Google
Analytics is a powerful tool that can show you who your website visitors are,
where they’re coming from and what they do on your site. By tracking visitor
behavior, analysts, developers and marketers can often identify problem pages
or processes, but diagnosing exactly what’s wrong -- or more importantly, how
to fix it -- is one of any analytics package’s greatest gaps. That’s why, of
the more than 20 categories of applications one can use to boost the power of Google Analytics, the “Experiments” category was most intriguing to me. It’s only by experimenting with various versions of pages and page elements
that Web developers and marketers can hit upon ways to improve their site’s key
performance indicators.
Experiments Within Google Analytics
As I mentioned, Google Analytics is a powerful tool in its own right and so actually does contain some A/B testing/experimenting functionality. Within the interface, you can create an experiment on certain metrics you’d like to improve, including any of the goals you already have set or other behavior like bounces, pageviews or session duration. Assuming you (or your Web developer) have already created your variation pages and added the standard GA tracking code to all, you can then simply enter the urls for the original page and up to 10 variations. The only extra step is adding a special experiments code to the one original control page.
As I mentioned, Google Analytics is a powerful tool in its own right and so actually does contain some A/B testing/experimenting functionality. Within the interface, you can create an experiment on certain metrics you’d like to improve, including any of the goals you already have set or other behavior like bounces, pageviews or session duration. Assuming you (or your Web developer) have already created your variation pages and added the standard GA tracking code to all, you can then simply enter the urls for the original page and up to 10 variations. The only extra step is adding a special experiments code to the one original control page.
As with
most GA functionality, reporting is fairly robust and visually very easy to
digest.
 |
| The reports available in Google Analytics Experiments tool are robust and very easy to derive insights from. |
So yes,
basic experimentation functionality is built into GA, but let’s be honest –
hitting your KPIs and getting visitors to convert is the primary job of your
site. Given such importance, is there any other tool or resource that can help
marketers better optimize their conversion rates in an easier, more efficient
manner?
Enter Marketizator, Now Omniconvert
Within the Google Analytics Partners “Experiments” category, I was drawn first to Marketizator, partly because of its name and partly because of its description: “Marketizator is an integrated conversion rate optimization platform. It empowers marketers to increase conversions by combining the most important CRO tools under the same roof: A/B testing, advanced segmentation,web personalization and behaviour insight surveys." This more holistic approach, especially the inclusion of visitor surveying capabilities, would seem to make this application a worthy addition to any digital marketer’s toolkit, so I dug a bit deeper. Upon further research, I discovered that, just about a month ago, Marketizator changed its name to Omniconvert, changing its plans for the future at the same time. It has the goal of becoming democratic, allowing users to decide where it goes next. But in the meantime, the core functionality listed on the GA Partners website remains intact – and highly valuable.
Within the Google Analytics Partners “Experiments” category, I was drawn first to Marketizator, partly because of its name and partly because of its description: “Marketizator is an integrated conversion rate optimization platform. It empowers marketers to increase conversions by combining the most important CRO tools under the same roof: A/B testing, advanced segmentation,web personalization and behaviour insight surveys." This more holistic approach, especially the inclusion of visitor surveying capabilities, would seem to make this application a worthy addition to any digital marketer’s toolkit, so I dug a bit deeper. Upon further research, I discovered that, just about a month ago, Marketizator changed its name to Omniconvert, changing its plans for the future at the same time. It has the goal of becoming democratic, allowing users to decide where it goes next. But in the meantime, the core functionality listed on the GA Partners website remains intact – and highly valuable.
A/B Testing
Since A/B testing is the main component of GA’s Experiments tool, I was eager to see how a third-party application like Omniconvert would compare. The primary advantage to Omniconvert seems to be the ease with which users can create variation pages to test. Whereas Google Analytics assumes one has the power to create variation pages through a content management system or some other method, Omniconvert offers a straightforward visual WYSIWYG editor that can change things like copy, colors, placement, etc. with just a few clicks. The objective of the experiment can be set in a way similar to the GA interface, but when it came to Omniconvert, there wasn’t a mention of having to embed or deal with any code. Still, that doesn’t mean it isn’t required. Also, Omniconvert’s website did not dig too deeply into the product’s reporting capabilities; instead, the site simply states that the product offers “advanced reporting” designed to “make sure you take the best decision every time."
Since A/B testing is the main component of GA’s Experiments tool, I was eager to see how a third-party application like Omniconvert would compare. The primary advantage to Omniconvert seems to be the ease with which users can create variation pages to test. Whereas Google Analytics assumes one has the power to create variation pages through a content management system or some other method, Omniconvert offers a straightforward visual WYSIWYG editor that can change things like copy, colors, placement, etc. with just a few clicks. The objective of the experiment can be set in a way similar to the GA interface, but when it came to Omniconvert, there wasn’t a mention of having to embed or deal with any code. Still, that doesn’t mean it isn’t required. Also, Omniconvert’s website did not dig too deeply into the product’s reporting capabilities; instead, the site simply states that the product offers “advanced reporting” designed to “make sure you take the best decision every time."
