Data Driving » Web Analytics

Use Case: What traffic sources are driving quality traffic?

Nick — Fri, 05 Feb 2010 12:04:53 +0000

In this series, I would like to provide real-life analysis from the site Mytripguru.com. This is my site so I am comfortable reporting the traffic figures and analysis. The aim is to show how web analytics can be used to drive real change on a website. Remember, analysis means absolutely nothing if it does not drive change!

The first thing in any analysis is to frame the goals of the site in terms of Key Performance Indicators. For the purpose of this post, the goal of MyTripGuru.com is attract high quality traffic with highly engaged visitors. How would you measure high quality traffic?

First, I set up a goal in Google Analytics to define my Key Performance Indicator. In this case, I want to see how many users sign up for the site. This goal is very simple to set up in Google Analytics. A visitor who successfully signs up for the site visits two pages – the sign up page and the confirmation page. These have to be included in the goal. In the case of Mytripguru.com, the pages end in signup.php and my-account.php. The goal looks like the following in Google Analytics:

With Google Analytics, goals are tracked once they are created, so you will need to wait for data to be collected before insights can be gained. I set this goal up a while ago so I have some traffic numbers to analyze. The first question is how many times this goal has been completed on the site. This is an easy analysis. Just pull up the report Goals Overview and look how many times this goal was triggered (expressed as a conversion).

But this data is still pretty meaningless because it is not actionable. What I really want to know, is how traffic from various sources are converting. Are my SEM campaigns effective? How good is the traffic from referrals? How are my email campaigns performing? What about organic searches and direct traffic? To analyze this traffic I like to pull the necessary metrics into Excel using the Excellent Analytics Excel plugin. This is a very useful plugin that let’s you pull data via the Google Analytics API to take advantage of Excel’s powerful reporting capabilities. The report I generated in Excel looks like this:

Now we can see some more interesting figures. Direct traffic (medium: none) has resulted in 27 registrations and referrals have resulted in 42. Not bad. But a quick glance at the figures show that cpc traffic is quite high. I don’t see many conversions here. To put things into perspective, some simple data crunching is necessary. In Excel, I added two Key Performance Indicators – Bounce Rate and Goals/Visit.

Now, the data is starting to become meaningful. The metric Goals/Visit puts things into perspective. On the low end, cpc and organic traffic is worrying. On the positive side, email traffic is converting very well, although the traffic numbers for this medium are too small to be meaningful. As a web analyst, the clear priority here is in optimizing cpc and organic traffic to generate more conversions.

The next step is to drill down into the cpc and organic traffic. Since I am running a cpc campaign with several thousand keywords and virtually no one signed up, I believe the problem is with my landing pages. In another analysis I also noticed that most of the landing pages for organic searches are the same as for cpc traffic. So, my landing pages clearly need work. Had the number of conversions been higher, I would take a look at specific keywords and ads that are converting well or especially poorly. My focus for optimization would then be on the SEM side – such as writing better ad copy or using better keywords. In this case, I think the problem is not with the quality of my ads but my landing pages. Now the fun begins. It is time to take a closer look at my landing pages. But that is the subject of another post!

Mytripguru.com launched

Nick — Tue, 29 Sep 2009 20:24:43 +0000

Gutenberg, Germany

Data Driving is proud to announce the launch of Mytripguru.com, an easy to use, intuitive online travel guidebook where members are recognized as experts, or Trip Gurus, for sharing their travel experiences. The Guru Rating system rewards top contributors and promotes them as experts in their particular regions and activities.

Mytripguru.com is built on the premise that everyone is a Trip Guru and has travel experiences or local knowledge to share. The more a Trip Guru contributes, the easier it becomes for visitors to the site to get travel advice from other Trip Gurus who share similar interests.

The simple navigation and user interface is designed to let Trip Gurus focus on their travel writing and readers to find relevant travel advice as easily and intuitively as possible. By promoting the top Trip Gurus in specific and marketable destinations and activities, Mytripguru embraces local businesses who hold tremendous knowledge about things to do in their regions and can profit from high level exposure on the site. Furthermore, by recognizing Trip Gurus for the quality and quantity of their contributions, local businesses are encouraged to add related content in their region to increase their Guru Rank while safeguarding the integrity of the site.

Mytripguru.com is founded and managed by Nicolas Johansson, owner at Data Driving. For more information about Mytripguru.com or Data Driving, please contact Nicolas Johansson.

