Today, Ann Smarty posted this blog at Search Engine Journal asking When (Not) To Use Target=”_Blank”. In all fairness Ann Smarty is totally awesome and really knows her stuff. In fact, she never actually endorses using target=”_blank” for external links, she was simply asking the question and opening the floor to debate on the issue. My reply simply was too long to make there comfortably.

So Should you open external links in a new window

No! The design decision to open external links in a new window is the wrong one for your users 99% of the time. There are a couple of key points to make on this, and I would welcome any thoughts anyone has on them.

User Reaction: Many users, myself included, have very strong negative reactions when a link opens a new window (or tab) without warning. I hate it and I’m not using that word lightly. I have never met or heard of anyone who has this same negative reaction when a link opens in the same window.

User Ability: Users have the ability to open a link in a new window or new tab as they desire already. This functionality has been built into every browser and is a normal function that users are already trained to do. Even better, many users can middle click the mouse and directly open links in a new window. There is no simple common browser function that will force a link to open in the same window if that is their preference (I could be wrong here). The key here is “user preference” not the website’s preference.

User Workflow: The internet has been around long enough that people have established a workflow for using it that works for them. Opening a new window or tab forces the user to change their workflow to accommodate the website owners’ narcissistic fear of losing page impressions. Making the choice to open a new window forces the user to locate a window/tab before they can continue with what they wanted to do. Nothing is gained for the user by this decision that the user can’t achieve on their own.

User Experience: Finally the last and most important argument against opening external links in new windows is that it breaks the established and expected norms for link behavior. This creates a negative experience for a large number of your users, while in turn enriching the experience for no one. There is simply no positive results from the decision to take the control away from your users and a great many negatives.

Exceptions and Best Practices

Like everything there are going to be exceptions to my position on opening new windows. In fact, Ann Smarty lists most of them in her post (just go read it already). I tend to prefer if .PDF files and other non-html files open a new window or tab. The key factor should be what will provide the most familiar and comfortable experience to the user and if at all possible defer to the users to make the decision for themselves.

For those times when you feel you absolutely must open a new window or tab it is best practices to make it very clear what you’re going to do. Just like you would announce that a link opens a .PDF so that those who would prefer to avoid them can do so, announcing when a new window will be opened is just common courtesy. Even users like myself who hate new windows will forgive the occasional one if it’s not a surprise.

My Response to Common Arguments

“I want to keep visitors on my site as long as possible to get more page views”. I have never understood this argument. Page views have no value, you might as well be raking up wompom points for all the good page views will do you. Why anyone would risk potentially running off a future customer or loyal reader just to artificially inflate a valueless stat is beyond me. Seriously take off the target=”_blank” from your site and email me. I’ll gladly set up a bot to give you as many page views as you want.

“I can’t vouch for the availability of external sites and don’t want visitors to follow a broken link.” This is from a comment on the post by Michael Martinez who is a search engine genius and because of that I actually stopped and considered the merits of the argument. After thinking on it I can see the merit if you’re linking to a lot of unreliable resources (why?). Hopefully though this isn’t an issue so often that this is necessary. If it is then I would strongly recommend taking a hard look at where you choose to link to.

“I don’t want interrupt the user from reading my content to check out my references.” I hear this argument most from websites that provide inline links inside of their articles and I can understand their position. No one wants to write 1000 words only to have readers jump ship and miss the last 700. The truth is that if the content isn’t compelling enough to make them want to come back, they weren’t staying anyhow.

A lot of SEO companies will stress paid search engine placement for your keywords, while we don’t discount that as a part of some SEO campaigns it won’t mean a lot if all of the traffic driven to your site doesn’t find the content they were looking for. That’s where knowing how to develop and implement your content comes in.

Of course, content strategies will vary based on the kind of site that you have. One thing that all content strategies should contain; however, is in -depth search engine optimization research. Every piece of content on your site should be crafted with your key words, competition, and potential customers in mind. While it takes a lot of work up front, it pays off with big SEO returns after launch.

A big no-no is simply regurgitating what you have found elsewhere. Not only do search engines frown upon this, but if your “borrowing” isn’t done following copyright laws with proper citation, you may find yourself in legal trouble. Keep in mind that if you create content (whether it’s blog posts, articles, sales text, or even just meta tag descriptions) that is unique to your site, uses your identified keywords (in moderation) and is relevant to your site and your customer’s needs, you will reap great SEO rewards.

It should also be noted that it is optional for search engines to follow the directive of the /roots.txt file. Although all major search engines tend to comply with the /robots.txt file there are plenty of spammy spiders that won’t. For that reason the ony sure fire way to protect content is to place password protection on the sensitive areas.

