1. The analytics data is not available in real-time, you have to wait until midnight passes the next day to see yesterday’s data
2. The data is not yours – it’s Google’s – and for all of the clean image that Google has (”don’t be evil”), forgive me when I say that I prefer to have my own copy of my website trend and visitor data.
3. Google Analytics works by including a JavaScript snippet in your pages, but search engine spiders (”crawlers”) don’t execute JavaScript when they load your site’s pages, so their visits aren’t logged by Google Analytics. So when spiders from GoogleBot, Inktomi (Yahoo), Bing (Microsoft’s new search engine), Ask and Baidu crawl your site, you know absolutely zero about it from looking at your GA reports.

To track these visitors that in the vast majority of cases go totally unnoticed, you need a server-side scripting package that logs all visits to a database, including search engine spiders. There are a few free ones out there, and one of the best of these is CrawlTrack – freely available to download at http://www.crawltrack.net/

The install for CrawlTrack is reasonably straightforward, and can be completed in a few minutes. All you need to do is upload the CrawlTrack files to a new directory on your hosting account, create a new database on your hosting account’s MySQL server, include a piece of CrawlTrack PHP code in your website code (in a header file, or another appropriate file, as indicated by the CrawlTrack documentation) and go through the automated installation procedure.

CrawlTrack does not interfere with Google Analytics, you can run them side-by-side no problem, and enables you to see which spiders are crawling your site and how often. Here’s a screenshot of the “Crawlers” activity view from one of my new websites just taken today.

This data shows the percentage of visits from certain spiders. A table at the bottom shows how often each spider is crawling your site, broken down by each individual spider. Note: a “Visit” is one individual page load by the spider.

You can click on the name of the spider (e.g. GoogleBot) to drill down into detailed crawl stats for that specific robot. This view will give you individual crawl stats for each page of the site crawled, as well as stats into what percentage of your site pages were not crawled.

This is great data for SEO as it shows the regard by which the search engines are giving to your site. This is very useful if you are publishing new content on your site, or updating it, and you want to see when the new content is indexed by the search engines, therefore enabling you to monitor the effectiveness of site changes, e.g. you update your site with a special offer, and wait patiently for enquiries to come in. But the thing is that you don’t know when the search engines will index the new content. With CrawlTrack, you can see exactly when this new content is indexed, therefore giving you the ability to correlate the offer with the response more accurately.

With CrawlTrack, you will also be able to see trends in your website crawl data that indicates interest from the search engines because of SEO activities like inbound link-building.

Then comparing your CrawlTrack data with your Google Analytics data may give you interesting correlates.

All comments welcome!

The name of the tool is Wonder Wheel and is available under the “Show Options” link to the top left of the main search results pane. Check out the screenshot below to see how to access it.

Now you get a sub-menu where there are loads of interesting filters for your previous search query, “fancy dress” in this case. There’s a Videos filter, Forums and Reviews filters and time filters which are useful for your average searcher. For SEO’s and keyword research though, the tool we’re interested in is the “Wonder Wheel” down near the bottom.

Ok now click on that baby and now we have a flash and ajax-ified wheel display which may resurrect memories of Set Theory in Maths class, showing search terms harvested from real search trend data in the Google database, which are related to the search term you had originally typed in. So for our original search term, “fancy dress”, we have the following:

We can click on any branch keyword from this star display to display both further search terms related to the branch keyword, which will also display new search results for the term we click on.

So how do we use this tool to help us with our keyword research and campaign setup? Well, we can start with a root keyphrase, such as “fancy dress” as we have in our example above, and split each of the keywords in the Wonder Wheel legs into their own ad group, using phrase and exact matching options. Of course this will also save us time, helping us to find keywords for our campaigns that we may not have found for a while through using analytics (see Building a Deep Keyword Google AdWords Campaign Without Third-Party Tools), or ever. Another great use of this tool is that it can help you to identify negative keywords for your ad groups.

This tool could also help us in a big way with how to lay out a website, helping to define hierarchy and categories etc, using a layout reflecting how the data is presented in the Wonder Wheel results.

