“A conversion is a lot like scoring a goal in a basketball game – it takes more than one player to make it happen”

Along with the new interface released last month, Google Analytics has released a great new feature that allows users to easily track all the ‘players’ that contributed to a customer making a conversion on your website. The ‘Conversion Dream Team’ can consist of paid advertising, direct traffic, organic search, referral websites, email campaigns, and social networking.

The new feature is aptly named Mutli-Channel Funnels and consists of four new reports that allow analysis of how each of your conversion players interacted to achieve the goal and what it takes to continue to score more goals in the future.

Using Multi-channel Funnels, you may discover that most of your conversions are achieved through your Social Networking efforts obtaining your client’s email addresses allowing for an Email Campaign which later leads to brand term clicks through Organic Search. Through this, you can continue to prioritise these marketing channels as you know they work together to achieve the most conversions.

Reports

Multi-channel Funnel reports are only available in the new Google Analytics interface, which by default is not switched on. To access the new interface, click the “New Interface” link which appears at the top of the Analytics interface next to Settings, My Account and Sign Out. The following are snapshots of the reports provided in the Multi-Channel Funnel section.

Assisted Conversions Report:

This report shows us the performance of each channel in assisting or achieving conversions (last interaction). You can see this in volume or monetary values.  The last column of the report shows whether each channel is more effective in assisting conversions (values above 1) or achieving conversions (values below 1). No longer do we have to guess if marketing tools are working or not – this report will assist in justifying the time spent on all online initiatives such as tweeting or responding to Facebook interactions.

Top Conversion Paths Report:

This report shows the top combinations of channels that led to a conversion. The visualization also allows for analysis of the order of the interactions that work best for your marketing strategy. Analytics also allows for the functionality to filter by number of interactions and conversion types (if your website has more than one conversion). Through this report, you can determine the most powerful combinations of online marketing and prioritise accordingly.

Time Lag Report:

This report shows exactly how long (in days) your customers are taking to convert. By creating Conversion Segmentations, you can filter sources to be included in the report which allows for powerful analysis that can assist in determining how long to keep various marketing campaigns live.

Path Length:

This report shows the number of conversions that resulted from the number of interactions with your website. Again, this report coupled with Conversion Segmentation can assist in improving the settings of your online campaigns.

Google Analytics is a free product that keeps on getting better. This new feature again aligns with the goal of Analytics – to assist you in achieving more conversions. Multi-channel funnels brings us into exciting times in online marketing, where instead of just relying on the goal scorer of the team, we can now create strong strategies involving the whole team with the ultimate return of increased conversions.

Earlier this week, a few of my colleagues and I attended the ‘Go Measure with Google Analytics’ event in Sydney. In this post I will  summarise  some of the key learnings and take-aways that we got from what was a very interesting day indeed (forgive the lengthy post, I have a lot to cover… ).

The agenda gave insight into some of the speakers we would witness over the course of the day, including several of Google’s own, namely John Jersin (Google Analytics Product Manager from the US), Alex Speziali, Peter Evans and Dave Smith (Analytics Specialists for Google AU/NZ), and Timo Josten (Google Analytics Certified Partner Manager, APAC & EMEA). Also on the panel were several Google partners – Christian Bartens, Director of Datalicious;  Daniel Rowan, GM of Internetrix, and Benjamin Mangold, Director of Loves Data (formerly Mangold Sengers), to offer their insight into some of the vast Google Analytics complexities.

New Google Analytics Features

The aim of the first half of the day was to educate the audience on some of the new features of Google Analytics, with most of the focus on the multi-channel attribution feature – the key to evaluating the performance of all your channels of online interaction. John stepped up first for the keynote address, and spoke of how version 5 of Google Analytics allows for easier data analysis, and provides more opportunity to leverage data. He wisely advised us all to stop thinking of just the last click, and start giving credit to channels/search terms playing their part by assisting conversions on our website. John also shared some stats on how the average number of pages per visit and average time on site metrics are dipping across the board in South East Asia – demonstrating lowering attention spans of web surfers and emphasising the need to capture their imagination as swiftly as you possibly can.

Conversion Paths

Christian was up next and certainly got the audience thinking. The topic du jour was attribution, and the message was to rethink attribution and look at the full path/journey of the user. Not just the different types of search terms that may have been used, but also the channels (touch points) that led them to your site along the way. There are many different ways businesses can engage with their consumer base, including e-mail, banner ads, through mobile content, paid search, affiliate marketing and social media. We need to start analysing more thoroughly how each of these channels contributes to overall conversions.

