SwellPath

Archive for the ‘Analytics’ Category

ACCELERATE 2011 in San Francisco

Published November 5th, 2011 Analytics, Events 1 Comment

Eric Peterson and the crew at Web Analytics Demystified have yet another project that benefits the analytics community. The first ACCELERATE conference will go down in San Francisco on November 18th. Web Analytics Demystified already is planning for more of these in other cities, but undoubtedly San Francisco will not disappoint for the inaugural event. Best of all, the event is FREE! As a result, it is also sold out. Hopefully you landed yourself a ticket, but if not, keep up on the next ACCELERATE announcement via the Web Analytics Demystified blog.

The conference is ending with an Ignite-like Super Accelerator session. I’ve thrown my hat in the ring and have a few words of wisdom of my own to impart on the crowd. If are making it to ACCELERATE, make sure you stay through the end and see the Super Accelerator session. The list of presenters is great, and it’s sure to be an entertaining hour. Hope to see you there!

For Real this Time: Google Analytics Real Time

Published September 29th, 2011 Analytics 4 Comments

In spring of 2010 I made the call that Google would (probably) have real-time reporting by fall of 2010. I did this publicly at a Web Analytics Wednesday here in Portland. I was off by 1…year. Today Google publicly announced real-time reports for Google Analytics.

Now everyone with Google Analytics can geek out on their real-time data like those of us with Chartbeat have been doing for the last couple years. In all seriousness, real-time reporting is an exciting new feature, and does provide some fantastic insight when there are “momentous” events driving traffic to one’s site. An example might be a rush of traffic from a blog post getting picked up on Digg Reddit. There is a lot of debate in the web analytics community about the value of real-time analytics, with many practitioners saying it is simply monitoring, and not analytics. But with the speed of change these days, and the quickness with which a swell (see what I did there?) can form and move, some type of real-time reporting is necessary.

So will Google put Chartbeat out of business? Hardly; Google has a ways to go before the features are as robust as Chartbeat’s offering. But most sites don’t need most of those features; those that do are already paying the $10/month that Chartbeat starts at. As Google did to the paid analytics platforms when it started to gobble up market share, this will simply force Chartbeat to continue to improve and enhance it’s offering. A challenge Chartbeat is up to, and has proven it is capable of doing for two years. Again Google is pushing the industry, and improving the conditions for all of us.

Finding out about the real-time rollout was one of the more exciting moments at the Google Analytics Partner Summit earlier this month. I am excited for our clients using Google Analytics, and look forward to playing around with it myself. It should roll out to all users over the next couple weeks. Kudos to Google for an impressive enhancement; real-time reporting with the volume of sites Google Analytics is on is no small technical achievement.

Google Analytics Premium – Validation for the Enterprise

Published September 29th, 2011 Analytics No Comments

Google Analytics today announced the release of Google Analytics Premium, a paid version of the platform targeted at the enterprise. For those that still doubted Google Analytics ability to handle enterprise level web analytics, this certainly should provide validation. We’re excited that our clients will now have the ability to implement an even more robust version of Google Analytics, and that Google is so dedicated to raising the caliber of the product.

I won’t go into detail on specific features, as most of that is still confidential. But here are some basic elements of premium:

  • The ability to handle larger sets of data, from collection through analysis.
  • The ability to gather more customized information about visitors and visits.
  • Deeper integration with Google’s advertising networks, providing richer information and more actionable data for advertisers.
  • A more holistic approach to digital analytics.
Ultimately Google has done a great job of making the offerring a win-win for those that can justify the cost. For those that aren’t quite at that level, Google Analytics “Regular” still will continue to improve, and there are some VERY exciting things coming in the near future. But for those that continue to doubt the ability of GA to handle enterprise level web analytics collection and reporting, this should be more than enough validation.

Recap of the Google Analytics Partner Summit 2011

Published September 23rd, 2011 Analytics No Comments

As most of the world has heard already, SwellPath has recently been accepted into a very elite group of Google Analytics (GA) professionals.  SwellPath has recently become a Google Analytics Certified Partner (GACP).