Only certain users have been selected to test the prototype for this feature, so if it isn’t available to you then all you need to do to get it is to go to Google.com and paste the following Javascript snippet into your browser address bar and then hit enter:javascript:void(document.cookie="PREF=ID=4a609673baf685b5:TB=2:LD=en:CR=2:TM=1227543998:LM=1233568652:DV=AA:GM=1:IG=3:S=yFGqYec2D7L0wgxW;path=/; domain=.google.com");

So that’s it in a nutshell. Please do comment, ask questions, bookmark and subscribe to updates. Till next time, take care.

This can give you a crystal-clear view of what pages in the checkout process of your online store are literally losing you sales hand over fist. This post is an adjunct to an earlier post on this blog which is at http://webreach.ie/blog/checkout-goal-and-funnel-analytics-where-are-we-losing-sales-on-our-online-store/.

Once you can then see which culprit pages are scaring your customers away (just as they have taken out their credit card), you are thus armed with the knowledge of which pages need changing in order to boost your conversion rate and therefore your sales.

Ok so when you log in to your Google Analytics account, click on the website profile name that you want to add the goal and funnel tracking to, and on the next page click the the “Edit” link on the right hand side of the website profile. Next appears the website profile for your Zen Cart site. Click on the “Edit” link for the first available “Conversion Goals and Funnel” profile.

Next, enter in the data into the required fields as shown in the screenshot below.

So firstly, you have an Active Goal, then use the Head Match option with the Goal URL (the checkout_success page), name the Goal something like “Checkout Complete” and then fill in the details for the checkout Funnel, which starts with the “login” page, then the checkout_shipping page, then the checkout_payment page and then the checkout_confirmation page. The Goal URL should be the exact same value as the one entered at the top of the page.

If your site is not built on Zen Cart, just replace the URL’s in the sample above with those from your site, starting with the URL of the first page which begins the checkout process, and the Goal URL being the page URL which is displayed to the user when they have completed payment (the “thank you” page). Make sure to leave out the dynamic query parameters part of the URL (i.e. everything after the “?” in the URL).

So that’s it, click “Save Changes” and you’re all set! Now you will be able to view all of the Google Analytics Goals reports for this website profile. It will take a day or so before any data is available of course, and may take a week or two before you have enough data to make accurate judgements about what needs changing.

Make special use of the “Funnel Visualisation” report which is shown in my previous post at http://webreach.ie/blog/checkout-goal-and-funnel-analytics-where-are-we-losing-sales-on-our-online-store/. This is a very powerful chart showing visitor behaviour in your checkout funnel, visitor entry and exit points etc. and will show you what pages along the way in the checkout process that people are abandoning their purchase, giving you vital information about changes you need to make to your site pages.

Your questions, comments and feedback are welcome, and please do bookmark these posts and share the link love!

In a blog post back in March, (“First Things First – Examine the Analytics Data & Establish the Facts” I wrote about a friend of mine who has an ecommerce website which had a fair bit of traffic but not many sales.

Since then we put a Google Analytics Goal Funnel in place on the checkout funnel pages to examine what site users were doing when they got to the point of checking out on the site. A screenshot of a one-month view of the funnel is below.

This shows us that almost 9 out of 10 people who click on the “Check Out” button abandon the process along the way. See that only 18 of the 90 people who attempted to check out proceeded on to step 2 of the process, i.e. 72 people abandoned the check out on the “Login/Create Account” page. So we have an almost 90% funnel abandonment rate!

So this Login/Create Account page is the major culprit, causing the loss of 80% of potential sales on the site). So we need to take a very good look at this page and redesign and re-test it.

This site is built on Zen Cart, which is a great open-source template-based online store system. Now the default templates that come with Zen Cart are usually just customised by adding page styling by the web designers who build the site, as is the case with this site, but the overall structure and wording of the page templates are not changed. These default checkout page templates leave a lot to be desired regarding conversion optimisation, and they really need to be changed to achieve better performance.

A great resource for getting off to a flying start with optimising your Zen Cart site (or an online store site built on any platform for that matter) is the free e-book written by Eric Leuenberger from ZenCartOptimization.com. The ebook is available for free download at http://www.zencartoptimization.com/2007/05/09/20-surefire-ways-to-increase-sales-using-zen-cart/. This ebook details ways in which you can change your e-commerce page and site design so that you achieve a much higher sales conversion rate.