I particularly liked the idea of measuring ‘assists’ (or influencers) as last click attribution modeling may not give those channels the credit they deserve for the part they played in the decision making process. The new multi-channel funnels feature in Google Analytics makes this process far easier.

Google Analytics Goals

The focus of Alex’ presentation was around site goals – specifically looking at more than just your obvious key goals e.g. sales. She recommended splitting your goals into micro (small) and macro (large), as this will give you a more complete picture. Many of you will be familiar with the cyclical purchase process: consideration – evaluation – purchase – post purchase/ongoing exposure. Define goals for each stage of the consumer journey, for instance, consideration goals (micro) might be time on site or pages per visit. Active evaluation goals (again these are micro goals) might take the form of watched videos, requested goals or viewed testimonials. Post purchase micro goals could be newsletter signups, Facebook ‘likes’ or Google +1’s.

Mobile and Social Plugin Tracking

Time was ticking along, so next we had a couple of Googlers, Peter and John, very briefly covering mobile and social plugin tracking. Unless you’ve been living under a rock for the last 10 years, you’ll very likely be taking part in the smartphone explosion (you may even be reading this on one!) – the prediction is that by the end of this year, half of all Aussies will have a smartphone. Amazingly 75% of large online advertisers don’t have a mobile optimised site – I’m sure this number will start dropping sharply when people start to understand how well this particular demographic can convert in the right circumstances.

With social plugins, you can see how visitors to your site engage with plugins such as the Google +1 button or the Twitter ‘tweet’ button. Since there is no new pageview when clicking these social buttons, some tweaks need to be made to register this data in GA.  For plugins that aren’t Google owned, you will need to make a call to the new _trackSocial() method – more on this can be found at http://analytics.blogspot.com/2011/07/social-plugin-tracking-in-google.html. We can also see from this what part the different plugins play in terms of ultimately converting visitors.

Landing Page Optimisation

The 2^nd half of the day focused more on the testing (LPO) side of things. The main hurdle to overcome with testing is the age old question of what to test…

Timo from Google suggests that you:

• Start where you are losing money!
• Use Analytics to identify opportunity
• PRIORITISE – start with pages losing most revenue (e.g. highest bounce rate)
• Rule out low quality traffic
• Think BIG! Then refine
• Ask your visitors!
• Find where users leave your site
• Do some competitor analysis for inspiration
• Ensure accuracy of your data

Following in a similar vein, Ben then talked through some examples of how Loves Data got some great results by making some very small changes for their clients including NIB, Amnesty International and BT Super for Life. For NIB, playing with the colour scheme and moving the quote form positioning resulted in a whopping 32% uplift in quotes subbed! Again, the message here is keep things simple, only ask for what you really need, and make it look easy.

Page Speed Optimisation

One of the more mind-blowing stats of the day came next when John stepped back up a 3^rd time to talk about page speed – every second of latency can affect your conversion rate by up to 7%! This one is particularly interesting, and one I intend to use to hammer home the importance of page speed to my clients. Next we saw another case study, this time a travel website called Amari, who improved their page load time by 11.5% just by removing whitespace in their HTML code and by actioning the low hanging fruit taken from the Google Page Speed tool. Even minor changes can have a big impact!

Live Examples

Finally, the panelists went through some live examples of audience submitted webpages to make some suggestions for potential test elements.

Much of what we heard were standard testing techniques – things like keeping important items above the fold, testing different offers/headlines/colour schemes/incentives, trimming off as many unnecessary fields as possible, displaying pricing, and demonstrating social proof ie. testimonials/case studies.

And with that, the day was wrapped up!

If you attended the event, we’d love to hear your thoughts on it, and what parts you enjoyed most. Reply and let me know how you intend to use the insights you learned on the day!

In this past month, Google (the “Googleplex”) has been busy announcing and rolling out several new search features, algorithm updates and innovative projects (literally keeping us SEO’s on our feet!). However Google have made yet another feature announcement, this time for Google Analytics – Social Tracking.

I’m sure everyone is jumping for joy, well I sure am . For the most part, the ability to track and monitor social interaction, or basically measuring what a website visitor really likes to read to the point of sharing it with their friends, had become pandemic for companies — even enterprise companies that are investing tens of thousands of dollars into social media campaigns.

More advanced implementation of Google Analytics code using virtual page views, event tracking or the dreaded “URL Tagging” method was required to view social media data in GA – these are required no more.