 

What’s a GACP? 

Being a GACP means that Google has validated SwellPath’s work in regards to advance implementation of the solution, advanced analysis and very importantly a proven and effective process for engaging our customers and understanding how their business needs translate into web data, insights and areas for optimization.  As a member of the GACP, we now have access to new GA features that are in the testing phase, access to the GACP forum for shared QA among partners and the Google Analytics team and finally an invitation to the annual Google Analytics Partner Summit.

While the majority of the amazing improvements to the platform are not public yet and therefore confidential, we can’t mention specifics on new features or release dates.  GACPs got an inside look at demos of new features that will be rolled out to the public soon and others that are still in the testing phase.  The main takeaway from the new feature demos is that Google is really putting a lot of resources into improvements in data collection, speed, UI and data access.

What does this all mean? 

The gap between the large enterprise solutions and GA will be diminishing over time.  The new features will make GA an even more robust solution and narrow that gap where the current large enterprise solutions have the advantage.  SwellPath has helped clients migrate from enterprise solutions to GA and the “easy” part was building a custom tagging strategy that met the client’s needs and provided, at the very least, the same data collection as they were getting from their enterprise solution.  The big difference was in the UI and reporting options/features/visualizations that the enterprise solutions offered once the implementation was complete.  The enhancements to GA that are currently in the queue will meet or exceed what the other enterprise solutions offer.

So what does Google get out of this? 

Google does like to do things for “free” but why are they aggressively making improvements to GA that will help all of us non-paying GA users do our job?  The writing on the wall was there if you looked at it.  Deeper validation and attribution of ad spend and social media investment.  Can you hear a CMO asking what was our ROI for that $20k social media campaign?

At the Summit, Google showcased some amazing features of Android and Google+.  Why? The data is showing that social spend needs to be justified and therefore measured more precisely (PostRank acquisition) and the dramatic upward momentum in the mobile world is where more and more people will be connecting with the internet, maybe the majority in the near future…  With these massive changes in how and where people engage with the web you can either be reactive and roll out solutions to solve the tracking issues once they surface or proactive and develop a tool that will be prepared to incorporate the changes before they are occurring.  An example of this is the release of the Multi Channel Funnels report.  The data is there for everyone to see but are we even ready to use this data to drive our marketing mix strategy?  Maybe and maybe not quite yet.  Have CMO’s been asking us for the top multi-session funnel paths?  Providing inherent value to each user touch point across multiple sessions from multiple sources is a pretty advanced tactic to try and determine your perfect marketing mix. What it does do is provide more data to validate marketing spend across different ad space via the metric of “assisted conversions”.   BTW – Google makes money selling ads.

There are lots of exciting improvements and announcements coming down the pipe from Google Analytics.  Keep an eye out for some big announcements coming soon and new reporting features being rolled out in the near future.

SwellPath is a Google Analytics Certified Partner

Published September 7th, 2011 Analytics, Featured, SwellPath 3 Comments

I’m very excited to announce that SwellPath is now an official Google Analytics Certified Partner! First, big thanks to Greg for helping put the application materials together, and Charles and John for weathering the interview with Google. Thanks to our clients for giving us the opportunity to really push Google Analytics to the enterprise level. Particular thanks to InFocus and Jive Software, who we used for the Partner application case studies.

Finally, thanks to Google for recognizing our expertise. Although we could have applied for the program last year, we chose to wait until we felt like our technical and analytical expertise truly matched or surpassed the other Partners. As our clients on Google Analytics know, we are huge fans of the platform, and we look forward to having a closer relationship with Google and more insight into the future of GA.

Here is a link to our official press release in a PDF format, and here is the copy:

SwellPath approved to become a Google Analytics™ Certified Partner (GACP)
September 7, 2011

SwellPath, a web analytics services provider based in Portland, Oregon, today announced it is entering into a strategic agreement with Google to become a Google Analytics Certified Partner.

