Whenever a visitor crosses from one domain to another, the information contained in the cookies must be passed in the query string to the second domain. Google Analytics documentation recommends using the provided method of _link().  This looks like:

<a href=”http://www.newsite.com” onclick=”pageTracker._link(‘http://www.newsite.com’);return false;”>Go to our sister site</a>

Following this link will cause the visitor to arrive on the page http://www.newsite.com/?__utmb=. . . .&__utmv=. . . .&_utmk=123456789&__utma=1.1234567.1234567. . .&utmz=. . .

With the values of those query parameters used by the GA code on the new site to create the necessary tracking cookies.

But when you have a very large number of links that cross from one domain to the other, it is difficult to find and modify every one, and to continue to maintain and update new links as they are created in the site.

To manage a situation like this you can use JavaScript to automatically determine which links are cross-domain links.  Once the script has determined which links are cross-domain links, it can create event listeners to wait for someone to click on that link.  When they click, the event listener will trigger the GA code that adds the cookie information to the query parameters.

There are several examples of scripts like this, and sometimes the script will need to be modified slightly to suit a particular web site or particular needs.

iQ Content published an autotracking script in this blog post:

http://www.iqcontent.com/blog/2009/08/google-analytics-automatic-tracking-across-multiple-domains/

An example of a script that I used for a particular client:

http://lunametrics.com/autoTrack-crossdomain.js

You can also find script examples in the books:

* Advanced Web Metrics by Brian Clifton (http://www.advanced-web-metrics.com/blog/about-the-book/)
* GA Short Cuts by Justin Cutroni (http://gashortcut.com/)

If you have an example of an automatic tracking script for GA you’d like to share, link to it in the comments below.

Looking at a website written in a foreign language is incredibly insightful. You may not get the immediate gratification of being able to easily find the location of a particular Prada shoe store, but you’ll find out a lot about yourself and how you use the site.

Let me preface this post by giving you all a little background about my recent activities. I recently went on a 10 day trip to Rome, Italy to visit my mother who has an apartment there for a month.

Rome, with its seemingly never-ending wealth of culture, beauty and delicious, delicious gelato and amazing boots kept me completely occupied about 16 hours of the day. But at the end of the day as it got dark and cold and damp, I would come home and indulge in a tiny cup of espresso and my internet addiction.

Though I was under strict orders from Robbin to avoid work at all costs, I still found myself dabbling around the web looking at where clients ranked in Google from Italy and paying attention to the differences in the types of results that came up when searching in Italy (for instance, the local 7 pack wasn’t as prevalent there even if I did searches for local shoe stores etc.) I must admit, I was also researching where to get fur/leather/boots and all the other things I’d made up my mind to purchase while in Rome.

I should also mention that I do not speak Italian at all. I took some Latin in high school, but that’s about as close to Italian as I’ve ever come. This profound lack of literacy made wandering around on Italian websites a somewhat confusing venture. It was only after three days of searching around for various things that I realized that what I was actually doing was usability testing.

Interesting Insights

It occurred to me that, in the nanoseconds of time that users give websites to prove their worth, the actual words on the page matter slightly less than the overall composition and clarity of the webpage. I realized what a unique experience it was to have to translate meaning from colors and shapes rather than actual words on the site.

Since all (yes all) of the Italian the shoe store sites I drooled over were flash sites I won’t use them for this usability blog post. However, I was on a few Italian travel sites trying to find good prices for tickets for inter-European travel (for funsies…I was just curious about how much Italians paid for their vacations) and quickly just how difficult it is to use the Italian versions of Orbitz, and Priceline. Then I stumbled upon www.edenviaggi.it and was immediately impressed.

The first thing that struck me about this site was its clarity. I felt that my eye could rest somewhere and, before I even knew what the words on the page meant, I felt like I could stop there and figure it out. It lacked the overwhelming info-packed clutter that plague so many other travel sites. The call to action is clear and was visible even on my tiny little netbook. The form was quick and easy to fill out and the resulting information was exactly what I wanted. It was the perfect research site, and I could immediately understand it without knowing a single word of Italian.

