>>TOM LEUNG: Welcome to the Google Trifecta Webinar.
This is Tom Leung, I am the Product Manager for Website Optimizer.
I'm very excited to be here with Michael Wyszomierski from the Webmaster Tools Team, as well as
Jeff Gillis from Google Analytics. This is the first time we've done a joint
webinar, and we think it's going to be pretty successful
because these are probably the three most important tools any website owner should consider
using from Google. What we're going to cover today, specifically,
is I'm going to talk for just a couple minutes about how we see the three tools complementing
each other, and then we're going to spend most of our
time digging into a nice introduction into each of the tools--the Webmaster Tools side,
Analytics and Optimizer-- and then we will take a couple of Q&As from
the registrations for questions that we saw coming up over and over.
So, without further adieu, when you think about these tools, you can think about using
them in sequence. Specifically, you probably want to start with
Webmaster Tools, because as Mike's going to tell you,
it's going to really make sure you maximize the exposure of your site to the search engine
so that Google can understand what's going on and what's on your site.
The next thing you do, once you maximize exposure, is to really start understanding,
"Okay, well I'm getting a lot of exposure; now let's see what's going on on my site--what
do people like? What do people don't like? Where are they going?"
And Jeff is going to go into great detail about that.
Now after you do that--after you get a lot of exposure, a lot of people coming to your
site, and then you start understanding what's working
and what's not working and it's time to make improvements, that's where Website Optimizer
comes in, and that helps you test different changes
to your site and see which ones are most likely to be well received by the visitors.
Another way to look at it is that each of these three tools can be used individually,
but they also make the other two even more effective.
So with Webmaster Tools, you're really going to make sure that as many people are able
to get to your site as possible, so-- because Google will understand all of the
content on your site and know what's going on--and that's going to allow more people
to find it. That is--the more people that arrive on your
site and the more things they do will give Google Analytics more data to show you--
interesting trends, interesting problem areas, interesting metrics--and that increased data
and insight into what's going on on your site, is going to allow you to get more conversions;
because then you're going to be able to really focus in like a laser and say,
"Well, I know from Analytics--you know--these are certain problem areas; I'm going to run
an experiment on that specific page." And so you--as you can see, then when you
get more conversions, those people will return more often,
greater word of mouth, more likely for other folks to link to you, and it kind of is a
virtuous circle. So specifically, I'm about to hand it over
to Mike, and what we're going to do for all three of the products is we're going to try
to answer three basic questions. Like, what is the big problem Google is trying
to solve with this product? What are the customers' benefits and features
for the product? And some specific things that you can do after
this Webinar, to really make use of it and resources you can consult to learn more about
these things. So I'm going to hand this over to Mike, and
he's going to tell you a little bit about Webmaster Tools in more detail.
>>MIKE WYSZOMIERSKI: All right, thanks, Tom. And I'm going to start off by explaining why
it's important for us to have Webmaster Tools in the first place.
And it really goes back to Google's mission, which is to organize the world's information,
make it universally accessible and useful. So in Web Search, we strive to provide the
highest quality and most relevant results for any query.
And these results come from our index of your pages.
So how can we ensure that we're presenting the highest quality pages to users?
Well, first you need to have good information and content, which will lead to better pages.
And these better pages are indexed by Google, which gives us better results for our users.
And then if we can better understand the content of your sites, the search results get even
better. So in order to make this information universally
accessible to our users, at Google, we actually rely on you--the webmasters--quite
a bit to make this content accessible to Google, and that's where Webmaster Tools comes in.
And the features of Webmaster Tools all relate to how your site is presented in Google's
organic, non-paid search results. So keep in mind that this tool isn't really
designed to help boost your rankings; that's more determined by your site's content.
However, it can help you with giving Google a better understanding of how your content
is presented to Google, so it can better understand it and present it to users.
So Webmaster Tools provides statistical and diagnostic information, as well as some control.
And I'll cover three main functional areas of Webmaster Tools today: diagnostics, statistics
and control. I'd like to start out with diagnostics, because
this is where you can find out if Google is even able to access your content in the first
place. So in the "Web crawl" section, you can get
a listing of errors that Google encountered as it attempted to download pages from your
site as part of our indexing process. And this is the part where Google reads your
pages to find out what information is on them. But sometimes Google has trouble getting in.
Maybe your webmaster, at some point, had blocked search engine robots from accessing your site
during development, and then upon launch forgot to remove that restriction.
That could prevent Google from getting enough information about your site to present it
to users. Content analysis goes a bit beyond basic access
and evaluates how well Google can understand the content of your site.
There are a few different ways you can describe your pages to Google and using--including
using title tags and meta description tags. And both these tags may be shown in the site's
listing in Google. And since that's what users see before deciding
if they want to click through to one of your pages or someone else's, it's probably something
you want to pay attention to. So let's take a look at a search result to
better illustrate what Google users--your potential visitors--see before deciding whether
or not to click. One way you can get them to look at your site's
results is by typing a site colon, followed by your site's URL, with no space in between,
and this will show a listing of pages from your site.
In this case, we're looking at pages from google.com/webmasters, which is our Webmaster
Central Portal. Looking at the first result here, you can
see that the title is Google Webmaster Central. We often use the same title which you provided
to us in the title tag of your page. And in this case, I think that Google Webmaster
Central is a good title for this page, because it describes to the user exactly what they're
getting. It would be less useful, for example, if all
of Google's pages simply had the title, "Google." If several of them were listed on a search
results page, it would be hard for the user to find what he or she wants.
I just don't go overboard with titles, because if you had too many words, it can be just
as confusing to the user. Below the title outlined in red is a short
description of the site, which here at Google we call a "snippet."
We generate these snippets from a variety of sources, including your page's content.