Clients working with SwellPath receive professional analytics implementation and consulting services for Google Analytics, a free, enterprise-class online measurement tool. Businesses of all sizes can work with a Certified Partner to receive professional, Google authorized training, implementation, and consulting services to make their websites and ad campaigns more effective. For more information, visit http://www.google.com/analytics/authorized_consultants.html

SwellPath offers custom, advanced and strategic measurement services to clients looking to increase their digital marketing and website performance. Whether you have an established in-house team or are just getting started, SwellPath’s experienced team applies a proven approach to web analytics for a variety of businesses.

“We are excited to become part of the GACP program. Since day 1 we’ve been helping our clients get the maximum value out of their digital efforts with Google Analytics,” said Adam Ware, SwellPath CEO, “ it is great to have Google validate our expertise and bring us into the program.”

“Google is really pushing the product by introducing features like multi-channel attribution, social sharing measurement, and custom variables,” said John Koenig, VP of Marketing & Strategy at SwellPath, “our clients are really able to understand and improve their ROI through the use of our web analytics offering and Google Analytics.

About Google Analytics
Google Analytics tells you exactly how visitors got to your site and how they use it. It can help you identify bottlenecks and gives you ideas for improvements, leaving you to concentrate on your marketing campaigns and increase your returns. Google Analytics is an advanced, easy-to-use tool. The software has all the functions that you’d expect from an advanced web analysis package. For more information, visit www.google.com/analytics

About SwellPath
SwellPath is a digital measurement and marketing agency founded in 2009. The company provides analytics and search services. SwellPath clients include Jive Software, InFocus, KEEN, and Nike Golf. SwellPath is a results-driven organization, focused on helping clients get better measurement and a better return on their digital marketing activities. For more information email info@swellpath.com, or call 503.224.9204.

 

 

SHARE ALL THE THINGS!

Published August 25th, 2011 Analytics, Industry, Social Media No Comments

Two exciting things happened last month in the social sharing space. First, Google released their Facebook killer Google+ and the now ever recognizable +1 share widget. Second, for you fellow Google analytics nerds out there, they made it easier to track all of these share widgets out there with their new social interaction tags. This got me to thinking about sharing as a concept in general. During a digital strategist meetup my good friend Alex C. Williams from Trendline Interactive and I were discussing this topic and how he was trying to increase sharing actions for his client’s email programs. What came out of this conversation was an undeniable fact, 99% of websites and emails have these features, and 99% of them were doing a pretty bad job with them.

As I started my research for this post I began to see what I can only describe as laziness. On most of the sites I visited these poor little widgets looked like they got placed haphazardly somewhere on the page. I could literally see the developer/designer checking it off the “best practices” list. If this is true, then why the hell are there literally billions of dollars being pumped into measuring how many shares, likes, re-tweets, +1’s, karma and all the other verbs people of came up with to describe one simple thought and action – “Hey, I think this content is neat and I bet others would think so too.”  Why are there not hundreds of posts about optimizing these features and functions like we do our button Call to actions (CTA’s) and form designs?

Here’s my opinion on it.

The first thing we must do is think about the placement of those widgets. I believe anywhere in the vicinity of the content you wish to share is good. They can be at the top of the article or after the last sentence of your post works to. Now, the second thing… for Christ oh-mighty please put a frickin headline by it to promote ACTION.  If you just have “share this” written by the buttons you’re better than some, but we both know that just an extra 2 minutes of copy writing could go a long way. If you stop reading this post now and go do this, I’ll have felt this post a success. If you wish to continue, I’ll give you some examples of some surprising sites who are doing this really bad and others that are doing it just right. For those of you who wish to read the whole thing, I’ll conclude with my personal thoughts on what social widgets I think are necessary and which ones you can throw in the trash.

 

Example #1 – So bad and you won’t believe what site this is.

