important for your ability to make decisions and understand later sections. To help develop your SEO strategy, the "Political Context" section provides information about how Google handles feminist-related terms and how anti-feminist sites use SEO. This section also includes advice for those with an existing site.

NOTE: Prominence in search results has a lot of advantages, but can also bring problems. If a group is more visible, it can also be attacked more easily. Find resources about security and hate speech on the web here:

TECHNICAL BACKGROUND

How do search engines work?

Search Engines first have to know which websites exist in order to show relevant search results. To figure this out, search engines use special programs called bots to continuously search the web and find sites. The process is called crawling. These bots are also sometimes called "spiders".

The bots crawl endlessly through the internet by following links that bring them to the next site. Sites they find relevant end up in a database. As soon as the bots find a site, they decipher the source code, try to understand the content, and decide which are the related search terms. This process is called indexing (cataloging) and the database is called the index. These databases are huge (over 100 million GB) and are saved in computing centers all over the world. Search engines can manage to give an answer to a search term in just a few milliseconds due to the fact that these index databases have already pre-sorted the internet.

Relevance and popularity are the most important ranking factors that decide whether a site is ranked first, second, third, etc. The search engines try to understand not only whether the results are relevant to the search term, but also how users behave on that site. They evaluate user signals to see whether or not users found the information they were looking for (popularity).

Search engines in detail

Google Ranking Factors

So how is it possible to design a website so that it‘s relevant for certain search terms?

Ranking factors are extremely complex – today Google uses over a hundred different ones. Unfortunately Google does not publish which ones it uses because it would hinder their competitiveness in the market.

Nonetheless, there are a few strategies that have proven efficient. They are covered in the following four sections. This background information is vital to understanding the practical "How to" section, which can be found in the last section.

1. Websites from a search engine‘s perspective

Search engines see websites differently than people do. If a search engine‘s crawler comes across a website, it follows its links to find the single pages of it, and then deciphers its source code in order to understand the site as a whole.

The structure of a website is therefore quite important. Only one URL can rank for one search term (or several terms, if they all serve the same search intent, but we will come to that later). Say someone searches the term „Feminist movement“, and the same website has five different pages related to this topic. If the contents of the five different pages are too similar to one another, the search engine will not be able to differentiate between them and in the end won't choose any of them. This is called duplicate content in SEO jargon.

To avoid this, each sublink should have its own clear topic. A tree structure can be quite helpful in this regard: a main site has links to its main topics and from there are links to sites with more detailed information:

Tree structure: Correct structure for SEO

It is much easier for a search engine to understand a website‘s content if it uses this tree structure. It should be easy for both users and search engine crawlers to get from the main page to the most important pages (those that should rank well) in as little clicks as possible.

How do the search engines understand what a site is about? They evaluate the previously mentioned source code. The most important thing to the bot is the text content – images, videos or other kinds of content are more difficult for the bot to evaluate and understand. They will be taken into account, but make sure not to have important content solely in image or video format.

During the evaluation, the bot looks for keywords in the content. They also weigh different sections differently – for example, the first paragraph and title are more important than the last paragraph. The bots assume that the last paragraph is not as relevant as the first, and so on.

The title and description of the site are also quite important. Both of these values aren‘t visible on the site, but exist in the source code. These show up in the search results list:

SEO title and description

Google creates its own title and description for your site, but you can also do this yourself, to increase the chances that Google chooses yours. The title and description are in the source code only. Keywords in the title are also important for the ranking factor.

It used to work quite well to hide as many keywords as possible on the site – for example at the very bottom of the site in the same color as the background. Today, this is not as easy because Google can reliably see through those kinds of tricks. The terms and keywords should be featured prominently in your text, but the priority should be that it makes sense and is appealing to readers, not search engines, otherwise it will appear forced.

Any text that links to another site – called anchor text – is also important. A link in source code looks like this:

If there are a lot of links on a site with the anchor text "Feminism", then it‘s a signal for search engines that this URL must be relevant to the search term "Feminism".

2. The right answer to the right question

If every subpage must have a clear theme in order to avoid duplicate content, it also means that every subpage can only be optimized for one search term, or for one collection of search terms that mean the same thing or serve the same intention.

For example, people who search "Feminism in the 70s" and "Women's movement 70s" are most likely looking for the same thing. A site about feminism in the 70s should therefore try not to optimize for all of those different search terms (e.g., three different URLs). It would actually hurt the site's SEO if they did that.