And one possible source of this snippet is your meta description tag.
And this allows you to give us input about how your page should be described.
Mostly the results here were generated without meta descriptions.
It looks like we did okay with the first one, again, outlined in red.
But if you look at the description for Google Gadgets, there's less useful information such
as navigation text and the phrase, "Terms of service."
And that page could probably benefit from the addition of a meta description tag.
Like titles, your meta descriptions are likely to be more helpful if they're unique to each
page, so the user can quickly determine which page is the one they want to click on.
The Content Analysis tool can bring issues about this meta data to your attention as
well as pointing out which content may be difficult for Google to index.
Here you can see that the Google Webmaster Central site has six pages with the same title
tag. Clicking on the duplicate title tags would
give you a listing of these pages, so you know which ones to change if you feel it's
necessary. Now since you're all here to also learn about
Analytics and Website Optimizer, I'm sure that you're very aware of the value of good
data. In Webmaster Tools, we provide a variety of
statistics to help you understand how Google users are finding your site.
But unlike tools which only analyze visitors that you already have, Webmaster Tools may
alert you to potential visitors by showing where you are in our results,
even if the user doesn't click through to your page.
One interesting application of this may be to look at your top search queries, in Webmaster
Tools, to see which terms you're appearing for,
and then check in Analytics to see if you're getting visitors for these queries.
You can also use the "What Googlebot sees" feature to see what other sites--possibly
your customers or your partners--are saying when they link to you,
and it can see what you're saying about yourself by evaluating which key words you're using
on your own pages. I mentioned links briefly on the slide before,
and we like to provide a lot of data about links since they can show how your site is
connected to the rest of the web. You can evaluate which of your pages are getting
links from other sites, and which of these other--and what these other sites are.
And again, comparing this data to Analytics--which is where your referrals are coming from--can
give you a better picture of how people find you out on the web.
The internal links data can give you an idea of how your pages are connected to each other.
They can also help out for very practical applications, such as learning where you have
broken links if you delete or rename a page. The third function of Webmaster Tools that
I'm going to talk about is control. Here are some examples of actions you can
take directly with Webmaster Tools. It can let Google know that if your site--or
a section of your site--is relevant to a specific geographic location,
so pay attention to this if you're running an international site.
It can remove outdated or private information from our results, can control Google's access
to your site by generating the robots.txt file,
it can monitor and control Google's crawl rate of your site, and it can even let Google
know about all of your URLs by submitting a site map.
This is what the Tools page looks like, and it's where you can take care of most of those
actions I just mentioned. So that's an overview of what you can do with
Webmaster Tools to help Google better understand and present your content.
And to get you started, here are three actions you can take this week:
First, verify that you own your site in Webmaster Tools.
This only takes about a minute and it will give you full access to all the features I
just talked about. It also helps us to notify you in cases where
we need to contact the site owner. Next, check for crawling errors.
If we can't get to your content, we'll have a pretty hard time indexing it and showing
it in our results. And finally, take a look at your titles and
snippets in Google and use the Content Analysis tool to see where they can be improved.
This may be your first and last chance to get a user to click on your site.
Next, I'm going to answer a couple of questions which were sent to us related to Webmaster
Tools. Does the link data on Webmaster Tools represent
all the link data that Google has? That's a good question, and the short answer
is not necessarily. One way you can get a sampling of links to
any site is to do a link colon search in Google. So that means that in the search box, you
type link: followed by the site, with no space in between.
This will give you an idea of what kind of sites are linking to it.
But if you verify ownership of your own site in Google Webmaster Tools, you can actually
see a much larger sampling of links to your site, down to the page level.
You can do this by clicking on the "Links" tab in the sidebar and then clicking on "Pages
with external links." However, even though it's a larger sampling,
the list may not contain all of the links which Google knows about.
What constitutes duplicate content? This is a very common concern among webmasters
once they start thinking about search, so I'm glad someone asked this.
Duplicate content refers to the same or substantially similar content across multiple pages.
At the most basic level, this means that if you have two or more distinct URLs which serve
the same page, then that's duplicate content. There are plenty of natural causes of duplicate
content across the web, including things such as tracking parameters in URLs, which would
change the URL, but not the content. And at Google, this is something we're well
aware of and try to handle properly. Duplicate content can be a problem for search
engines, because if we don't detect it, the results may not be as diverse as we'd like.
We wouldn't want to serve a list of results where every page had exactly the same content.
So what we try to do is detect which URLs are the same and then group them together.
We then show users what we feel is the best URL to represent that content.
To help us out in reducing duplicate content and to also help consolidate the links to
duplicate content on your site-- which may affect your ranking--we have multiple
blog posts on the subject that you can find on the Google Webmaster Central blog.
There's a link to it from Webmaster Central and I'll give you the link to that in a minute.
Getting back to Webmaster Tools specifically, there's one very common cause of duplicate
content. Many sites serve the same pages for domain.com
as well as www.domain.com. And since www is technically a subdomain,
and could have different content, it may be considered as a different site or page than
the other version. And we actually have a function dedicated
just to this in Webmaster Tools, called "Set preferred domain."
And here you can tell Google which URL you prefer if you serve the same content on www.yoursite.com
and yoursite.com. In addition to this setting, we also recommend
that you set up a 301 redirect on your server to the version of the URL which you prefer.
This will give another signal to Google and also help out other search engines which wouldn't
have access to the data in Webmaster Tools. So I'll end with a few helpful resources,
and I hope that you're all excited about trying Google Webmaster Tools if you haven't already.
So all of these resources are available at Google.com/webmasters, which is our Webmaster
Central Portal. We have the Help Center, containing comprehensive
articles about your site and Google. Our blog is used for feature announcements
as well as tips for improving Google's understanding of your site.