I speculate that the introduction for this feature was due to the following reasons:

• Social buttons have become more common to webpages and sharing content, hence the need to understand and track this type of interaction has also grown.
• Google has been executing a reactive flanker strategy in the ever expanding social media landscape. Simply where Facebook has been succeeding, they have been responding with flawed social products i.e. Buzz, Wave in order to keep up in the “game”.
• More obviously Google’s acquisition of Post Rank earlier this month, and its technology to monitor and aggregate social media has been used to extend the functionality of Google Analytics.

How to View Social Interaction Reporting for Google Analytics

The Social Tracking feature lets you track people who socially engage in your content, from +1s to Facebook Likes, to commenting and tweeting content. Google adds a new option to the new Google Analytics interface for “social” under the Visitors menu with the following options:

1. Social Engagements – This overall report demonstrates site behaviour changes for visits that click on any social sharing button. +1 is added automatically; however,  tracking other buttons will be clarified in the “how to track section”.

2. Social Actions: Identifies the number of +1 click, Likes, Tweets, etc.  used on a website. A helpful report to prioritise the sharing buttons used for articles and blog posts (like this one!)

3. Social Pages: This reports drills down to identify social interaction activity for individual pages- very useful to identify the most viral content shared by website visitors.

In order to view these reports it is essential to correctly setup the GA tracking code, more of which will be clarified in the next section.

Setting Up Social Interaction Tracking for Google Analytics

For Google Social Tracking – +1

Google have stated that reports automatically include information on any Google +1 interactions that occur on a website, however to track non-Google interactions, you will need to modify your tracking code.

For Non Google Social Interaction Tracking

As mentioned from Search Engine Land; the _trackSocial method will only be used for creating social tracking as mentioned from the following syntax:

_trackSocial(network, socialAction, opt_target, opt_pagePath)

1. Network: Name of the social network (google, facebook, twitter, digg, etc)
2. SocialAction: Type of action (like, tweet, send, stumble)
3. opt_target: Subject of the action being taken. Optional, defaults to the URL being shared (document.location.href). Can be manually set to anything: a different URL (if they’re sharing content that “points” to another URL), an entity (e.g, product name, article name), or content ID
4. opt_pagePath: The page on which the action occurred. Optional, defaults to the URI where the sharing took place (document.location.pathname). Can be manually set (like a virtual pagename).

For more information on best practice for technical setting up and implementing non Google social interactions tracking (Facebook, Twitter), click here.

Barriers for using Google Social Media Tracking

The main problem Google face is with tracking other social elements, due to the lack of consistency each social network has taken when designing their social buttons. Google has done a great job of creating some working examples of Facebook and Twitter integrations, but even with this there is a large amount of legwork when trying to modify the Google Analytics code.

None the less, Google have again broken the barrier in providing useful features that businesses can use to make actionable business decisions for social media campaigns.

Those that haven’t signed up for Google Analytics in the last year may not have noticed that Google have updated their Google Analytics tracking code! The migration to the asynchronous (async) tracking code is probably the most significant change to the data collection mechanism of the analytics product since Google Analytics was launched in 2005 – for the first time allowing users to place the tracking code anywhere on the page without interfering with page load times.

The benefits of the new tracking code include: -

1.       Faster page load times,

2.       Increased accuracy in tracking of visitors,

3.       Less page interruptions,

4.       IMPROVED SEM AND SEO RESULTS!

Should you upgrade to the new tracking code?

Read on to find out the benefits to you (The Good), why you might not want to upgrade (The Bad) and instructions on how to upgrade tracking code (The Ugly – for those that don’t code).

The Good:  Benefits of the Async Tracking Code

Improved SEM and SEO Results: For us at Outrider, Australia’s largest search engine marketing and optimisation agency, we think the best part of implementing the new tracking code is the improved page load times and the effects of this on SEO and SEM. Google is no stranger to enhancing online user experience and has publicly stated that they use site speed in determining Google organic search engine rankings, as well as using page load times to determine quality score for ranking of Google sponsored links. By implementing the new tracking code, you’re improving your SEM and SEO and you’re opening up the opportunity of increased ROI through lower CPCs and increased rankings in your paid and organic listings!

Optimised User Experience: Compared to the traditional ga.js tracking code which stops all other activity while it’s loading, the new Async tracking code improves page load time by using code that is able to load in the background while your page content loads. Page load time tests by Position 2 show that the Async code does provide faster load time of websites compared to the traditional ga.js tracking code. By having a website that loads fast without any errors, your visitors are more likely to be in a good mood and ready to convert.