This comes from one of the leaders in tech and internet news, and you think they would know better. The first thing we see is that the share buttons for the article are 435 pixels downward from the actual article content and then to put the cherry on it, there’s a giant rule dividing the share section. I’m sure the UX people out here would agree, at least put the damn buttons by the damn content. Without some visual cues, how am I really even know what I’m sharing? Rule number one, put these buttons in context with the actual content. The second major fail is the lack of a headline CTA. Where’s the reason I should share?  Why should I click one of these buttons? Rule number two, at least tell me what these buttons are supposed to do. This might be suprising, but this is actually a screenshot from Wired.com. All of their article pages are seriously like this. :(

 

Example #2 – How can there be so much creative thought put into this blog, but absolutely none into their sharing section?

For this article I asked my good design friend Elliot Olson what some of his favorite design blogs were, and he directed me to Fastcodesign.com. I myself was very impressed with the overall layout and elements. And then I went into a post detail page. Much disappoint ensued.  How can they be doing so many things right visually and than when it comes to sharing this beautiful site to the world, this sound is sadly appropriate. Considering the creativity of the content and overall design of the site, I feel this section should be just as cool as every other piece on this site, but it looks like it just got slapped on because there was whitespace available. See what I did there my designer readers? At least they are better than Wired’s buttons and are placed  in context of the article.

 

Example #3 – Now we’re getting somewhere.

 

I did say earlier that there were a few sites out there doing a good job and the Huffington Post is one of them. While, I would agree the amount of real estate taken in the actual article seems a bit much and the share headline is a bit weak. Take a second to look at how much utility they get out of these boxes, AWESOME! They do an amazing job with 8 ways to share, plus a one-click action email signup…bravo. Notice that this does not end up looking like a Nascar stock car like we see in some sites out there. Designers and developers take note, with a little bit spacing and background colors these pesky buttons can not only look great, but the amount of utility in such a small space can be dramatic. The placement is also great with absolutely no confusion about what article you will be sharing. Huffington post you definitely get my “SHARE ALL THE THINGS AWARD”, nice work.

 

Example #4 – My personal fave.

Now this last example might not be the best one out there, but I think it illustrates the points I’m trying to make, plus it’s on my favorite analytics blog, Occum’s Razor by Avinash Kaushik . Besides just being overly simple and to the point, it exemplifies what just 2 more minutes thinking about the “headline CTA” can do. Why yes Avinash, I did like this post. Ok, I’ll share it. See, this isn’t hard to do. His placement is also nice, right at the end of the article, no design elements, just a relevent CTA, some buttons and voila. We’ll also give extra points to for having some old school share buttons for the paper pushers out there.

I promised some thoughts on what actual share buttons you should have and what you can throw in the trash, so here it goes. No matter who you are or what your site is about have the big 3 – Facebook’s Like, Google+’s +1, and Twitter’s Re-tweet. They make it so easy to add these buttons and everyone uses at least one of these sites, so stop being lazy and do it. Now comes the hard part, what about all the rest? Is your audience b2b? I would say LinkedIn would be a smart choice. Is your content weird photos in Portland? Perhaps you should have a Stumbleupon button or maybe a reddit link (Digg is dead. Sorry Kevin.). Better yet, look at your data that you collect on your site and see if you historically have gotten referral traffic from one of these sites, if so ADD IT!!! Lastly, for those of you who use Sharethis or Addthis icons, personally I believe these are relics from when we were trying to figure out what this share thing was going to be about. Personally, the metrics I have seen on these being used on sites have been abysmal at best. Last but not least, email sharing. I’m going to get some flack for this, but honestly when was the last time you actually wrote someone an email to share some web content? That’s what I thought. Now let’s go SHARE ALL THE THINGS!

How to Measure the Effectiveness of Internal Campaigns Part 2

Published August 5th, 2011 Analytics 1 Comment

Part 2: Analyzing Your Internal Campaign Performance

Now that you have thoroughly read through my previous blog post Part 1 – How to tag internal campaigns with Google Analytics, its time to tackle the question of, “What to do with all this awesome new data?”.  This post will provide some insight on how to use the data collected on your internal marketing campaigns to measure their effectiveness and optimize your campaigns.