Translate Your Own Site

Now, if you really want to screw with your head, take a look at your own baby: the website whose content you crafted so carefully and whose design you agonized over. Go to translate.google.com and enter in the URL. Make sure you’re translating into a language you don’t know. Then take some time to click around your site, or a competitor’s site. Without text to guide you, can you figure out what to do? Is the website set up in such a way that you intuitively know how to get where you’re going?

It’s likely that without directions and prompts it’s going to be really hard to figure out where to go and what to do. Try checking out or filling out a contact form. Ask yourself how you could make that process more intuitive

In conclusion, I want to make a few clarifications. I’m not trying to imply that if you can’t get through your site in another language you (and your site) fail. The actual words on the site are integral to the user, and can’t be discounted.

I guess I’m just offering a bit of a unique perspective here. It’s sort of like how some famous artist (I wish I could remember who!) used to draw faces from the bottom up instead of the top down in order to free himself from the preconceptions inherent in doing something in an entrenched pattern. By removing one part of the website, you can gain insights into how you view other things and maybe some flaws will jump out at you.

I got this question the other day in my email (and I promise you, I could never make this up.)

I have a question.  Should you insert Google Analytics code in every single page if you want to track goals correctly?

What I mean is this:
My main opt in (squeeze page): /optin.html
My goal page: /thankyou.html

If I insert the GA code into EVERY web page of my domain, then GA is going to report my goal conversion inaccurately.

What I’m saying is this… Take a look at this scenario:

Visitor A visits /subfolder/subpage1.html
Visitor B visits /subfolder/subpage2.html
Visitor C visits /subfolder/subpage3.html
Visitor D visits /subfolder/subpage4.html
Visitor E visits /subfolder/subpage5.html

And all the above 5 pages are being tracked by GA as visits.

Then Visitor F visits /optin.html (which is my opt in/squeeze page),
and then he opts in and is directed to /thankyou.html

This means the conversion rate counted by GA is: 1/6 = 16.7%
(6 because subpage1.html, subpage2.html, subpage3.html, subpage4.html, subpage5.html, optin.html)

but the TRUE conversion rate is: 1/1 = 100%

Does this mean that in order to track my goals correctly, I should ONLY insert the GA code into my optin.html page and thankyou.html page?

So, dear questioner, this is actually a rather interesting problem, and is all about how you define “conversion rate.” Most people are interested, initially, in the conversion rate for an entire site, and then they start to segment (“How did our PPC visits do? How did the visits from Sydney convert?”)  But if I understand your question, you want to know, “How did people who landed on optin.html convert?” and you don’t want to see that polluted by visits that landed on other pages and converted. (Hence your use of the phrase “squeeze page.”)

So here is my advice:

1.  You should always code the entire site with your GA. Companies that pick and choose pages to code usually get inconsistencies in their data.

2.  There are numerous ways to capture the conversion rate by landing page. My favorite would be to create an advanced segment.  {Note for nitpickers, advanced segments are session-based and not page-based, but landing page is a ’session identifier.”}  To the left, you can see a screen shot of a landing-page advanced segment.

Notice, in this next screen shot (my favorite report, btw, traffic >all traffic > by medium > see goal tab) that the data make sense (at least to me, since I know this particular landing page well.) The majority of traffic comes by email. I was surprised to see how much comes organically, although it doesn’t convert as well as the email does (something I already knew.) The great thing about looking at this report, though, is not for me to learn about our landing page, but to answer the question: should I code all my pages? Yes.  Is there still a way to look at conversion rate by landing page by source etc? Yes.  (And just think, there are other ways, too!)

Robbin

HTML5 is the next generation of HTML, the markup language that the web is written in. Its features are slowly being adopted by browsers), and there are a few that have some ramifications for analytics.

<video> tag

Embedding video in your site these days usually involves a Flash-based player of some kind. Often this is an easily embeddable player from a site like YouTube or Vimeo. But what if you want to track when your visitors click the play button, or whether they watch the video all the way to the end?

Well, it’s certainly possible. Google Analytics offers Flash tracking, for example. So you have to take the video file, wrap it in a Flash-based player of your own devising (or maybe an open-source one like FlowPlayer), and implement some Event Tracking, for example. As you can imagine, even at its easiest, this is a lot more difficult than just pasting some code from YouTube to embed a video.