So that is where we are at now with this site. We are going to totally overhaul the Login/Create Account page, and keep an eye on the Goal Funnel, to see how the new design performs. From that we can expect at least a doubling of sales on this site.

Other pages such as the Checkout Payment page, also need an overhaul (its causing 30% of visitors to bail, just as they are about to pay), so we’ll implement the suggestions from Eric’s ebook on this also.

We will update this series of posts with a post on how our newly optimised pages are performing, and I will also be writing a separate blog post on how exactly the funnel is implemented on the Google Analytics account in a short while. Its pretty simple really, when you know how.

We appreciate your comments in the meantime.

One of the challenging things about setting up an AdWords campaign which is geo-targeting the Republic of Ireland only is that a lot of the third-party keyword tools out there have little or no keyword volume for searches performed in Ireland so therefore aren’t nearly as useful as they would be for analyzing say UK or USA audiences. You could use data collected from these high-volume countries to build out your keyword lists but, from my experience, often this does not reflect the search trends for Ireland.

So for this reason amongst others I have often set about building out a good deep keyword list for a campaign as the wider related keyphrases and synonyms are getting so little search volume in Ireland that they were virtually inconsequential.

So, in steps, here is what I find myself doing: Load up Google’s Keyword Tool (https://adwords.google.com/select/KeywordToolExternal, or by selecting “Tools -> Keyword Tool” from within AdWords) and from there typing in either the domain name of the client that I am building the campaign for, or, even better still, the domain name of a more comprehensive competitor’s website. Do check the box “Include other pages on my site linked from this URL” and click “Get Keyword Ideas”. From there we will get a good list of keywords.  Save/”Add” all of the appropriate keyword groups (some groups may be totally unrelated to your campaign, ignore these) that are displayed to your list which you can later download as a text file when you are finished adding.  Then underneath the saved keyword list on the right-hand side of the page, select “Download these keywords: text” link to get your complete list of keywords as a text file. Save and open using a text editor.

Now what we want to do is use this large list of keywords as an input into the Google Traffic Estimator tool and see which keywords get some search volume in Ireland.  From within your AdWords account go to “Tools -> Traffic Estimator” or go to https://adwords.google.com/select/TrafficEstimatorSandbox and from here you can copy and paste all of the keywords in your text file that you downloaded, into the “Enter Keywords” field. Choose Euros as the currency and leave the Max CPC field blank, and leave the Daily Budget field blank also, and then select and add Ireland to the Selected Countries box and click “Continue”. Now the Traffic Estimator will display a table detailing the search volume estimates for Ireland for the keywords selected. Sort this descending by search volume by clicking the “Search Volume” column heading link. Here now we can see what keyphrases (from the ones that we have gathered) are going to be worthwhile building out ad groups for.

From this I build out an ad group for each of the top trafficked keyphrases, using the keyphrase in the ad heading, and again in the ad text and/or display URL for the ad. This is called the “peel and stick” method. In each ad group I will write 4 different ad variations, making sure that I edit the campaign settings and change the ad serving option from its default setting of “Optimised” to “Rotate”, so that each ad is displayed with the same frequency. This technique is commonly called “split-testing”. Set the Ad-Group-Level Max CPC to be that of the estimated Average CPC (from the Traffic Estimator) for the particular Ad Group for that particular keyphrase. That way we know what our max CPC is going to be for that ad group.   To begin with, set all of the keyword match types in the campaign to be “Phrase”. This way we will avoid the poorly performing “Broad” match types (generally speaking, always to be avoided), and also enabling our ads to be triggered for the exact keyphrases in our ad groups that are searched for and also all searches for keyphrases containing that exact phrase but also with prefix word(s) and/or postfix word(s), for example, the keyphrase “fancy dress” with “Phrase” match type will trigger when the user searches for “fancy dress costume” and also when the user searches for “halloween fancy dress”.