And finally, we have an interactive discussion group where you can ask specific questions
about your site and get answers from other webmasters interested in search.
It's also monitored by Googlers, including myself, who collect feedback and answer questions.
Now once you have your site indexed by Google and people are finding you, you probably want
to get some more information about these visitors-- what do they search for? Where are they coming
from? Are you getting visitors from other search engines or sites?
And to answer these questions and more, I'm going to throw it off to Jeff, who is going
to talk about Analytics.
>>JEFF GILLIS: Thanks a lot, Mike. Great, so now we'll move into Google Analytics.
So first of all, a little Google Analytics overview.
Google Analytics is a free hosted web analytics tool.
It helps website owners understand how visitors find and navigate their site, and also what
types of visitors they are and we'll get into all that in more detail a little later.
And it also shows all traffic from all sites--not just Google--which is a bit of a differentiation
point between Google Analytics and Webmaster Tools.
And I want to touch upon a couple other differentiation points, as well as some commonalities.
So Webmaster Tools--as Mike said--it's an indispensible SEO tool for Google.com searches,
including things, like he said, search query impressions and how those relate to actual
visits from those search query impressions. And how does Google Analytics compare to that
particular bullet point? Well, Google Analytics is an SEM and SEO tool
for all traffic, plus website content evaluation, which is kind of where it ties between Webmaster
Tools and Website Optimizer. So you can see the queries that then lead
to conversions on your site. Where Webmaster Tools talks about geo-targeting
of certain websites where you can try and target, say Japan, for that website and those
Google searches-- with Google Analytics, you can then go on
to see actual geolocation of the people who came and visited your site, to see if that
had any affect. Where Webmaster Tools offers a URL removal
tool, if you don't want a certain section of your site to appear in Google search results,
Google Analytics will show you the popularity of certain content and folders and directories
on your site, and then navigation between that as well.
And so I'll get into all that stuff a little later.
Webmaster Tools also offers some other things such as Google robots reporting--which Google
Analytics doesn't offer-- and information about how your site is optimized
for mobile searches--which Google Analytics doesn't offer currently--
but Google Analytics offers a host of things which I'll just run through here--this is
the most text heavy side--but I wanted to go through all the benefits.
So first of all, with Google Analytics, you can understand how visitors engage with your
site, and it's almost any type of content can be tracked on Google--
from rich content to flash to outbound links and more.
You can also compare performance of all your marketing efforts, such as emails, keywords
or even offline campaigns. We have integrations where you can do a radio
ad and then see how that affected your website. Integrated with AdWords, but also tracks non-Google
campaigns such as Yahoo Search Marketing and adCenter in Microsoft,
it helps you identify areas for improvement on your site.
You can pinpoint breakdowns in conversion process use, which is very important--we'll
talk about that--it's setting up goals. And then also, lastly, and a little bit more
sophisticated, you can actually track e-commerce metrics and get a host of reporting on that,
and really do an in-depth ROI analysis on keywords and then how those relate to product
purchase. So for what type of websites is Google Analytics
appropriate? And the reason Google bought Urchin in the
beginning and turned it into Google Analytics, is because it really works with all types
of websites-- from e-commerce, retail websites--you can
see below each column the types of metrics that people who own those websites might be
interested in within Google Analytics-- lead generation, where you're possibly having
people sign up for a newsletter, trying to figure out conversions and analyze visitor
behavior, and then also branding, where you might be
more interested in a metric-like time on site--did they look at a movie that we have or are they
just spending time on our site? So first, I wanted to ask, what is your website?
So taking a step back, for many people, it's--at first, it's a webpage.
And then as you become a more sophisticated business owner or you're getting more visitors,
it becomes more than just your server and you've got a website--"Look, mom, my website
is up, here's the URL--" then you become a Google search result and
you say, "Oh, people are actually coming to my site; well, how do they find it?"
Well, it's on Google. Google's crawling it. And then, even further, now you're going to
start to possible try to get traffic directed--targeted to your site by buying key words.
So you become your statistics--CPC, CTR--maybe your website sales.
But what more? Well, usually nothing more. That's where people
leave it. And so when you're ready to look, there is
a ton more that your website is, and that you can actually use to get more visitors
and understand your audience. Without Google Analytics--I love this picture
and these characters--so that's obviously Professor Xavier, he can read minds--it's
a movie character--and that's a dog--his dog. Without Google Analytics, there's no communication--he
can't read your mind and you can't hear his thoughts either.
And--but, with Google Analytics the dialogue begins, so let's dive in and look at some
examples. I own a web design and hosting company.
You can see the web page; it's a beautiful, really polished web page.
We design web pages for you and we can host them as well.
Google Analytics says, "We've got some interesting news for the marketing team--1,000 visitors
came to the site this year--from another website, sitepoint.com, and converted."
Well that's really interesting to know. And you can see in this report, which is called
"Referring Sites," SitePoint is the number three largest referring URL.
It's bringing in over 1,000 visitors under this--in this time period.
What is SitePoint? Well, let's take a look at the website.
Here it is, you can see on the right, there's an ad.
Now for some reason, SitePoint is linking to your ad--your web design company--your
URL. So why not throw an ad up there, if possible?
And you can do placement targeting with a lot of the PPC products out there today, so
put some ads on SitePoint, it says on the right.
All right, let's move on. We redesigned our home page in September,
it's beautiful. What affect did it have? With Google Analytics, you can see after the
home page redesign, percentage of returning visitors went up 60%, so great job.
Here's a Returning--New vs. Returning Report within Google Analytics and you can see the
blue--that spike was the current time frame, compared to the green line,
which is the previous time frame. You can see down in the data table, the percentage
change in returning visitors, so that's a great indicator and just some metrics you
can see. The dog says, "Nice webmaster work."
