Search engines judge how relevant your web page is to a searcher’s request by evaluating many criteria, but the very first thing they’ll encounter and subsequently evaluate is your page’s URL. It’s your first, best opportunity to establish why your page meets the needs of the person searching for it.

You might think that’s intuitively understood by web developers, but the internet is teaming with examples of database-generated URLs impossible to understand by humans and search engines alike. If your service provider generates these kinds of URLs by default and provides no means for you to craft better URL structure, find another service provider.

Consider your page’s URL an introduction to people and search engines. It instantly tells them what to expect when they view your page. Your URL should be clear, concise, and use plain human-readable language. Beyond this, there are a few additional ways to make a good URL a great URL. Here are 7 of them:

Use a few words to say something relevant.

In three to five words your should be able to give searchers and search engines an understanding of what content your page delivers. Be direct and as brief as possible. A phrase like “improve-your-search-engine-ranking” will work far better than “how-to-increase-your-web-page-search-results-with-google-and-bing” even though they may each be describing the same page. Crystallize the essence of your page’s content into a diamond of a URL.

Keep it short.

Try to keep URLS under 60 characters. Roughly speaking, the shorter the URL the better – assuming it’s clear enough to be understood. This even applies to BITly and other URL shorteners that allow customization of shortened links. Often, they generate links that have no intrinsic meaning. Make sure you modify the shorthand so it makes sense to readers.

Use “https” rather than “http”.

That extra “s” tells search engines and the rest of us that the connection to your page is a secure one, encrypted from start to finish so you’re more assured that visiting the page is safe, and you’ll be protected from data thieves.

Establishing trust up front is even more important on e-commerce sites, where financial transactions are conducted. Google has gone so far as to present Chrome users with a “NOT SECURE” warning for pages that don’t leverage the secure protocol. That’s not something you’ll want your customers to read as they enter the front door of your online store.

Omit “stop” words when possible.

Stop words include words like “the,” “in,” “an” and “or” which most search engines simply ignore. While that may not penalize you in search engine ranking directly, it does add unneeded characters that contribute weight to the URL’s length. Remember, you’re striving to be concise. The exception? If stop words are needed to make the content of the linked page clear to people, use them.

Keep directory structure simple.

Directory structure is indicated by the number of “/” symbols in a URL. In a sense, they’re a holdover from the days when content was held in directories and sub-directories (folders and sub-folders) on a server. While content today may be dynamically served in any number of ways, the “/” symbol still provides a sense of logical hierarchy and is still used. The key here is not to get carried away. Using many folders creates the impression that content is buried deep below layers of navigation and content, when really it’s just a click away. Better to keep it simple.

Use a keyword – or two. But not three.

Okay, I have to say it: don’t stuff your URL with keywords. Don’t stuff your content with keywords either! In each case, clear, concise and relevant words are the real keys to search engine optimization. Do include keywords in your URL, but never at the expense of readability or relevance. Don’t repeat keywords either – you only get credit for one instance, and you only keep credibility by limiting a keyword’s appearance in a URL to one instance.

Don’t hang out with “unsafe” characters.

As w3schools.com explains, URLs can only use the ASCII character set. Those URLs containing other characters will be converted into a valid ASCII format. The encoding process replaces unsafe ASCII characters with a “%” followed by two hexadecimal digits. Instead of spaces, use hyphens. Don’t use underscores, because Google prefers hyphens, and always use lowercase letters. It’s consistent and avoids “Page Not Found” errors.

So which characters are “unsafe”? Characters which can be misunderstood in the context of a URL string and which will always be encoded. They are the space character and the following characters:

" < > # % { } | \ ^ ~ [ ] `

Special characters are another set of characters that have special meaning in the context of a URL string, and they’ll be encoded if used outside of their intended roles:

$ & ' ( ) * + , / : ; = ? @