Safari 4 Beta distorts Web Analytics

Nick — Fri, 06 Mar 2009 14:06:46 +0000

There is an interesting article over at Formatur.de (in German) that talks about Apple’s release of Safari 4 beta and its potential impact on web analytics. The contention is that due to Safari’s new Coverflow-like feature to display the most visited sites, Safari makes regular “hidden” visits in the background to update its screenshots in the Top Sites. This means a website that is listed in the Top Sites section gets visited even if the user does not actively visit the site. The implication for web analytics, Formatur goes on to explain, is that Safari 4 users will have more page views, higher bounce rates and more ad impressions (therefore lower conversion rates) due to these hidden visits.

According to Formatur, there is no known workaround to this issue for web analytics providers as there is no method, using JavaScript or otherwise, to differentiate between a virtual visit and a real visit.

I tried to replicate this phenomenon on my site using Safari 4. On Wednesday, I visited my site several times to ensure it appeared in the Top Sites in Safari. On Thursday I used Safari for my web browsing but made sure never to visit my site. I then pulled up some reports in Sitestat and Google Analytics to check if there were visits from Safari 4 on Thursday to my site. I then segmented these reports to isolate my visit – in Sitestat I used my IP Address and in Google Analytics I segmented by City and Provider. While the segment in Google Analytics does not necessarily exclude other potential Safari 4 users in my city, the data from Sitestat validates.

Top Sites in Safari 4 Beta

Safari's hidden visit to Datadriving.com with in Sitestat

Safari's hidden visit to Datadriving.com in Google

These reports confirm Formatur’s suspicions that Safari 4 is making hidden visits to update its screenshots and these are being tracked by analytics tools like regular visits. The question for me is how often does Safari make these hidden visits and how will it affect web analytics if Safari 4 gains market share or, as Formatur predicts, the Top Site feature is copied by other browsers?

Is Google Analytics Being Evil?

Nick — Wed, 04 Feb 2009 08:45:01 +0000

There is an interesting article over on Sitepoint and originally broken by Michael VanDeMar about recent tests Google has been conducting to offer AJAX search instead of the standard HTML search it currently offers. Essentially, AJAX is supposed to improve the user experience by speeding up search. Besides some Firefox plugins possibly no longer working, there didn’t appear to be any major reasons to be concerned. However, the implications for web analytics tracking tools, it turns out, could be significant. From VanDeMar:

Every single analytics package that currently exists, at least as far as being able to track what keywords were searched on to find your site in Google, would no longer function correctly.

The reason, VanDeMar goes on to explain, is that search URLs in this test include a hash sign (#) which is then followed by the search parameters. Since browsers do not include anything after the hash sign in the referrer string, this information would not be sent to the tracking servers. Every visitor coming from a Google search, therefore, would appear to be coming directly from Google.com without search term information.

Web Analytics providers will be understandably upset if these developments are true. Search engine traffic is a huge part of any site’s traffic so losing detailed referral information would be critical to say the least.

Even if this test is rolled out and the implications turn out to be a problem, I am sure the web analytics community and Google will make nice and come up with a solution. And, while I doubt that Google is intentionally being evil, it is once again a reminder how even routine tests by the internet giant can have huge ramifications throughout the industry.

Whitepaper on Managing Web Analytics Projects

Nick — Wed, 28 Jan 2009 09:32:18 +0000

I am proud to release a white paper today together with my colleague Markus Hällgren from the Umeå School of Business and Economics. Despite the abundance of high quality white papers, blogs and books on web analytics available today, we continue to observe web analytics implementations that fall short of expectations. This paper is largely a result of our observations with web analytics implementations and project management consulting. It is our sincere hope that this paper will contribute to the industry and help take web analytics to the level where it can truly add value to organizations.

Download the paper now!

How to access Google Analytics on your iPhone

Nick — Sun, 30 Nov 2008 13:02:30 +0000

What’s the best way to get your Google Analytics stats on your iPhone? I have used three methods, each of which has benefits as well as limitations.

The first and most obvious method is to simply log in to your account via Safari, and it works surprisingly well. Essentially all reports are available including the new enterprise features such as segmentation. Since the iPhone does not support Flash, however, the date range is stuck on the previous 30 days. So, it is not possible to view visitors of the current day or for a specific date range. Viewing stats for multiple accounts is fairly cumbersome since there are quite a few clicks necessary to change website profiles. All in all though, this option gives you the most detailed information on your visitors.

If it is important to get detailed stats for a specific date range, for example for yesterday, then the best option is to define a dashboard in Google Analytics and set up a scheduled daily email. This works great as long as you know what you want and aren’t interested in doing any ad hoc analysis on your iPhone (not that you can, anyway!). This is a suitable option if you want to avoid the cumbersome browsing process with Safari.