A good metric to start with is click through rate (CTR) of your on-site promotions.  To get this metric, take the total clicks on your tagged campaign from your event tracking report and divide this by the total pageviews of the page the promo was located on (internal campaign clicks/total pageviews of campaign page = CTR).  Make sure that your reporting periods are aligned when you pull these numbers.

The CTR will give you insight into the ability of the internal campaign to generate clicks and visitor interest.  Some factors to consider adjusting to drive up your ad’s CTR would be page position, messaging, call to action copy, creative or the page you chose to place the ad on may not be appropriate for that message.  The CTR is also a good metric to use if you are doing comparative analysis on multiple campaigns within a single page such as a homepage or custom landing page.  In this scenario, positioning within the page becomes even more relevant.  When you tagged your campaigns, you added in a positioning reference into the value making your life much easier when it came time to analyze the data.

Clicks are great but are they converting?  That is the main reason for non-branded advertising whether it’s internal or external.  Even with a branded message, your goal is to drive some kind of interaction or content consumption which can be typically be defined and measured against.  With ecommerce or B2B lead generation, it is easy to get the complete picture.

Visitors who click on your internal campaign and thus fire off an event tag, custom variable tag or both, have entered themselves into a nicely defined custom visitor segment.  Let’s use an example where the internal ad is for 50% off Troy Polamalu jersey’s at the NFL store.  The CTR would obviously be through the roof and there would be a ton of people rockin’ Polamalu jerseys thus making the world a better place.  However, a couple people might click on the ad and not actually make a purchase – losers!  How can we map the action of clicking on the internal campaign with the end conversion point of making a purchase?  This can be done through custom segmentation.  You can build an advanced segment to apply the value of the event tag to your visitor segment.

In this scenario we would build out the advanced segment to include > Event Label > Exactly Matching > Polamalu Jersey. 

Then simply apply this custom segment to your ecommerce and/or goal reports to get your conversions and conversion rate.  But what if someone clicked on the Polamalu Jersey promo then bought an Brian Urlacher jersey?  Like that would ever happen…..You can apply your custom Polamalu Jersey segment to your product performance report and see how many Polamalu jerseys (and anything else) this visitor segment purchased.

If you were exceptional with your tagging and utilized a custom variable tag for your internal campaign and set it at the visitor level, you could then measure how many visitors clicked on your internal campaign and made a purchase or converted throughout multiple sessions.  Maybe your visitors are going off-site and doing some comparison shopping but can’t find a better deal than 50% off so they return 3 days later and purchase.  The custom variable would still be attributed to that visitor and the sale can be associated to your internal campaign.  The custom variable report provides data on your goals and ecommerce within the standard report so you only need to build out a custom segment to get more granular information such as total conversions or specific products purchased.

If you internal campaign is pushing a message for a quick sale, you may find value in applying your newly built custom segment to your goal funnel visualization report.  If the campaign message is a pushing a short term conversion (Sale is Today Only!!!!), you can apply this custom segment to your goal funnel visualization report and to see how your conversion path is holding up to visitors who “should” have a strong desire to convert.  You may find that your offer is great but once visitors see the shipping price there is a large increase in abandonment.

Finally, apply your custom segment to visitor attribution reports such as organic and paid search keyword reports.  This can provide  you with insight into which keywords visitors are using to find your site that would be interested in the specific offer you are promoting.  You can scrape out some findings from the keyword reports to target some new keywords for similar promotions you may run in the future.  Also see which keywords are more likely to lead to a conversion if the visitors are provided a direct path to conversion and a compelling offer.

These are just a few examples of how to make excellent use of your properly tagged internal campaigns.  There are many other ways to slice apart the data from your campaigns and build out more detailed custom segments.  You can aggregate your findings over time across multiple campaigns to find the bigger trends and build out a framework of what a successful internal campaign looks like for your business.