All right, that's a sizeable investment. I want to decrease CPC keyword costs for my
site, which is a volunteering opportunity portal, so I can--people come to our site
and they look for volunteering opportunities. Well, people are coming to your site on the
search term "volunteer abroad" and then bouncing--you don't offer abroad programs, just domestic.
Let's take a look at the report. Here we go, we can see bounce rate--this is
a visualization of comparison to site average, so some are green, some are red--these are
your key words on the left-- and you can see "volunteer work abroad--"
just like the guy said in Google Analytics--is getting a high bounce rate.
So people are coming and then leaving right away from that URL--that's what bounce rate
means--they come, look at one page and then take off.
They don't do anything--they don't look for more opportunities or sign up for anything.
So, we're going to want to get rid of that keyword or do something with it--
why not add "abroad" as a negative keyword to your PPC campaign so you're not paying
for those--or not showing up for those searches? I want to expand my business.
You want to expand your business? This dog can really read minds!
Well, you're getting lots of visits from New York and France.
And let's take a look within Google Analytics--this is called the Map Overlay Report.
And you can see the little bubbles are kind of a heat map--they show what cities across
the world that you've gotten--that you receive visitors from.
If you hover over a bubble, it shows you the number of visitors.
The bigger the bubble, the more visitors. And there's more data on this in a data table
as well, just numerically kind of spelled out--this same information.
But this is a great visualization to show customers if you have--if you're representing
clients, if you're an agency to say, "Here's where your visitors are coming from."
As you can see, there was a big bubble over New York, and also there were visitors coming
from France, as the dog said. So why not create a location-targeted campaign
and create New York-targeted key words and ad text.
You know--you're really trying to grow your business--you're getting visitors from there--why
not start a campaign like that? Also, perhaps offer a French version of the
site if you can translate it. Couple--one more example--I need to check
my laundry. Your laundry's done. I mean, if you're a really sophisticated user,
maybe this would work--just kidding! Moving on, let's dive into the product now
and take and look--I wanted to show some of the features that everybody could use today.
First of all, here's the dash board. This is what you're going to see when you
go into Google Analytics. There's a lot of reports and so I just wanted
to show you--in a simple way--this is what's offered.
So you've got the Visitors Report--which I'll show--Traffic Sources--where people are coming
from, Content, which is what they're doing on your
site, and then a Goals and E-Commerce section where you can actually configure it and tell
us what's valuable to you. Here are all of the reports maximized, just
to show you how to navigate. Where to start? I would just start at the
Overview of the Visitors section--the Traffic Sources and the Content section.
Here's the Visitor's Overview. As you can see, before I asked what is your
site? Well, here's what your site is--it's not just
CPC and CGR--it's also visits, page views, unique visits--which is people who--
it's not counting people twice who came twice in a certain period--it just counts them once--time
on site and more. Here's the Traffic Sources stuff--this is
all Traffic Sources, so it's showing how are people getting to your website?
What are the most popular ways they're getting there? What sites are directing traffic to
you? And then you've got a--this is the Navigation
Summary. This is the in the Content section.
This shows a certain webpage and where people came to get to that webpage, and how they're
kind of surfing through your site. Let's look some more at the Content stuff.
This is a--this is a list, again, of keywords and this was the keywords that actually brought
people to a specific piece of content on your site.
So I wanted to show this to you to show you that you can actually slice and dice information,
so that you can see one metric as it relates to another metric.
So sometimes we have a very important site, are keywords actually bringing them to that
site or not? Again, this is what your site is--it's the
keywords that bring people to your page. It's not just, again, CPC, CTR or just Google-centered
keywords, these are keywords from all search engines.
Great, so here two keys to Google Analytics: First of all, look at data in context.
We--and then secondly, set up goals. So the reason I put this guy on here, again,
is he's the guy who reads minds. And we can't read your mind unless you kind
of help us out a little bit, and we've given you tools to do that.
First of all, data in context. Data isn't an island. It should be read in
context, and that's in order to give it meaning. And so this is a technique for you to get
more out of Google Analytics, and I'll show you how to do that.
Context insight for the right questions--so here's a graph of keywords and basically the
popularity of keywords. Underneath you can see search sent one million
non-paid visits via 300 thousand keywords. Well that's great, but is it better than before,
is it good? I don't know what it means--it seems like
a lot of keyword activity, but I'm not really sure.
So what you can do is you can click this "Visits" button and actually compare it to another
metric that's important to you, such as--as you can see in here, there's a
whole list--why don't we compare it to conversions, so you can compare two metrics, and we'll
do conversion rate. So it seems like the keyword activity is going
up and to the right a little bit, but when you compare it to this other metric,
conversions, that's actually going straight or maybe even a little bit down, so that's
something that we should be aware of. We're paying the same amount in keywords,
our conversions might be tightening up a little bit.
Another way to get context out of data is compare date ranges.
This is a really easy and visual feature within Google Analytics.
So you just drop down any--you just click on the date and then you'll see you can compare
two date ranges by clicking that drop down menu right there.
And then you'll see that in this graph up above, there's a blue line and green line,
and they're plotting the same data, but different time periods.
You'll see that, interestingly, July 2, we have a spike in conversions in a previous
time period. So what did we do to get that spike?
You can also see that there's like red and green next to the data in the middle there,
underneath the "Visitors Completed" text, and that's showing whether we've gone up or
down in this time frame. And so it can show you what the areas that
you need to concentrate on are. So, again, context--you can see the question
mark area is without a comparison to previous timeframe, whereas the one with the exclamation
point is showing us context. Here is a Content by Title--and this is the
last area of context I wanted to show. There's some pretty mature visualizations
within Google Analytics that can show you real actionable areas.