The final option is to buy the iPhone app, myAnalytics. This app costs $1.99 or €1.59 and gives a quick but simplified view of all your Google Analytics profiles you have under your account. The reports are limited to Visits, Pageviews, Views/Visit, Bounce Rate, Time on site and % New Visits. The available date ranges are Last year, Last month, Last 2 weeks, Last week and Last day. The main advantage of the app is quick and easy access. With one click you have an update of the metrics outlined above for each of your website profiles. But don’t expect to get any actionable insights from this data!

Source: http://www.iphone-analytics.de/screenshots.htm

So these are the options I have used in the past. Anybody know of any other ways to get the data on your iPhone? In reality, all of these alternatives leave me wanting more. I guess it is time for Google to develop a Google Analytics App!

Nuconomy – Next Generation Web Analytics?

Nick — Wed, 12 Nov 2008 17:05:55 +0000

More Web Analytics tools are beginning to crop up as the industry continues to show signs of growth and maturity. Nuconomy is a new player in the field that places the emphasis on tracking and measuring the social web.

From nuconomy.com:

Go beyond traditional metrics like the page view and instead measure engagement. Understand all of your users’ activities and how they interact with the various features of your site. Automatically optimize your site to achieve your business goals.

As our social lives increasingly move online – to places like Facebook, Twitter, Netvibes, Digg, Pownce, Delicious, Flickr, LinkedIn (the list goes on) – and as websites become more like applications, these types of measurements will become essential to (more) accurately measure visitor engagement.

In addition to traditional metrics, Nuconomy tracks page activities such as AJAX, Flash and Silverlight applications. Data is available down to the user level, promising very detailed analysis and ultimately, richer insights (one would hope!). There is also a 2-way API that will enable a site to push dynamic content and ads to specific user groups and even to individuals.

The product is currently out of private beta but due to the volume of requests, not quite in public beta yet. There are some questions I am especially interested in learning about first hand:

How difficult is the implementation, especially for application-rich websites?
How accurate are the statistics?
Will this tool supplement or complement traditional web analytics?
How will traditional web analytics vendors react?

The next few months should hopefully reveal answers to these questions. In any case, this is a definitely an exciting time to be in web analytics!

Web Analytics in a Recession

Nick — Tue, 04 Nov 2008 15:05:04 +0000

There has been a lot of talk about the current economic climate, the credit crunch, the stock market tanking, government bailouts, the impending recession or even depression. Everywhere you look there seems to be bad news. But it doesn’t do any good to dwell on these things. At the same time, it doesn’t mean we can safely ignore them. How, then, should companies react to the situation we are currently in? In terms of website optimization, there are few things companies should be focusing on:

Focus on conversion rates
Conversion rates are the rate at which your visitors convert any defined goal on your website. Conversion rates are likely to go down as consumers tighten their belts.

Goals are the cornerstone to measuring conversion rates. There can be different types of website goals – for example to sell a product, to gain a sales lead, to drive traffic to a special offer, or to build awareness about a new feature. Each goal is defined by the confirmation page, or the final page when you consider the goal to have been reached. In the case of selling a product, the goal would be defined by the confirmation or thank you page after placing your order. In the case of my site, whose primary purpose is lead generation, the goal I should be measuring is the Thank You page after a visitor submits the contact form.

If you are using Google Analytics, set up some goals if you haven’t already done so. They are very easy to set up and will offer tremendous insight into visitor behavior.

Test and test some more
Good companies use the lean years of a downturn to optimize the hell out of their businesses. They invest in themselves and focus on getting internal processes running efficiently. The strong companies survive by coming out of a recession even stronger. The same applies to websites, which have a virtually limitless optimization potential.

With free tools such as Google Website Optimizer, it is cheap and easy to set up A/B and multivariate tests to optimize your site. This is the time to try out different landing pages or to move around elements on your home page. Try simplifying sign-up forms or reducing the number of steps in the checkout process. Not all optimizations will work but you will find your website to be continuously improving. The best sign of improvement – higher conversion rates!

Get innovative
Some of the best companies today were born during the last downturn. They used this time to innovate. Rather than lament the current situation, why not get creative and try some bold new ideas? Try ideas you ordinarily wouldn’t even consider. A/B testing offers the perfect environment to do this since you can easily try out an idea with a small sample of your visitors.

Joann.com successfully did this several years ago when they ran a series of product promotions and measured the results of each with a multivariate testing tool by Offermatica (later acquired by Omniture). They found, much to their surprise, that the least “serious” offer – buy two sewing machines and receive 10% off – was in fact the most successful. It turned out that people were ordering together with friends and neighbors to take advantage of this promotion.