HTML5 offers a <video> tag, much like the familiar <img> tag. (There’s also an <audio> tag.) Instead of relying on a Flash-based player, it uses browser plugins to support native video files. And best of all, it supports a variety of events. You might be familiar with the “onclick” event attribute, for example. For a video, there are event attributes such as “onplay” and “onended”, so we can employ pretty simple JavaScript to track these events. No more Flash, just standards-based HTML and JavaScript.

Browsers are already starting to support the <video> tag, but not all of them are there yet. Over time support will become more widespread and tracking this sort of thing will be much easier.

rel=noreferrer

While the <video> tag represents a way that HTML5 can potentially make analytics easier, the second issue I want to cover potentially makes it more difficult.

We’re all familiar with links; they’re the currency of the web, after all. A link looks like this:

<a href="http://www.example.com">Some text</a>

In addition to the href attribute (which tells you where the link goes), there’s an optional attribute called rel, which tells you how the linked page is related to the page the link is on. There are a variety of possible values, but HTML5 introduces a new value that’s of special interest for analytics.

rel=noreferrer tells the link not to pass referrer information along to the destination page. So, for example, if a link from example.com went to your site, you would usually see “example.com / referral” as a source and medium in Google Analytics. However, if example.com used the rel=noreferrer attribute on the link, you wouldn’t get any referral information and the link would be counted as “direct / (none)”.

Is this a big deal? rel=noreferrer is really intended to protect private resources. For example, suppose the link was in someone’s gMail, or in an internal wiki, or something like that. There’s the possibility that passing along the URL where the link came from could reveal some private information or present a security risk, which is why rel=noreferrer was introduced. It does present a potential problem for accuracy in analytics, especially if it is used for reasons other than its intended purposes. But let’s remember that (1) analytics are never perfect, and (2) we should have a healthy respect for the privacy choices that people make on the web.

In any case, no browsers yet support rel=noreferrer, but you can expect it to emerge eventually and you should be aware of it.

There are a lot of ways one can make their website into a conversion driving machine. Site design tweaks, endless A/B and multivariate testing, exit surveys (gasp!) are some of the hardest to pull off. However, some of the most obvious usability issues often get ignored. There have been countless roundups of these issues, and a lot of them are really technical in nature.

I’m going to keep this one really simple and try to make these tips as actionable and specific as possible. That might mean that they don’t apply to a specific kind of site, but there’s something here for everyone.

Usability Issue #1: Browser Size and Your Call to Action

Just because you (or your web designer) has a 40 inch wrap-around monitor does not mean that your users do. Making sure that your primary call to action is included at the very top of the page. If you have a variable width website, great. If not, make sure that button/request/link/etc. is close to the top left hand corner. Everyone sees that corner. Additionally, frontload your keywords in your content. The sooner they see the words on the page that likely brought them there, the better.

Usability Issue #2: Where’s the form? You mean I have to do all that?

One of the most important types of conversion for most websites is data gathering. Therefore, one of the most popular soft conversion goals is to have the user fill out an information request form. Whether its demographic information that you’re after or email addresses for your newsletter, you want people to fill out this form.

To encourage users to fill out this form, webmasters are usually willing to jump through some hoops. Maybe they require a form to be filled out to access parts of the site or tantalize users with premium membership perks. However, everyone knows that requiring users to fill out forms is just another barrier surrounding the sale. Added to that, Most of the time, the form is buried more than three pages in the site and is hard to find, and a lot of the time, the form is lengthy and takes lots of time to fill out completely.

This is the point in time when priorities need to be set. What is the the MOST important piece of information that the website it supposed to garner through the form. If it’s the email address (and it usually is) put a simple, three line form on every page of the site.  First Name. Last Name. Email Address. That’s it. It’s visible and it’s so brief that filling it out won’t interrupt the user’s experience enough to prohibit them from doing it. If you want age, street address, phone number etc. by all means create that form and bury it wherever you want. But make an abridged version available as many places on the site as possible.