So when you click on this "Views" area, this one is probably the best.
It's called "Compare to site average," and as you can see right away, what's in red and
what's in green. These are articles by our friend, Alvin (S/L
Ashcowsheek) on his blog, and you can see which ones got the most page views,
sorted by "Page views," if you look above the column on the right, there, and then it's
sliced by "Time on page." I can see they got a lot of page views, but
are people actually sticking around and reading these posts? That's what I really want to
know. Another area for this visualization that's
really helpful is in optimizing keywords and campaigns, under the "Traffic" section.
So you can see here, you've got a list of keywords and you can immediately see which
ones have a high average time on site or a low average time on site.
So just to race through this, we've got a couple of visualizations, but the best one
is the one on the lower right, which is the "Compare to site average."
Lastly, Goals--take your data to the next level of usefulness.
So this you can see--this guy--there's a purchase page and we say, "Thanks for your purchase!"
And so that's the goal page. And what you do is, you put that in Google
Analytics--very easy, just copy and paste that URL into Google Analytics--
and then you can also--that's up at the top where the circle is--and then you can also
tell us the steps to get to that page, such as, enter your name, view your shopping cart.
And what Google Analytics will do, is we'll populate this beautiful report called Defined
Funnel Navigation, and so you can see at the bottom there, we have completed order.
Great, okay, that's my conversion page--that in itself is--will populate a lot of data
within Google Analytics to show you what keywords are converting and things like that.
But if you have a funnel, though, the top will be populated--funnel one step, funnel
step two--where are people coming in and then going to the next step in the funnel,
and which funnel steps are causing people to leave?
And so this--if you do a little bit of optimization, it can save you a lot of money and customers.
Some funnel step pages might be too confusing. You can just make minor changes and this is
an area where you can see that. And Tom will talk a little about this with
Website Optimizer as well. And then just to finish up, this is, again,
the Compare to Site Average--if you have a goal configured,
you can see that you can actually use the Compare to Site Average to sort by goal on
the right. So before we did time on site and bounce rate,
but now we're looking at our keywords sorted by popularity, and then segmented by which
ones had conversions. So it's more than just CTR and CPC, it's actually
conversions. You might be getting a lot of traffic from
one keyword and paying a lot, but not having any conversions.
Lastly, for best results, as I said, get started with the Visitor's Traffic and Content Overviews--those
are the reports to start with-- and then view data in context--we had those
looking at--comparing the site average, looking at those visualizations and looking at data
over two time periods as well. And then lastly, creating goals is probably
the most powerful and very easy thing that you can do within Google Analytics,
to figure out what your valuable key words and content pages and traffic sources are.
And then take action on those things. And speaking of which, I'm going to turn it
over to Tom and Website Optimizer to talk about how to take action.
>>TOM LEUNG: Thanks very much, Jeff. So let's make believe that you've set up Webmaster
Tools, Google sees all of the content on the web that you offer,
and then you use Analytics and you see all these wonderful insights from Anayltics.
So the next question would logically be, "Well, what should I do about it?"
And that's where Website Optimizer comes in. And specifically, you can look at a bunch
of reports and get a lot of insights and a lot of understanding about what's happening
and what has happened. The next step would be to say, "Well, as a
result of that," let's say I'm looking at a Funnel report, like Jeff was just showing
us. If I find that a large number of people are
not proceeding after the first step of a funnel and they're kind of dropping off,
well, how do I increase the rate at which people will move on and continue the process,
and ultimately complete whatever action it is that we're looking for them to complete?
Be it purchasing something, filling out a lead generation form, watching a video, reading
a white paper--you know--signing up for a newsletter--it really could be anything.
So in the old days, there are a number of ways that you "improve" a page.
One, is you kind of go by your gut feel and you just say to yourself, "Oh, wouldn't it
be cool if we made the button a little bit bigger and moved it on the left?
I think--you know--that would be better for users," or "Wouldn't it be cool if I, instead
of having one image, I had an image kind of rotate between a number
of different images?" That's one way to do it, but it's often wrong,
because you don't really know if it's going to work or not,
sometimes it helps, sometimes it doesn't make any difference and sometimes it actually hurts.
We'll talk about how Website Optimizer shines a bright light on these kinds of decisions.
The second way a lot of people make decisions is with, what Avinash Kaushik--one of our
colleagues--calls the HiPPO--the highest paid person in the organization.
Now this is usually the head of your design department, or the CEO or owner of the company--
and he'll just--you know--walk over and say, "Hey, I was looking at the website and I'd
like us to add this chart and I think that's what the customers want,"
or--you know--the hotshot designer will come by and say,
"Oh--you know--I read an article and it showed that this kind of pattern is the best way
to encourage people to complete a transaction, so we're going to redesign this form of it."
That is fine, but it's also subject to the same risk, which is basically that those are
people's opinions. Now they may be highly informed and highly
compensated opinions, but they're not necessarily--you know--metaphysically certain to be the right
thing. In fact, what we find is almost always, there's
a better way than what your gut is telling you.
And then the third way is this before and after, which is a slightly more evolved way
of making an improvement to your site, which is you look at your metrics for this
week, you make a change and then you look at your metrics for next week, and then you
see if the metrics went up or down. The problem with this method is that if it
went up, you don't know if it went up because of the changes you made,
or because it was just seasonal, or because--you know--one of your colleagues downstairs in
another office was optimizing their AdWords campaign,
and so was getting better traffic to the site, or--you know--you got mentioned in a major
media publication--so it's really hard to know what lead to those improvements.