If you have something that you want to disallow search engine spiders access to, but sill allow unrestricted surfer access the best practice is to create a /robots.txt file in your root directory. That means the url to it should look like this:

http://www.yourwebsite.com/robots.txt

The following are examples of what should be contained in a /robots.txt file:

If you want to exclude ALL robots from your ENTIRE site:

User-agent: *
Disallow: /
If you want to exclude ALL robots from some content on your site:

User-agent: *
Disallow: /cgi-bin/
Disallow: /tmp/
Disallow: /junk/
If you want to disallow a single robot:

User-agent: EvilRobot
Disallow: /
IF you want to let ONE robot in:

User-agent: GoogleBot
Disallow:
User-agent: *
Disallow: /
The user-agent field is used to identify the bot you would like to allow/deny access. The * character is a wild card character and stands for “any”. Please note that if there is a part of your site that is private it should be password protected even if you use /robots.txt to disallow crawling of the page. /robots.txt is NOT enforceable and although legitimate spiders of search engines will follow its directives, not all bots are benign.

It’s also important to know that just because you have excluded content from being crawled, search engine results can still contain the urls to that content. This is because the spiders may have spotted links to your content from other sites with anchor text suggesting your uncrawled url may be good for that term. Below is a great video from Matt Cutts showing how that works.

Uncrawled URLs in Search Results – Matt Cutts

Best practice is to take those old URL’s, and all their traffic and inbound links, and redirect them to a new working URL. This results in a win/win for your users and your search engine rankings.

That is where the 301 server redirection comes in. Without getting technical a 301 redirect is a HTTP response from the server that means that the page has permanently moved to a new URL. Setting up 301 redirection is pretty straight forward and easy to manage.

301 Redirects and SEO

The permenant redirection indicated by a 301 error is honored by search engines. What this means is that when a search engine encounters a 301 redirect while crawling the web they will delist the old broken URL and update all the incomming links to the new URL.

Here is a short video by Mike Cutts from Google explaining how Google treats 301 redirects and anchor text.

301 Redirects Best Practices

Currently what we suggest to all our clients is that any page that gets search engine traffic or that has a significant number of inbound links get a 301 redirect to a working page with similar or the same content. This provides the best experience for surfers and maximizes the SEO value of their existing inbound links.

Any old pages that don’t get traffic or that have little to no inbound links should then be sent to a well designed and helpful 404 error page. Since there is little value in these pages to be lost the purpose of this method is to maximize the user experience for those who stumble upon one of these pages.

Another popular alternative is to redirect these lesser used pages to the home page. Doing this allows you to direct the inbound links and search engine rankings to a valuable page on the website.

The downside of this method is that potential customers may follow a link expecting to find something and instead find themselves on your home page. This isn’t the best user experience and may result in a lost customer.

Redirects on Apache Servers

The simplist and most common ways to set up 301 redirects is through the .htaccess file in the home directory of your web server. Unlike javascript redirects .htaccess redirects are invisible to the surfer because they are executed by the server when the page is delivered.

.htaccess Single Page Redirects

Single page redirects are best used when the content on the old page is still available at a new location. This way not only do you maintain all the inbound links and pagerank but visitors who arrive through external links find the information they were expecting.

When writing the redirect rule for a single page redirect you leave off the http://www and instead just include the path that would appear afterwords. For exampe to redirect from the old page of http://www.site.com/old/page.html to http://www.site.com/new/page.html you would use …

Redirect 301 old/page.html http://www.site.com/new/page.html

Other 301 Redirects

There are many other ways to set up 301 redirects besides using an .htaccess file. Steven Hargrove has a great resource that covers all of them here. If possible I highly recommend using .htaccess though because of how easy it is to set up and maintain.

I’m not going to bother writing yet another of those articles. Personally I HATE this tag.  Not because of the tag itself, its really nice and non-offensive.  No I hate it because of all the inane questions, bad information, and scammy tactics that are the results of this tag.  If your wanting some information on the current state of the tag here are 3 wonderful examples if your wondering.

Here is Matt Cutts from Google saying that “Google doesn’t use the keyword meta tag in web search”

What’s The Best Keyword Meta Tags Formula? by Shaun Anderson (hobo)
Even Yahoo, who supported the tag much longer than anyone else are dropping support (or are they?).

So with no search engines of note using the keyword meta tag to influence rankings anymore what should we do with it?  The easy answer is to forget all about it.  In large part that is what I have done on a lot of our sites and the sites we manage.  So far we haven’t noticed any negative effects at all.

But, there is one use that I will highly recommend for the keyword meta tag.  Misspellings.

Currently Yahoo and most likely the others as well are indexing the keyword meta tag as normal text with a really low ranking factor.  Because of this I would recommend using the keywords tag for common misspellings of your business name, product name, or anything else that you would rank for normally.  Since you wouldn’t want to use the misspellings on your page hiding them in the keyword tag allows the search engines to relate the misseplling to your page.

Either that or just forget about them.  Thats what I do.

As an internet marketing professional it is your responsibility to manage those expectations and to help guide the client through the process.  These are the 5 keys to managing the client’s SEO expectations.