Details
may be somewhat sparse, but a key testimonial is included on the A/B testing
page, which reinforces the importance of the functionality to marketers
everywhere. It focuses on the idea of doing everything you can within the
website to increase conversions, rather than simply spending more money to
reach more prospects/more qualified prospects. “We used Omniconvert to increase our sales and conversion KPIs without extra costs with media advertising,” said Dragos Oprea, Project Manager at Telkom.
Surveys
Just as it does not provide solutions to the problems it points out, Google Analytics also does not gather qualitative data. That’s why such systems as Customer Experience Management and Voice of Customer exist, to get at the who, how, what and why as a supplement to the when and where provided by analytics (Peterson, 2008). A key part of Voice of Customer systems are surveying tools, which are included in Marketizator/Omniconvert. As Marketizator puts it, such functionality provides “the ability to get answers from specific segments of visitors from the top of the funnel about their needs or barriers."
Just as it does not provide solutions to the problems it points out, Google Analytics also does not gather qualitative data. That’s why such systems as Customer Experience Management and Voice of Customer exist, to get at the who, how, what and why as a supplement to the when and where provided by analytics (Peterson, 2008). A key part of Voice of Customer systems are surveying tools, which are included in Marketizator/Omniconvert. As Marketizator puts it, such functionality provides “the ability to get answers from specific segments of visitors from the top of the funnel about their needs or barriers."
There seem
to be few barriers to setting up sophisticated on-site surveys using
Omniconvert. The surveys can be triggered four different ways -- on click of a
certain page element, on load of a certain page, on scroll, or on exit. Various types of questions are easy to create in the interface, as
is a certain level of branching (serving up different questions based on
previous answers). Users are easily able to choose what segment
of visitors they want to receive the survey and how frequently they want to
serve it up.
Such
qualitative data can help fill in quite a few gaps and give marketers even more
actionable insights. Omniconvert cites several possible insights its surveys can provide, including how likely visitors are to recommend the site to
friends, how qualified leads are, and exactly why people are abandoning their shopping
carts. Visitors can more deeply explain why they’re not
converting, or even if they ever intend to, assuming your survey asks them very
directly, “why are you here?” (Kaushik, 2010, p. 155).
Personalization
Omniconvert offers an opportunity to become more attuned to website visitors not just through surveying but also through Web personalization tools. The application says it offers the ability to “Target specific segments with on-load, on scroll or on-exit intent with creative overlayers or pop-ups that you upload or build in Marketizator editor."
Omniconvert offers an opportunity to become more attuned to website visitors not just through surveying but also through Web personalization tools. The application says it offers the ability to “Target specific segments with on-load, on scroll or on-exit intent with creative overlayers or pop-ups that you upload or build in Marketizator editor."
Essentially,
these “custom overlays” include certain content or calls to action on various
parts of the screen based both on the visitor and one of those same four
triggers used in the surveying functionality. These custom overlays can include dynamic text that changes based on several factors such as
user’s location or weather or any other text on the page. Omniconvert’s
reports then let you see if those overlays are impacting any of your KPIs.
In most
instances, personalization should boost the all-important KPI of customer
conversion. In a 2015 Chief Marketing Officer (CMO) Council study, more than 40% of CMOs believed the conversion of more customers was one of the primary benefits of personalized Web content and digital interaction. Omniconvert’s own case study of the Avon company seems to bear
this out. Avon included a pop-up on certain pages prompting people to answer a question
in order to find out what type of makeup is best for them. Depending on how
they answered the question, which had to do with one’s eye color, they saw
different content once they were at the cart stage of the site. By using such
personalization, “Avon optimized their visitors’ journey on site without
changing anything in the website’s design (like it happens in A/B testing)." Plus, the conversion rate on their products increased by more than 96%.
Conclusion
Web analytics in general, and Google Analytics specifically, has some shortcomings. One of those, testing solutions to discovered problems, can be addressed fairly efficiently from within the GA platform, thanks to its Experiments functionality. Still, if marketers really want to improve their KPIs and increase conversions across their site, they may be better offer implementing a more holistic conversion rate optimizer (CRO) tool such as Omniconvert to gain deeper and more actionable audience insights.
Web analytics in general, and Google Analytics specifically, has some shortcomings. One of those, testing solutions to discovered problems, can be addressed fairly efficiently from within the GA platform, thanks to its Experiments functionality. Still, if marketers really want to improve their KPIs and increase conversions across their site, they may be better offer implementing a more holistic conversion rate optimizer (CRO) tool such as Omniconvert to gain deeper and more actionable audience insights.
Sources
Kaushik,
A. (2010). Web analytics 2.0: The art of
online accountability & science of customer centricity. Indianapolis,
IN: Wiley Publishing. ISBN#9780470529393.
Peterson,
E.T. (2008). The voice of customer: Qualitative data as a critical input to Web
site optimization. ForeSee Results.
No comments:
Post a Comment