Better Tracking Accuracy: The new Async tracking code increases accuracy of reporting by tracking website visitors as soon as they visit the webpage compared to the traditional code which tracks visitors after the whole page content has loaded. Google recommends that the new async tracking code be placed at the top of the page in the <head> section of the source code, which allows for tracking of visitors that may exit or click a link before the whole page loads. After installing the new tracking code, your bounce rates may rise but at least you will know your true weakest pages and maybe even identify several that don’t work!

The Bad: Why not upgrade to Async Tracking Code?

Flash websites: If your website is mainly flash, it’s likely that you’re using the Adobe Flash component approach to track your site and simply inserting the new code in the <head> will not work.

The Ugly: The Tracking Code

Place the following Async Tracking Code at the bottom of your <head> section on every page of your website and get rid of the old ga.js tracking code.

Customised tracking tags:  Custom tracking tags such as virtual pageviews, e-commerce, and cross-domain tracking will require basic recoding. This doesn’t mean you shouldn’t upgrade just because you have custom tracking tags.  However, be warned that tags will require updating for features to be retained.

Note on Site speed: The new Async tracking code doesn’t automatically track page load times, however to activate this feature you need to modify your traditional code anyway. So, why not kill two birds with one stone and update your tracking code as well as insert the Site Speed code to your website. Read Outrider’s blog article on Google Analytics Site Speed Feature.

Google have recently released a new reporting function within Google Analytics  to measure ‘Site Speed’.  The report has been developed to measure page load time and I have to admit, it is insightful.

The report sits within the Content section of Analytics reports:

• Avg Page Load Time—The average amount of time (in seconds) it takes that page to load, from initiation of the pageview (e.g. click on a page link) to load completion in the browser.
• Pageviews—The actual number of times the page was viewed for the selected date range.
• Page Load Sample—The actual number of pageviews that were sampled to calculate the average page load time.
• Bounce Rate—As for Pages report, the percentage of views to this page in which this page was the only one viewed for the session.
• % Exit—As for the Pages report, the percentage of views to this page in which this page was the last page in the session.

How does it work?

Although there are other products in the market that provide similar reporting, they only measure part of the picture.  With Google’s Site Speed report you can measure page load times from the time the page was actually requested, until all the elements of the page have finished loading.  

How does this compare with traditional methods?

Traditional methods: Document loaded, start tracking time, all objects loaded, stop tracking time

Google’s method: Page requested, start tracking time, document loaded, all objects loaded, stop tracking time

Site Speed Tracking set-up

In order to activate the site speed report and start the data collection process, you need to make a small adjustment to your current Analytics code.

Simply insert the additional tracking code snippet shown below to your existing Google Analytics code on your website, and away you go.

_trackPageLoadTime();

Sample Google Analytics snippet:

 <script>
 var _gaq = _gaq || [];
 _gaq.push(['_setAccount', 'UA-XXXXX-X']);
 _gaq.push(['_trackPageview']);
 _gaq.push(['_trackPageLoadTime']);

 (function() {
   var ga = document.createElement(’script’); ga.type = ‘text/javascript’; ga.async = true;
   ga.src = (‘https:’ == document.location.protocol ? ‘https://ssl’ : ‘http://www’) + ‘.google-analytics.com/ga.js’;
   var s = document.getElementsByTagName(’script’)[0]; s.parentNode.insertBefore(ga, s);
 })();
</script>

It is worth noting that before you enable Site Speed reporting for your website (by the addition of the additional code snippet) both ‘Page Load Time’ & ‘Page Load Sample’ will show zero values in the Site Speed report.  You may need to wait until at least several hundred pageviews have been accrued to view a small portion of data in your report.

Data should start to appear within the report a few hours after publishing changes to the code snippet.  (Just select the current day, from within the date range selector and if you see data, you know set-up has been completed correctly).

How can Site Speed data be used?

The report is most useful for measuring the latency (delay) on your critical pages.  Poor performance could be caused by browser type selection or poor internet connection speeds in a given geographical area.  The additional insight will allow for a more targeted approach to tackle overall site performance.

For example:

• For those areas with poor internet connection speeds, trial rendering alternative landing pages,  better- suited to the reduced speeds

Good News for Advertisers and Clients!

Sub-optimal landing page load times, in a given geographical area or when using a particular browser, lead to a poor user experience.  In turn, this may lead to negative sentiment towards a brand or a potential loss of conversions.

By reviewing data provided within the Site Speed Report, small changes to landing pages could provide the win-win situation that we all strive to achieve.  What do you think?