Search engines use user signals to measure if a site fulfills the user‘s search intent. User signals include: how far users scroll, if users follow links on the page to other sites, if users use interactive elements on a page, and many more.

Fast load times and readability are therefore relevant to a site's SEO as well. Search engines prefer sites that are fast, easy to navigate and whose content reaches the broadest variety of users.

3. More links to your site, more love

Besides user signals, search engines measure popularity based on how many external sites link to your site.

Every link is like a recommendation: because a person linked to your site, it means that they read the site and found it good or at a minimum found it relevant. But every link doesn‘t have the same value – if the site linking to you also has high-quality links on it, then reflects good on your site.

From Google‘s perspective, it‘s much better if a site like Wikipedia or BBC links to your site than if smaller blogs or spam sites link to your site. It‘s also better if the site linking to your site has a thematic relevance to your site. If the sites are too different, then it‘s possible that Google doesn‘t understand the connection between the two.

To make it more complicated, there are multiple types of links, which are all evaluated differently. An HTML link looks like this:

The user only sees the anchor text and clicks it to follow the link. One can also add another attribute to the link to manipulate how a search engine sees the link. A nofollow attribute means that a search engine should not follow the link. The link only counts towards the SEO of a website (positive and negative) if it does not carry a nofollow attribute, that means if it is a normal link without any "rel" attribute. They are also called dofollow links. Those are the good ones. In HTML, a nofollow link looks like this:

Generally, the more "good" links, the better ("good" and "bad" in reference to Google‘s definition). For instance, if a site is linked to from spam, it itself could also potentially be spam, thereby devaluing its value in Google's eyes. Therefore, the links aspect of SEO attempts to not only prove popularity, but the trustworthiness of a site.

Graphic of external links pointing to two different websites, one where the external websites don't relate to the topic of the linked website at all and some links are spam, and one where the links are high quality and make sense thematically

[[ Proceed with caution: Links can cancel each other out. If two websites link to each other, it doesn‘t help the SEO of either site (of course, those links might make sense for other reasons, this is simply the SEO perspective). This is one of the unfortunate side affects of SEO – it leads to unfair competition in the internet where sites are trying to be linked to as much as possible, but themselves link to very few other sites. In the left scene, however, websites in solidarity with one another could reach agreements. ]]

How often a site is linked to on social media (especially Facebook and Google+) also increases a site‘s popularity.

Links with problematic content: If you want to link to a site that you do not support, it should always have the nofollow attribute. This prevents the site from getting a higher search ranking from your link. Another option is to use the service Donotlink.it. This service generates a link that in addition to carrying the nofollow-attribute prevents the linked site from knowing where the clicks came from.

4. How effective is SEO?

SEO is not a straightforward process and the results aren‘t straightforward, either. There is no proven method that will result in 100% success. Sometimes the reasons behind your rank in search results is like a black box – you throw a bunch of stuff inside and it‘s unclear what‘s happening behind the scenes. For one thing, the ranking factors themselves aren‘t transparent. The rules of the game are also constantly in flux.

[[ To learn more about Google‘s updates to its algorithm: Overview of Google updates [english] ]]

Competition is another huge factor – if a lot of websites are competing for relevance to the same search term, it becomes a lot harder to move up in the rankings (and stay there). Hard fought search terms are ones that have the largest search volume, meaning that lots of people search them and that they are connected to specific interests, for exmample search terms that have a specific consumer intention, such as "Where can I find the best shoes“, which are relevant for businesses. What people search for is also constantly changing.

Newer websites have it considerably harder than others established ones. If it doesn‘t work to reach first place for a tough search term at once, it is no reason to worry. Don't make abrupt decisions and start changing your page; instead, stick to your strategy if you're confident that it is good. If the page is ranking poorly, but it is ranking, Google occassionally tests it on higher positions, and the tendency is that it is moving upwards, that is a good sign. Then you just need to have some patience and maybe update the content every once in a while slightly, so it is always up-to-date. If the page is not ranking at all over a course of two weeks, especially if it is not even indexed (that means, you can't find it if you search for the URL in Google), that indicates a problem. Another indication is when the page is constantly ranking poorly, not moving up at all, not even fort short tests, over a period of several weeks. Only then should you evaluate and revise your concept.