Website Optimizer kind of changes that, and essentially what happens,
is visitors arrive at your site and Website Optimizer will redirect those visitors to
an entirely different page that you tell us about,
or if you use a more advanced, multi-variant testing we'll actually change the content
of the page, as the visitor's browser renders that page;
so that effectively, every visitor to your site gets assigned to an experimental group,
and then we will report back to you as to, "Well, group A had this percent that made
it to the goal page and group B had a larger percent,
and group C had the smallest percentage." And those goals could be a thank you page,
they could be reading--filling out a form, it could be just going deeper into the site--you
can put those goal tags wherever you want. But the bottom line here is that instead of
guessing, instead of using sort of office political power to make changes to your site,
why not do an experiment, randomly assign visitors to each of the different ideas that
you have, and then let the visitors tell you-- if they all convert because they like--you
know--the image without the model and with the green buttons,
then let's go with that, and let's not sort of spend time kind of using voodoo web design
strategies. So Website Optimizer really helps make all
this possible. It is the web's most popular testing platform;
it's free; very easy to use. Often times, we see increases in conversion
rates--easily over 25%--a number of other resources that we make available,
and basically let's you do any kind of testing you can think of; works with all of your traffic,
not just your AdWord traffic, but your organic traffic from other ad networks;
it doesn't require any other of the other tools in this webinar--although as I mentioned
earlier, it really--when you use all three together--it's
really a powerful combination. So let's talk a little bit about what it would
be like if you decided to use Website Optimizer and you said, "Okay, I want to start the wizard
and create an experiment." The first thing we ask you is whether or not
you want to do a A/B or multi-variant test. A/B is the simplest way to get started.
It's great for low-traffic pages, also works great for high-traffic pages and helps you
test layouts very easily. And as long as you can create another page,
you can do an A/B test. Multi-variant is a little bit more advanced;
it lets you actually carve your page up into multiple sections and will basically swap
out content within those sections as visitors hit the page.
Let's make believe you decided, "Well, I'm just going to start with a simple A/B test,"
and you have your current page, you create another version of that page, you publish
it to the web, and maybe it's--you know--you have your index
page, and then you have your index2.html for your B page.
You just click "I want to create an A/B," you give the experiment a name, you give us
the URLs of the original and the test page; if you a have more than one test page, you
just say, "I want to add more," and then you just give us the URL of your conversion page--
usually your thank you page, or--you know--your contact form,
or whatever it might be--whatever page you consider to be a good outcome for someone
who saw that test page first. Then you can either have someone else add
the tags for you, and we'll basically give you a URL that you could email to your IT
Department, or if you are the IT Department, then you
say, "Okay, I'm going to do the tagging myself," and the way the tagging works is very simple
for A/B. You just have a control snippet that you put
at the top of the A page and then a tracking snippet that you paste at the bottom of the
A and B page, and then a goal snippet that you put at the
bottom of your thank you page or your goal page, and that's basically it.
So it's really three very simple copy and paste operations.
And then one on the goal page and--you know--it typically takes a few minutes to set up an
A/B test. Then you--after you add those tags and update
the pages, you basically click "Launch experiment," and then after a period of time when we start
seeing people visiting the site and we start dividing the traffic,
we'll report back to you and tell you exactly which version of the page did better, by how
much, with what confidence, what was the raw data,
what percentage improvement was the winner over the original, and really taking a lot
of the guesswork out of website design. So instead of kind of arguing around a conference
room and lobbying people, you can just say, "Well, why don't we just run a test?"
And everybody can come up with ideas and then we can see what the customers like best.
So here's an example of a test that we ran at Google for the Picasa team, and basically
they were curious about whether or not this page could be improved,
which actually had been performing pretty well.
As you can see, it's got an image of the product, it uses the word "free," pretty clean layout,
very action oriented headline. This was an alternate version that actually
had a 30% higher download rate. And what's most interesting is that we actually
took out one of the images, we changed the link to a button, changed the headline--
this was one of many different versions that we tested--we would never have picked this
one by ourselves. We probably would have picked one of the ones
that had more imagery, a little bit more content--but as it turns out, this one was the one that
won. So even at Google, we think we do a pretty
good job designing web pages, but we test everything, and Website Optimizer allows you
to do the same thing. If you ever use Gmail, you may notice that
sometimes you'll see things changing on the page--you know--
you might come in a couple weeks later and notice that, "Hey, I notice the login box
is now above or below." That team is always running experiments--they
probably run dozens of experiments--they test everything from text changes, to layout changes,
the order of the box, the design of the buttons, and they have had
great success with Website Optimizer. And so it's important to know that this is
a tool that is not only available to yourselves, but it's also a tool that we use and that
we've been using for the last couple of years. So here's another example.
This was provided by our partner, Wider Funnel, which is an agency that consults on onlining
pages and conversion optimization. Here their client had a pretty good looking
homepage, but they wanted to test a number of different layouts and--you know--
the naked eye doesn't really know immediately which one would do better, and certainly can't
tell which one would do better for all of the visitors.
What's most interesting is you'll see there's two different overall designs; there's a two-column
and a three-column. What's most interesting is that one of them
actually did better than all of the other ones.
You would think two would do similarly and then the other two would do similarly.
So one of them actually did double the conversion rate of the original--all of them did slightly
better--but one really pulled away and it was this version here.
Again, really hard to know until you run a test.
Here's another example, and this was provided by another partner, LunaMetrics.
Their client had a pretty good page right here; they sell stair supplies.
What they wanted to do was run a test where they provided the menu item, already expanded,
and then they had a friendly picture of an operator standing by to take your order.
One would not have expected this to make a major increase in your conversion rate, and
actually one would not have expected it to hurt it either.
But as a result of the test, they were able to find that these hypothetical improvements
that they wanted to test actually hurt their conversion rate.