Last month Google made the announcement that in addition to its regular search choice they would start offering a new “secure” alternative.  The new offering can be found at https://www.google.com and the idea behind the new service is rather simple. By using SSL technology you are able to create a secure and encrypted connection between your browser and Google, ensuring that no third parties on your network will be able to intercept your search results.

However, this new enhanced privacy potentially comes at a cost for those of us out there who rely on analytical software to gain valuable insights into how users come to discover our site. Let’s say for example you owned a website that sold home theatre systems, traditionally when a user searched for the phrase “buy home theatre” in Google they would be taken to the address http://www.google.com/search?q=buy+home+theatre. Then when a user clicked your listing in the search results and was taken to your website, using analytical software you could then see what page the user came from, and in turn, discover what search phrases they used to discover your site.

Conversely though, due to the way modern browsers handle the transition from a secure SSL connection to a regular webpage all information that would regularly be sent in the browsers referrers string is lost, leaving website owners seemingly with no way to determine what phrases are used to discover their site in Google. Some in the industry have been screaming that this is the death of analytics; however this is simply not the case.

It would be worthwhile noting that Google’s new SSL anonymous search alternative is in no way the default option and while it is too early to determine any meaningful data in terms of usage and adoption rates among regular web users, the general consensus so far seems to be that this is not going to be the next “big thing”.

Furthermore, something that seems to have been lost thus far in the panic is that for years now the millions of users running one of the many Internet security programs from vendors such as McAffe, Norton and Symantec tend to strip the referral string by default, albeit to zero fuss.

While it does seem at this stage rather unlikely that this new offering from Google will have any major effect for most website owners at this stage, for those who wish to err on the side of caution, a solution exists.

As previously mentioned the information contained within the browsers referrer string is only lost when the user’s browser makes the transition from a secure SSL connection to a regular unencrypted connection. Simply by offering users a secure connection when visiting your site you will be able to continue to capture the valuable analytical insights contained within the referrer string.

In the end it remains to be seen exactly what, if anything, this means for analytics. In the mean time, it’s nice to know that a solution exists.

A topic that comes up quite often in SEO discussions is the merits of outbound linking, whether you should link to your competitors and if outbound links encourage visitors to leave your site. You should consider the user experience first and foremost when thinking about outbound links. If the link directs the user to relevant content that they would be interested in, then the link is worthwhile. If you’re worried about excessive loss of traffic through external links, then it makes sense to keep a close eye on the click activity around these links.

There are a few different ways of measuring the number of clicks on outbound links from your site. The one I wish to highlight is a fairly neat way of tracking these clicks in Google Analytics, that some of you may not be aware of. Unfortunately, the ‘Site Overlay’ feature of Analytics only gives you click volume on your site’s internal links, and not the links pointing away from your site.

For the purposes of illustration, yesterday evening I tagged up one of our previous blog posts, David Boyd’s thought-provoking ‘Viral Videos and the SEO Payoff’. This particular post contains 6 links to YouTube videos in the second paragraph, the first of which is a clip called ‘Charlie Bit My Finger’. A regular link to this YouTube page might look like the following:

< a href=”http://www.youtube.com/watch?v=_OBlgSz8sSM”>Charlie bit my finger< /a>  (Spaces added for display purposes)

However, I’ve tweaked this slightly (along with the other links), so I can see how many visitors to this blog post click the links to each video. If you look in the source code for the post, you’ll see the link looks as follows:

< a href=”http://www.youtube.com/watch?v=_OBlgSz8sSM” onClick=”javascript: pageTracker._trackPageview(‘/outgoing/CharlieBitMyFinger’);”>Charlie bit my finger< /a>

The part bolded above is the addition that’s needed to register the click information for this link in Google Analytics. When someone clicks one of these links, it’s registered as a page view in Google Analytics. Note that the part in brackets ‘/outgoing/CharlieBitMyFinger’ can be anything of your choosing, but it makes sense to keep it meaningful. When you wish to see how many visitors are leaving your site through each of the links you’ve added, you need to run the ‘Top Content’ report in Google Analytics. Here’s my example:

Tagging each link with /outgoing/ makes it easy to filter your report like in my example above. What I can see from the above report is that, in the last day, 2 people have visited the Outrider blog post on ‘Viral Videos and the SEO Payoff’ and then clicked on the link to the YouTube video ‘Charlie Bit My Finger’, with one click each on the links to ‘Beached As’ and ‘Mean Kitty’.

Note: this only works for sites that are using the new version of Google Analytics tracking script, not the old legacy version. To see how to implement this functionality using the old tracking script, please see Google Analytics Help.