Usability Issue #3: Information Overload

Deciding how to present necessary information in a user-friendly way is a challenge for every website in every industry. It’s usually really hard for webmasters to decide what the most important information is and where to place it. Figuring out the overall purpose of the webpage in question is a good first step.

If the home page is supposed to introduce the company in question and showcase a specific product, limit the information to these two tasks. Keep in mind that your navigation bar will take up space and attention. Keep the copy as sparse and concise as possible and get to the point right away.

Avoid overloading every page with extraneous elements like rotating testimonials, banners and the like. There are specific places for such things, where they won’t detract from your focal elemetns.

Usability Issue #4: Sense of “Place”

Many times, especially with large, multilayer, thousand page sites, the user looses track of where they are and how to get back to something else they were looking at. This problem is not limited to huge sites though, it occurs on the smallest of blogs as well. There are many ways to create a sense of “place” for your user. In my opinion, the most overlooked tool is breadcrumbs.

Breadcrumbs (horizontal navigation links) are a great, noninvasive way to give customers a sense of place when they’re 8 levels deep on a specific product page or feature list. Breadcrumbs don’t clutter a page up, they don’t eat up a lot of real estate and they provide the added bonus of more anchor text rich links to other pages on your site (if you use them the right way.)

NewEgg.com uses breadcrumbs incredibly effectively, allowing me to navigate from a very specific product page to a more general category page quickly instead of having to find the category all over again in the side nav bar.

Usability Issue #5: Cross Browser Compatibility

If I had a dollar for every site I work on that has elements that work in Internet Explorer but not Firefox or Safari, I’d buy myself some over the knee Prada Boots. If I had a dollar for every site I come across (searching for boots?)  that works in Internet Explorer but not Firefox or Chrome, I’d probably be wealthy enough to start my own fashion footware company.

It’s a sad fact that browsers interpret code differently,but that doesn’t change things. Just because you see your site one way doesn’t mean everyone does, and if your navigation is broken or your CSS has text all over the place in one browser or another, you’re limiting your audience drastically and creating a bad name for yourself. Take the time to test the look and functionality of your site across all browsers.

Did you know you can make a coupon for your business through your Google Local Business Center? Not many people are aware of that, because the coupons have been nearly invisible to searchers since they were introduced back in August of 2006. Until now.

I present to you, Exhibit A:

I found this coupon in the wild while doing a search for “wedding photography Pittsburgh“. If you scroll over the coupon, it expands to show you more detail:

This little addition could have tremendous impact for local businesses. Just look how much the listing with the coupon stands out; the bright yellow definitely catches your eye. Of course, it’s only a matter of time until everyone jumps on this bandwagon, so it remains to be seen how Google will show the coupons when every listing has them. I can’t imagine them showing the same yellow coupon next to each listing – that would be cluttered and hard on the eyes.

Here are a couple more searches that I found interesting:

This is a search for “hotels Pittsburgh“. Notice how the listing in the top right stands out? It’s ranking at the #9 position (letter I), and it would be interesting to know if it would rank at all on the first page without the coupon.

Hey Pittsburgh attorneys, how would you like to get a few more phone calls from your Google Maps listing?:

That’s right, not a single coupon… yet.

Here’s another one, this Pittsburgh plumber has the idea:

All those dots, and he’s the only one with a bright yellow coupon.

And finally, a bit of a surprise. Usually locksmiths are all over this kind of thing, but that doesn’t seem to be the case (at least not in Pittsburgh – I’m busy right now doing a much more thorough search of cities and categories).

How long do you think it will take until more businesses start utilizing these coupons?

Today, we got a phone call from a potential customer, who had been recommended by another customer. I took one look at his site and was fairly certain that he had paid no attention to SEO, had no analytics, and had no calls to action.

“So,” he challenged me, “Why should I use your company?”  Well, I answered, maybe you shouldn’t use our company. I don’t recommend that everyone spend their money without having strong goals. Maybe you should start by telling me what your needs are.

The man on the other end of the phone said that all his company’s sales had been face-to-face or by referral up until now.  They had a little site, but did they really need to spend money? Did I know if people really found his type of services through the internet?  In fact, his needs weren’t so much SEO or PPC or GA or GWO.  He needed to be persuaded that those alphabet soup of internet services and tools were worthy of his budget dollars.

