Over the last few decades, the Internet has changed quite a bit. One of the most heavily changed aspects of the web have been meta tags. Today, some meta tags are essential to the proper functioning of your site. Others are indispensable for SEO. Many others are outdated and obsolete. Knowing which is which, and how to properly use the ones that remain, is important for a properly functioning website.
The meta title tag specifies what the title of your page will be. This title appears in the top of your web browser, anywhere the post is shared dynamically – such as Facebook or Google+ – and in any search engine. In general, you just need to remember to write your tags for human readers, but to optimize them with search engines in mind. Your title tag should be no more than 70 characters long, with keywords up front and brand names last. Always strive to make every title tag on every page different to avoid duplicate content penalties.
The description tag is the text in black that appears under your link in Google search results, as well as in other search engines. It’s also what appears when your posts are linked dynamically on Facebook, Google+ or other social media profiles that generate previews.
If you leave the meta description tag blank, Google and other preview generators will create a description for you. Generally, this description will be pulled from the first major bit of content on the page. Sometimes as an intro paragraph this is just fine. Other times it’s too long and rambling to make a good snippet. It’s generally a good idea to customize your description whenever possible.
Limit your description to around 150 characters. Google and other sites won’t display much more, though it’s not a hard limit. Use keywords, early and as often as you can without looking like you’re stuffing keywords. Above all else, make sure the description for every page is completely unique. Don’t forget to write your description with humans in mind; writing for the search engines just earns you a penalty.
Social Media Meta Attributes
As a quick note, you can customize your meta title, preview image and meta description on Facebook specifically. If you specify them with normal meta tags, they will appear the same in Google and Facebook. If you want your page to look different on Facebook, you can use the Facebook Open Graph attributes.
The meta property og:title, og:image and og:description tags are what you will use in these cases. The title can be customized to be similar but not identical to the title you otherwise use. The image will link to an image anywhere on your domain, and it doesn’t have to be an image in the article. Finally, the description can be unique and distinct from the meta description you use for Google search results.
Twitter Cards are a similar feature for Twitter. The meta attributes are twitter:site, twitter:creator, twitter:title, twitter:description and twitter:image. Site and Creator are typically both your @username, complete with the @, though if your site and your author have different handles, they can be specified individually. Title, description and image work the same way they do on Facebook.
You can also use Pinterest Rich Pins to customize images pinned to Pinterest.
Several years ago, Google was much worse at determining what the meaning behind any given post was online. Back then, the best way to get your posts indexed for certain keywords was to use the meta keywords attribute. In it, you could make a list of keywords nearly as long as you liked, separating them with commas, spaces or both.
This method of sorting pages quickly became prone to spam, and Google decided to remove its influence altogether. Today it is not used at all. If you have an old site that still has a meta keywords implementation, you may want to consider removing it. Google ignores it entirely, while Bing uses it as an indication that a site may be using old black hat techniques.
If you have meta keywords and they aren’t actively harming your site, you don’t need to rush to remove them. Simply remember they are obsolete and remove them when you next audit your site.
Meta Robots Directives
You can use robots directives to guide how Google and other scrupulous search engines crawl your site. This should be rarely used, because using a general robots.txt file works better. When in doubt, don’t use a robots attribute at all; it’s too easy to accidentally noindex an entire website and earn yourself a search ranking removal.
There are four attributes that can be used in conjunction in meta attributes.
• Index: Tells the robot to index the page. This is default behavior anyway, so this is redundant.
• NoIndex: Tells the robot not to index the page. Only use this on system pages, category pages and other pages that should not appear in search.
• Follow: Tells the robot to follow and pass PageRank to links on the page.
• NoFollow: Tells the robot not to follow the links on the page. Typically, the bot will still understand those links exist, but will not assign PageRank to pass from your page to the destination pages.
Index/NoIndex and Follow/NoFollow are mutually exclusive. You cannot tell a page to have both follow and nofollow, the code will be considered malformed and will be ignored, possibly resulting in a noindexed site.
Obsolete Meta Tags
In addition to keywords, there are a number of meta tags that are no longer used or are very rarely used. You don’t need to worry about any of them. These include, but are not limited to:
• Meta subject, to describe the overall subject of your site, like an overarching description.
• Meta copyright, which is typically assumed or posted in the website footer or about page in plaintext.
• Meta classification, which is typically valueless.
• Meta revisit, which instructs the search engine when it should return to reindex the page. This is replaced by robots directives and the fact that robots crawl fairly often as it is.
• Meta refreshes and redirects, which are better done with http codes rather than meta attributes.
Required Meta Character Set
A quick note; every web page, when created by anything other than a .txt file and hand coding, starts with a meta http-equiv tag labeling the content type as text/html and the character set as UTF-8. This is essential and should not be removed.