This is a case study that shows how crazy ideas can turn out to be pretty good ideas. And by using A/B and multivariate testing, there is little risk in not trying them out!

Everydaycelebrations for Web Analytics

Nick — Mon, 06 Oct 2008 11:48:16 +0000

Everydaycelebrations.ca is a program created by General Mills for families, catering to every day family needs such as what to eat for dinner and what activities to undertake with the kids. By providing recipes and ideas for family outings, the site helps position General Mills as a trusted brand for families in Canada.

On the site, this program is described as follows:

Whether it’s advice on teaching your kids everyday life skills, ideas for keeping them occupied on a road trip, or a delicious dish that will treat their taste buds, General Mills Everyday Celebrations can be your resource to stir up family fun.

What makes this site effective?

As a branding initiative, the site works well. There are calls to action peppered throughout the site and between useful content for the target audience. The navigation is clean and simple and lends itself to surfing and exploring.

Everydaycelebrations does an admirable job in subtly reminding visitors that General Mills is behind the site. By offering various product promotions and including branded products in the recipes, visitors are gently guided towards these products in a way that builds trust rather than coming across as an advertisement.

The copy is written in the active voice, helping to entice visitors to click deeper into the site. It is clear that the titles such as “Love Your Thanksgiving Leftovers” were written for people and not search engine spiders!

How can the site be improved?

The sign up page has some optimization potential. The contrast between text and background makes reading the text a bit strenuous. There is no footnote for the “opportunity to save up to $5 on great General Mills products!**”, leaving the visitor wondering what terms apply. The form itself is quite long with mostly required fields. This could scare off some visitors. Finally, the large spacing between the form descriptions and the form fields makes it difficult to follow the form.

From a design perspective, there are a couple of minor elements that could be considered for optimization. For example, the search bar contains the text “Enter keyword”. A clearer instruction would be “Enter search term” or better yet, “Search”. This would eliminate any potential confusion. Also, the information about the program is buried in the footer and could feature more prominently on the site, since some visitors might need this information to make an informed decision to sign up on the site.

Web Analytics

Obviously, only General Mills can know how much uplift they are getting from this initiative. Web Analytics data could certainly shed some light onto the effectiveness of the site. For example:

Web Analytics Tip: As a marketing/content site, visitors should be staying on the site as long as possible. So, check the average time on site and look what content visitors are consuming the most and the least. Optimize or remove the low-performing content and add to or enhance the best content!

Web Analytics Tip: Track the number of sign ups as a conversion rate – number of sign ups / total visitors. This metric will show how the site is performing on one of the site’s main goals.

Web Analytics Tip: Track the number of times a recipe is viewed and printed. This will give a good indication of popular dishes and possibly even insights into which products to promote on the site or even offline.

Conclusion

In conclusion, this is an excellent site that effectively guides visitors towards specific goals. As is usually the case with well-designed websites, the site is in a good position to profit from more web analytics and site optimization measures such as A/B testing and form optimization. This is when the fun starts!

Sales Tactics from a Web Analyst

Nick — Thu, 18 Sep 2008 12:47:35 +0000

“I’ve got the perfect car for you!” (Photo: idreaminir)

Yesterday I visited the trade show, Online Marketing Düsseldorf (OMD). Generally, I was impressed with the quality of exhibitors. I noticed, to my delight, that the stands of web analytics vendors were especially well visited. But I also noticed that many exhibitors at the show did not take the time to ask what my needs were when I visited their stands. Instead, I received one canned sales pitch after another.

It occurred to me that a web analyst has some of the tools to make a good sales pitch. A good web analyst makes no assumptions, asks questions and listens carefully to the needs of the customers (website visitors). Obviously, a good salesperson does the same.

This service approach is hard work. It takes effort to listen. But it pays off in the end. And it is the only way to be successful. As a web analyst, that means taking the time to get to know the visitors on your site. Why are they here and how is the website meeting their needs? Understanding web analytics data requires a certain level of empathy to give meaning to the stream of data flowing in. As a salesperson, the same applies. What are my prospect’s needs and how can I assist?

So, my concluding insight after the OMD is that salespeople and web analysts can learn a lot from one another. Web Analysts often spend the better part of the day thinking about the needs, desires and motivations of visitors to their website. Salespeople, in turn, have tremendous insight into the pyschological aspects of selling, which is difficult to read from numbers. And, just like successful selling, the experience at OMD was a great reminder why it is so important not to lose focus on the customer!