Search Engine Optimization (SEO) is an integral part of the design and development process. Getting the public side of your website to rank well in search results should be something you consider from the very beginning, and not an afterthought once you're getting ready to deploy.
There is no magic formula, and major search engines like Google keep the specifics of their ranking algorithm a secret to prevent excessive manipulation of the system. Furthermore, these factors change over time, so a strategy that was useful in 2013 may not be as useful today. Nonetheless, there are some general principles that most SEO experts agree upon as a baseline for getting a page to rank well.
A few things to keep in mind:
We'll go over the major things to keep in mind here, and talk about how this fits into the architecture of UserFrosting (where appropriate):
This is perhaps the most important thing to focus on if you want a page to rank. Modern search engines have sophisticated algorithms that go beyond keyword counting in determining how relevant a page is to a user. The three main things to keep in mind are:
Google can tell when you're copying content from another page. When it sees copied content, it decides that your page doesn't really contribute anything new to the list of results it will be showing to the user. Just like in the movies where a staffer presents three different manila envelopes to a president, each representing a different course of action, Google wants to give its users as rich a set of options as possible for their query. It wants to give them options A, B, and C - not options A, A, and A.
Having lots of words on a page gives Google more to work with when it determines how relevant that page is. While you may not need a lot of words to get your point across, it signals to search engines that your pages' contents have real value. At the same time, you need to avoid keyword stuffing - putting an unnatural number of keywords in your content in an attempt to game the ranking algorithm. This used to work in the past, but nowadays search engines will detect and penalize you if every other sentence is "Garden shed painting."
To summarize, you need to write the text of your pages so that they have a healthy balance of your target keywords without appearing overly spammy. You do know what your target keywords are, right?
As it turns out, where you put your content is also important. Search engines pay attention to your page title (as in, what goes in the
<title> tag) more than anything else. Your page title should not only say who you are, but what you are as well.
Mama's Kitchen | Authentic Korean barbecue and restaurant in Bloomington Indiana
Notice how we follow the name of the restaurant with a brief tagline that tells us what and where the place is. This lets us target people who are searching for things like "Authentic Korean restaurant", "Korean barbecue Bloomington Indiana", "Korean restaurant Bloomington", etc.
These same rules also apply to other "header" information on your page such as the
<meta name="description"> tag contents and the header tags (
If "content is king," speed is the queen. Pages that are slow to load will rarely rank well. This is especially critical in ranking on mobile devices, where bandwidth is limited and users have less patience. You can see what Google thinks about the relative speed of a page using their PageSpeed Insights tool.
While lots of factors influence page loading speeds, we'll go over the biggest problems that we see over and over in sites that are struggling to rank.
I can't even tell you how many times I've seen a 3MB logo image slapped hastily on a website. Don't use a massive 3000px x 1000px image as your logo, only to "scale" it to 300 by 100 using the
height properties of your
img tag. Your users' browsers will still end up downloading the 3000x1000 image with its massive file size, considerably slowing the time it takes to completely render the page.
Make your image files exactly the size at which you intend to display them (in some cases, you may want multiple different sizes for different devices) using an image editor. Use Photoshop or a web-based tool like Compressor.io to apply a lossless or reasonably lossy compression to further reduce the file size.
uf-bundle commands solves all three of these problems:
public/directory, so that they can be served directly by your webserver instead of going through the Slim application lifecycle.
Caching should happen on a number of levels throughout your application, on both the server and client sides. On the server, UserFrosting automatically caches route resolutions and fully-rendered Twig templates when you use the
production configuration profile. You can also configure the webserver itself to cache entire responses. For example, see nginx's caching documentation.
Cache-Control response header, which is the server's way of telling the client's browser how long they should cache the response of a particular request.
Setting the value of the
Cache-Control header is also typically handled directly by the webserver (though of course you could set this in your application codebase instead). The webserver configuration files provided with UserFrosting contain some common default directives for configuring this behavior on your server.
The main issue with client-side caching is that you need some way of forcing the browser to refresh cached assets when they have changed (for example, when the site developers add a feature, fix a bug, or change some content). This is called cache busting, and the most common approach is to simply change the name of the asset so that browsers assume that it is a new resource to be loaded.
Fortunately, UserFrosting's build tools take care of this as well. Each time you compile your assets, a random hash is used to name the compiled files. References to these assets in your pages are automatically updated to reflect these new names.
In the past, a common SEO strategy was to simply get as many links from other websites to your page as possible. This is because of Google's characteristic PageRank algorithm, which determines the importance of a page based on the number of pages that link to it, and their PageRank as well.
This algorithm was easy to game though, and "link farms" developed to artificially boost the ranking of pages with purchased and often irrelevant links. As a result, the relative importance of PageRank in search engine results has diminished over time.
Nowadays search engines still take links into account, but they are careful to exclude (or even penalize) "spammy" links. Furthermore, the quality of a link is more important than the quantity of links. Links to pages act like letters of recommendation, and Google is only going to take a "letter of recommendation" seriously if it comes from a reputable source and is contextually significant. In determining the relevance of a link, Google considers:
All in all, this means you should focus on getting good links to your site, rather than many links.
Page content, speed, and links are the Big Three in any search engine optimization strategy. There are many other factors as well, such as structured content, geographic factors, social media presence, user engagement, and security which we won't go into here. You should regularly monitor and record your site's search engine rankings, and don't be afraid to frequently experiment with your content to discover what improves your rank!
Finally, we recommend setting up Google Analytics and Google Search Console (formerly Webmaster Tools) to monitor your site's traffic and search ranking performance. You can set your Google Analytics identifier in your Sprinkle configuration file with the
site.analytics.google.code setting, and UserFrosting will automatically generate the tracking code for all of your pages.