It is one thing to sell someone on the value of, say,  Google Analytics.  It is something else entirely to take someone who doesn’t have a strategic or emotional commitment to the work we do and convince them to spend their money there. I guess that’s a game I don’t want to play.  I feel like, you decide that you care on your own time, them come to us (or to another consulting company.) Consulting on the Internet is hard enough when your point of contact is committed (because s/he may have other priorities, may have to answer to his/her boss, may get moved around the organization.) Imagine how hard it would be if the person writing the checks really isn’t sure that the Internet matters.
Robbin

Whether this is a new decade or not, there is no question that it is a new year. Enter website copyright issues. So we have again my annual copyright post.

You can go here to read the original article, but in short:  “…the year corresponds to the date of creation of the material. Copyright law doesn’t have much to say about exactly how the date works. A range usually represents the idea that some material was created on X date, some on Y date, and some on dates in between. A single date is supposed to represent the idea that all of the material was created on that date.”  Remember, we aren’t lawyers.

So there you have some of the legal issues, but you know, there are other issues, too.  One of the things we learned this year (and always suspected) is that some people really do look at the copyright range to see how long you have been in business.  Bet you never thought your copyright was part of your conversion strategy, eh?

Robbin

“If I use keywords in the link title attribute, will that help me rank higher?”

I recently had a client ask this question. They were under the impression that using keywords in the link title would help in terms of SEO. Unfortunately, that’s not necessarily the case. There are several issues to consider, the big one being usability. Let’s take a look at this link title, shall we?

What is the link title attribute?

Put simply, the link title attribute gives additional information about the page being linked to. Here’s what the W3C has to say about it:

The title attribute is used to provide additional information to help clarify or further describe the purpose of a link. If the supplementary information provided through the title attribute is something the user should know before following the link, such as a warning, then it should be provided in the link text rather than in the title attribute.

When should I NOT use a link title?

You should not use a link title if you’re just repeating the anchor text. Not only does that not provide any additional information about that link and the page it goes to, but you could be annoying your visitors (if they’re anything like me). Have you ever scrolled over a link and the title pops up, covering surrounding text. Yeah, that’s what I mean by annoying.

When should I use a link title?

Use the title attribute for your links when you can provide additional information about that link and/or the page it goes to. For example, if your anchor text just says click here, give your users a better idea of what they’ll get if they click on the link (scroll over the “click here” link to see what I mean).

Of course, I’m not the first one to talk about this. If you want to find out more about the title attribute for links, here are some helpful resources:

How to Use Link TITLE Attribute Correctly

Supplementing Link Text with the Title Attribute

How to Customize Your Link Title Attribute

Do you use the title attribute on your links? If you do, let me know in the comments if you have any cool examples, or if you do it differently.

I was recently asked via Twitter (follow us @lunametrics) how to use Google Analytics to match landing pages with exact keywords sending traffic to those pages.

Simple enough, but first, let’s talk about why you would want to do this in the first place:

1. If you haven’t yet started optimizing a site, you can get an idea of which keywords the pages already get traffic from.

2. To check that the organic keywords sending traffic to a page are the keywords that you’re optimizing or link building for. It’s always good to measure your efforts.

To match your landing pages with the keywords used to get to those pages, it’s become much easier to do with the recent release of the new Google Analytics features – specifically Secondary Dimensions. Start off by going to the Traffic Sources report, and from there go to the Keywords report.

If you’re only interested in organic visits (which, for this example we are) then be sure to click non-paid.

Then, all you have to do is click the Secondary Dimension dropdown menu and choose Landing Page. It’s really that simple.

Tip: You’ll get a lot of branded keywords showing up at the top, mostly landing on the homepage. It helps to filter these out of your results using the keyword filter (Filter Keyword: excluding [insert branded term here]).

BONUS: For you analyst ninjas, the fun is just starting. Set it to show 500 rows and export your keyword/landing page report as a CSVfor Excel file. Now you can sort and organize the list any way you want.