And so had they not run a test, they may have just made these changes and let them run,
and actually harmed their sales dramatically. But because they ran a test, they were able
to basically roll it back and stick with the original, and then test other things that
might help increase your conversion rate. So one of the things to keep in mind is that
testing in not just about improving your conversion rate,
it's also an insurance policy, so that even the best intentions and the best changes that
appear to be good in a conference room-- you want to make sure that they're really
not going to hurt your sales--and so testing helps you kind of give you a safety blanket.
We got a lot of questions about Website Optimizer, let me roll through some of them.
One is, how can a communications based site benefit from Website Optimizers?
So, you don't sell widgets--maybe you are a professional services firm.
Every site has an objective. It may be not necessarily to sell something
and get someone's credit card information and sell them a product,
but it could be having someone contact you and filling out a lead generation form--
maybe you are a law firm and you want more people to contact you and fill out the "I
would like to talk to a lawyer." Well, you could test a whole bunch of things
on that homepage and see which one had the greatest result in getting new clients.
You may be a publication and your interest is actually people subscribing to your monthly
newsletter, or even more interesting, spending a minimum
amount of time reading a page. So whatever your goal is, you can run a test
for it. We've designed the tool to be very flexible;
it's not just for folks who are selling things. Will this affect my search engine ranking?
Website Optimizer is about testing and increasing your conversions, it's not for increasing
your search engine ranking. Will I be penalized for cloaking?
We have a whole FAQ about that. Testing is not cloaking.
Cloaking is where you're trying to mislead GoogleBot and then search engines and showing
them some piece of content, and then showing your users something completely
different. So--you know--if you're telling search engines
that you sell baby carriages, but you show visitors to your site stuff about online gambling,
then that is absolutely cloaking and it doesn't matter what technology you use--you will not
be protected for doing that kind of thing. But if you're just trying to improve a customer
experience and do the right thing for customers, and using a tool like Website Optimizer, that's
encouraged. What if I have a dynamically generated page?
No problem, as long as our tags get on the page, you can just add it to the template,
you can have your server add our tags dynamically as well.
Basically when the visitor's browser renders the page, that's when our tags do their magic.
They don't have to be--you know--officially part of the source code of the page from the
very beginning, they can be injected there by your server as well.
So, a few things to do this week. If you decide that Website Optimizer is a
good tool for you, one of them is look at your top landing page or homepage and ask
yourself, "Yeah, does it look professional? Is it going
to let me get my goals done? Does it make sense?"
Probably you're going to answer, "Kind of" to those questions and there's probably going
to be things that you can do to improve it. Well, make those changes, but don't make it
directly to the page in question, but make it to another page.
Then do a simple A/B test; maybe change the image, the text, the layout.
Then we'll split the traffic and then we'll help you compare them.
And then you can even make it fun--maybe have--you know--print out both of the mocks and have
folks in the company bet on which one they think would be better,
and make it a little bit more interesting. Or have two designers--you know--that are
always competing against each other--have them come up with their best version of the
page and see who's really better. Now for those who want the VIP treatment,
there are a number of partners that we have-- that we have authorized consultants as well
as technology partners that make it a lot easier for you.
The authorized consultants basically are available if you need spot technical assistance and
you want to talk to someone on the phone, or you want someone to actually do it and
to have it as a managed service; they can do all of that for you.
We also have contact management partners that make it really easy to set up those tags and
get the experiments running. The last thing I want to talk about for this
particular section is the resources available. So if you go to google.com/websiteoptimizer,
you're going to find a number of articles, tutorials, video demonstrations, links to
our user forum, case studies, a partner directory--it's the most information and resources on the
web for testing, so I hope you'll go there.
So now that brings us into the next section, which is a panel discussion, and these are
questions that kind of apply to all of the different products,
and the first one is how much the tools cost and why, and I was thinking maybe Jeff could
answer that.
>>JEFF GILLIS:
Yeah, sure. Well, obviously, as we've all said that these
tools are all free, and there's questions from everybody that we got recently is why
these tools are free; and one of the answers is that--and a very
legitimate answer is that Google's very user focused and all of these tools are going to
help webmasters make better websites, and also help users find the websites that
they're actually looking for, so Webmaster Tools will make better results in the Google
Search queries. Google Analytics will help with both content
and also you to understand where users are coming from and who your audience is;
and then obviously, Website Optimizer is going to help you really make appropriate web pages
that are very user friendly and navigation friendly as well.
But there's also another answer. All these tools can also help advertisers
and help our AdWords advertisers buy better keywords and do better AdWords campaigns,
And so there is also a benefit to Google's bottom line, but mostly it's user focused,
advertiser focused. And then none of these tools are--you know--without
some backend kind of effort. We'll talk a little bit, hopefully, at some
point about the partner network--I think Tom might have mentioned it--some of their partners
already-- but these tools are free, but it's a bit of
an investment to get something out of them. And so hopefully that answers that question,
Tom.
>>TOM LEUNG: Thanks, Jeff. The next question--maybe Mike, you can answer--whether
nor not these tools are for small companies or big companies.
>>MIKE WYSZOMIERSKI: Yeah sure, I can take that.
So the nice thing about these tools is that really anyone can use them.
But the applications of how they use them may depend on probably the size and the traffic
of the site. So starting with Webmaster Tools specifically,
a small business may not have a large advertising budget,
and really relies on organic search or non-paid search for their customers, so they really
need to use Webmaster Tools to make sure that they're showing up in the results.
A larger site may use the more advanced features of Webmaster Tools to monitor crawling or
manage site mass if they have a large number of URLs.