Don’t Promise Anything You Can’t Guarantee:

New clients have seen the other guys making promises for #1 rankings and most will expect the same from you.  Even though it may make it easier to sell your services; never promise anything you can’t personally guarantee. Instead focus on what you can guarantee; that you will bust your ass for your client and that you will use your expertise to market their service to the best of your ability.

Set a Realistic Timeline:

Internet marketing and SEO are long term investments that (should) start before any code is ever written and (should) continue indefinitely.  Lots of times clients want to cut a check once and see immediate eternal results.  Therefore, it is important to set forth a realistic timeline with the client and for the project.  Explaining this upfront, and reminding as necessary will keep the client thinking in the same timeframe as you and this will go a long way to managing expectations.

Set and Track Goals:

Goal setting in internet marketing establishes the parameters for success and failure.  Which means that it is essential to work with the client to set realistic, achievable goals very early on.  Once set, goals should be regularly tracked so that you can identify what is working and make adjustments.

Back Up Everything With Hard Data:

Any time you are providing information to the client do your best to back it up with data.  This is almost the easiest of the 5 keys to actually do.  With the high quality free analytics packages available and SEO tools at our disposal getting hard data is easy.  Preparing and presenting that data in a way that clearly answers the client’s questions will go a long way toward building their trust and making them feel at ease.

Keep the Client in the Loop:

This is probably the most important of the 5. Since your clients are probably not SEO and internet marketing experts, they may be wary of spending money on something that they don’t understand and may not see immediate payback from. By making certain that they know what you are doing and where their money is going you are fostering trust. Be specific with how their dollars are being spent, and how that will transfer into profit for them.

If you do follow these 5 keys then the client will be able to relax, trust your expertise, and focus on making more money to pay you with.  This means that they will have less questions and concerns for you to address.  Freeing up more of your time to focus on marketing for them.  Best of all by following these 5 steps, when it comes time to negotiate for a new contract you will have built the trust, relationship, and documented success that leads to a bigger paycheck.

I should point out that this process only covers the planning, design, and development.  Contracts, creative briefs, and payment arrangements will all have been completed before the first stage of the project begins.

Our goal is always to provide the best experience and service possible for our clients.  For this reason we never begin a project until an agreement is in place and everyone is happy. Maybe we have been lucky but we have yet to have problems when we stick with this policy.

Planning Phase

Key Events:

• Set Goals
• Market Survey*
• SEO Research*
• User Research*

The work put in at this phase sets the foundation for the rest of the project.  First we work with the client to establish strong goals that the project will deliver upon.  Every action taken afterward is evaluated against how it furthers those goals.

We can also do a market survey, SEO research, and user research if requested by the client.  This research is used to determine best practices for the market, user expectations, keywords to target for search engine traffic, and much more.

Information Design

Key Events:

• User Scenarios
• Site Map
• Content Inventory
• Content Creation / Delivery

Despite all the pretty graphics and bells and whistles of the modern website, information is still king.  In the second phase we start deciding how all that information will be communicated to the user.

We do this by “mapping” out all the content that will be put into the website and then figuring out how to organize it in the easiest way for users and search engines to navigate through it all.  While I wish I could say that this is very technical; in truth, there is a lot of scratch paper and scribbling involved.

At the end of this phase we will have a solid understanding of what the users will need to accomplish their goal or find what they are seeking.  This is another reason why doing the market survey and user research is a good idea.  The better we understand your customers, the better we can plan for their needs.

Visual Design

Key Events

• Wireframe
• Photoshop Mockups

After we know how the website will be organized we sketch out some basic ideas on how to display that visually on the screen.  Key events and important page elements are designed first, with less important elements designed afterwords.  This ensures that the proper visual weight is applied to each screen element.

Prior to beginning the project we will have discussed colors, mood, and other design elements with the client while filling out the design brief.  Using that information we will present several design options for discussion.  This process can be very collaborative and can require multiple redraws, or it can be very much up to us.

Development

Key Events

• Programming
• Content Management

With an approved design in hand we start the process of transforming all the planning and designs into valid code.  During this stage we will also develop any custom applications or databases required for the project and set up any software on the server.  If a CMS, such as WordPress, is being used we will set that up as well.

Testing

Key Events

• Browser Testing
• Usability testing
• Code Validation
• Pre Launch Checkoff

Once we have completed all of the code, graphics, and everything else we upload the project and start testing.  Our testing process will ensure that the pages are valid, that they are usable, and that they load quickly and properly on all modern browsers.  We work very closely with the clients at this point to ensure that everything is looking and working as they requested.

Launch

Key Events

• Project Launched
• Post Launch Evaluation

When we have satisfied the client (and my own obsessive compulsive need to triple check everything) we finally get to launch the website.  Typically we include a 2 week warranty on every website (extended warranties and retainers are available) to ensure that our clients can sleep soundly knowing that if they need anything we are only a phone call or email away.

This is our design process in a nutshell.  I know I didn’t go into too much detail on exactly what goes into every phase but I do hope this helps answer any questions you have.  Please leave a comment below or use the contact form, we would love to hear from you.