Now, and this is the smart part: When I am starting out an an AdWords campaign like this, using Phrase match type on a broad search term, like “fancy dress”, I always tweak my Google Analytics profile for the target website to expose the exact keyword phrases that the users are finding my ads using. So for example, in Analytics by default, when someone finds and clicks on your ad after searching using the keyphrase “fancy dress costume” and the keyphrase in the AdWords campaign that triggered the match is “fancy dress” with Phrase match type, Google Analytics will only record this as an entry for the keyword “fancy dress”. Not very useful. To see how to configure Analytics to record the exact keyphrases that are finding your ads refer to Semvironment’s excellent article at http://www.semvironment.com/ppc-management-adwords-keyword-data-exposed-with-google-analytics/

Now, as the exact keywords are recorded over the first few weeks of the campaign, we know exactly which keyphrases people are searching for and finding our ads using. To improve the effectiveness of our campaign, and to keep it well ahead of the competition, we then create new ad groups in our AdWords campaign that are based upon the exact keyphrases that are recorded by Analytics Traffic Sources->Keywords report.

So, to be exact, we view this Keywords report in Analytics and take the top trafficked phrases that we don’t have any ads directly targeting and create a new ad group specifically targeting that phrase. So from the screenshot below of a live campaign that I am running, we can see that in #5 is the exact search term “halloween fancy dress”, which is finding the my ad which is targeting the phrase “fancy dress”. I don’t have an ad group targeting “halloween fancy dress” specifically so I will create one specifically for that and also I will add the exact match negative keyword ( -[halloween fancy dress] ) to my Fancy Dress ad group so that I am sure the correct ad group will be triggered when people search for “halloween fancy dress” in future.

Of course, other AdWords basic rules apply, such as always targeting specific landing pages on the target website which is most relevant to the keyword and therefore ad that the user clicks on, as well as using SplitTester.com to monitor competing ads within each ad group to see which one to drop and replace.

If you like this post please bookmark and/or link back to it and share the link love!  I really appreciate it.

So anyway, my friend mentioned to me that he wanted to get an AdWords campaign going in order to get some traffic to the site and generate more sales as business from the website have been slow. I remarked “ah, ok, well how many visitors are you getting to the site per day at the moment?” to which he looked at me perplexed and replied “don’t know!”. So we hopped up to the computer and logged into his Google Analytics account.

We were presented with the Google Analytics dashboard:

Ok, now correct me if I am wrong, but 1535 visits and over 7000 pageviews in the past month is not bad going, given that the site has not had any inbound link building done or any real SEO of any kind. So I suggested to my friend that we should first concentrate on getting more sales from the existing traffic before we go throwing money into Google AdWords when the conversion rate is abysmal.  

So what does he do to improve the conversion rate?  Well, we can see that most of the traffic is from the UK, coming in through the parked .co.uk domain. So the first thing my friend is going to do is to take a look at his pricing, taking a good look at his main competitor’s pricing for the products on the most trafficked product pages on his site. He can see which are his most viewed products by going to Content -> Top Content.  So an experiment would be to slash the prices of the top 5 or so viewed products and then see if sales on those products pick up.

But there is more that can be done of course, using Google Analytics more advanced features.  So, from a scientific metrics point of view, it will be useful for us, using his most trafficked product pages as a starting point, to use Google Analytics’ Goals and Funnels to chart the purchase path (the series of pages starting at the product page on through to the order “Thank You” page), and see using the Funnel Visualisation tool which pages along the way are causing the user to abort the purchase process. If most are aborting on the product page itself then either the page needs to be changed and re-tested, or the price needs to be reduced, or perhaps both.  Other pages along the way to the “Thank You” page may be leaky also. We will use the modify and re-test approach for these also until we are happy that the conversion rate is reasonably good. There is no limit to what you can do with the right tools and the commitment to using them.  We can also use a website optimisation tool like Google Website Optimizer to test variations of the pages on the site, analyse,  improve and re-test. 

But one thing is for sure, we can improve the conversion rate using Analytics data and tools. And this is the best approach at this point. Once we have the conversion rate improved to a satisfactory level we can then consider sending more traffic to the site with organic SEO methods and Google AdWords.  I’ll write an update post at a later date to tell you how we are faring on this project.  

Update: Part 2 of series of articles is available here http://webreach.ie/blog/checkout-goal-and-funnel-analytics-where-are-we-losing-sales-on-our-online-store/

Please feel free to leave comments if you have any feedback.