For Analytics, it's always great to know your visitors and both through Analytics and Optimizer,
a small business might just appreciate the fact that such a powerful tool is free,
while a large business would like Analytics because of how the data is presented in an
easily readable format and has all the various reporting options.
A large business might like Website Optimizer because it allows changes, which at a large
scale, could really affect a business positively or negatively,
to be tested safely and accurately, even in a complex environment where multiple factors
are affecting conversions. For all these tools, a large business may
decide to outsource the actual use of these tools to agencies who are experts on them,
but it's really up to you--anyone can be a user.
So Tom and Jeff, did you have anything to add on this?
>>TOM LEUNG: No, I think that was spot on. I think the next question is, I'm not that
technical, could someone use these tools for me?
I'd love to take that one. So all of these tools are designed to be kind
of do-it-yourself tools in the Google way that we've tried to really make the user experience
as intuitive and simple as possible. There is some technical involvement, as an
example, with Website Optimizer, you have to add a few tags to the page;
with Analytics, you put one tag at the bottom of the page; with Webmaster Tools, it's even
easier to get started. But some people may feel that, "Wow, I just
don't want to even kind of dip my toe in the water."
All of our tools are highly used by agencies and web design firms and SEO firms and web
development firms, and you could just ask whoever built your
site or runs your technical department to do it for you;
and we also have formal partner programs for the Analytics and the Website Optimizer businesses.
So that kind of is a good segway into some of the things that you can do that are really
exciting. And these may be things that you do or you
ask your professional services firm of choice to do for you,
but I thought maybe we could start with Mike and then we'll go to Jeff and myself,
as to what's the one coolest thing that people should be using from your tools that may not
be something that's entirely obvious, that's a good tip?
>>MIKE WYSZOMIERSKI: Yeah, sure. Starting with Webmaster Tools, I'm actually
going to combine two features into one super feature,
and I would say look at your external links in the "Links" section, and combine that with
the "What Googlebot sees," in links to your site feature as well,
and using this data together, you can find out who's linking to you and what they're
saying when they're linking. And if you look at this, you can get things
like product feedback or do reputation management--you know--
if you would find that a customer had blogged about your product--you know--find their link,
go to the blog and leave a comment on their blog,
and--you know--do it right and don't spam and they might just love to see a human face
from your business.
>>JEFF GILLIS: Thanks. For Google Analytics, this is Jeff.
I talked about a couple of the advanced features in the presentation, such as using those visualizations
like a Compare to Site Average, and segmenting by goals--it's just such an
easy way to see the performance of your keywords and your content and your campaigns and things
like that. But the coolest advanced feature right now
that I'm hoping people start using is the Site Search report.
So if you have a search on your website, again, not Google Search,
but a search box on your website where people can put in a keyword and maybe search for
a product that they're not finding easily, Google Analytics will show you a number of
reports about the data--about what people are searching for, what content they're actually
looking at after they do a search, what pages caused them to search and then
whether they're converting after a search, or doing another search--if they have to like
do multiple searches-- and it's so powerful, one of the ways you
can use that information is if you look at the terms that people search for on your site,
those are great keywords to possibly purchase or to look at and see if you're using them
at all, because that's what people are really looking
for and it's also stuff that they're not finding. Another way to use it is you can see for a
certain keyword, what page the person actually went to,
and you can use that to make Webmaster Tools know which landing pages you want certain
content to get, or create a campaign with that landing page
for that specific keyword. And then you can also see--you know--at its
most basic, what people are searching for, what content needs to be bubbled up and surfaced
because they can't find it easily. So it's another way to do some website optimization.
Thanks.
>>TOM LEUNG: So I think for Website Optimizer, my favorite feature is our Time on Page testing
capability, and this is one where instead of saying, "Well,
I want to find the page that is most likely to get people to some thank you page at the
end of a funnel," which is perfectly fine. What you do is you say, "I want to see what
version of this page is most likely to have people actually consume the content."
And so the way you set up the experiment is the goal tag is actually put on the test page,
and we give you a little code sample to have basically a little timer,
and you can say, "Well, if someone spends more than three seconds on this page, I'm
going to call that good--that's a good result." And so you can imagine like if you really
want people to read the "About Us" page, or if you really want people to watch--you
know--the 30 second flash demo of your product, or if you really want people to read some
article that you've written, and you're not sure if the layout should change,
or if a different version of the video is more likely to have people stay on the page,
you can run this Time on Page test, and the other hidden benefit to it is that
it has--usually will have a high conversion rate
in that a very large percentage of people will stay more than the number of seconds
you set it for, and that's going to allow us to give you much more rapid results as
well. So it's great for people who don't have a
specific commerce goal in mind, it's great for getting a lot of data quickly and having
a test run, and then getting to a winner soon, and it's also great for people who are interested
in making sure that people aren't just bouncing off of their--off that first page of the visit.
So that would be the Website Optimizer one and I think I'll end this with just a slide
showing you some of these great URLs. So if you found that Webmaster Tools is something
that you haven't been using, but now you understand why it's critical, go ahead to that URL and
get started. You've got nothing to lose, and it's free,
and you should help Google understand your site as best you can,
and that's going to likely increase the exposure that you get in the long run.
And then you can head over to Analytics and really understand everything that's going
on on your site; and why are people coming? And what's going
on? And be able to talk intelligently about what your site's doing and what the trends
are looking like, and then--you know--hopefully after that you'll
decide, "Oh, well it seems like this page is not doing as well as I thought it could
be, or our conversion rates are a little bit low,
or our cost for acquisition is getting up there,"
so then use Website Optimizer--start doing some A/B tests or get a little fancy and want
to do some multi-variant, that's great as well.
So hopefully this was useful. We really appreciate you taking to time to
learn about these tools and look forward to talking to you again soon.
Thanks very much.