Perfect Harmony – Data Discovery with turntable.fm

Published July 22nd, 2011 Analytics, Industry, News 2 Comments

 

Like some of you over the last couple of month I have found a new internet addiction…turntable.fm. Being a on and off again club DJ for the last 10 years,  I was a bit skeptical about the site, spending hours dropping my favorite tracks to see if my musical taste could serenade the electronic music snob hordes. I had a few misses and a few big wins that first night, but in the end I was hooked. Alas, this is not an opinion or how-to blog post about turntable.fm. This is a SwellPath blog post, and I’m going to focus on the one common cord with everything we do here, THE DATA. But, if you haven’t used turntable.fm and you’re interested in more details, please read this very informative post before continuing.

A few hours into my first session at turntable.fm I was utterly fascinated with the data the boys at Stickybits were collecting. Not only are they getting an obscene amount of music uploads, they asre tracking listening habits, music ratings, genre popularity and God knows what else through FB connect (I didn’t read the terms, did you?) from people around the WORLD.   The best part was all of this data was not through survey’s or prompts, the game, app or whatever this new thing is itself just produces it on it’s own. Using a beautiful mix of crowd sourcing and game mechanics, they are recording true , un-solicited feedback by a self-segmented random population. For you stats folks out there I think this is possibly the most potent and statistically relevant data you can get, right? Where is this data though? They probably have it under lock and key I bet. I can only imagine how many music related firms would love to get their hands on this data in the pure form. Luckily for people like me, there are plenty of developers out there who love to use “scraping tools”  to get  this data and are creative enough to create applications to satisfy these needs. For turntable.fm the app is called … ttdashboard.com.

TTdashboard.com is credited as being created by @alain_gilbert and I give him major props for putting a site like this together. It already brings a lot of cool real-time reporting features starting of course…with a dashboard. Like most dashboards, it shows high level data like who the top DJ’s are in terms of points and fans, the most played song (I think this feed still has some bugs), and actually a pretty accurate update of the top 6 rooms with a display of the current song and its ranking in the room.  Personally, I dig the fact that you can actually see the number of negative votes, which is not visible information in turntable.fm interface for some reason. Personally, I didn’t find the dashboard to interesting, but as I dug into the site I found the deeper pages much more fulfilling.

 

The Room Detail Page
When you dive into a room detail page, you have a long listing of the tracks that we’re played, with the likes and dislikes for songs. You also get a decent data visualization example, a trending line graph showing room occupancy over the last day. Sadly, data visualization only looks at the current day and not over the life time of the room. I think a couple of data visuals showing time peaks and visitors over the lifetime for the room would have been far more compelling. A lot of work can still be done on this page, so let’s see where it goes from here.

 

 

The Profile Page
The profile page has the least amount of functionality right now, but I think the most potential for tracking great data points.  Currently it’s only has your avatar, current fan and points total unless the system has tracked you. Again I’m not sure how this is tracked, but the ability to categorize song’s you have liked and not liked to revisit later for possible purchase would be an awesome feature. I personally have gone through hours on Turntable.fm and kicked myself later for not documenting a song I liked to look up later. I also would love the ability to easily access DJ’s who I have fanned to see what tracks they have been playing to be better prepared for my next turntable battle.

 

The Song Page
My favorite page is the song details page, which can be accessed by either typing in your favorite song into the search bar or just clicking on a song title from any page. The ability to see the aggregate of likes and dislikes for a song is sweet, but since ttdashboard.com only tracks the most popular rooms right now, the data they show is just the tip of the iceberg of what is probably available. This is where I could see showing some really great visualizations around likes and dislikes, or play frequency. I’d be interested to see some dynamic comparison graphs that show how songs fair across different rooms, in particular if certain songs are generally favored across all rooms, not just their specific genre. Again, we’ll see where this goes, my fingers are crossed.

 

 

I’m really interested in where the information on turntable.fm can take us. I can already see the potential for great global and regional visualizations based on music genres, or maybe a real time trending music graph showing the top 10 hottest songs on the site that could be adjusted by time. The amount of the global music insights you can get if turntable could get to the size of a twitter or facebook are really astounding if you think about it. Beyond this I have already tried to prompt my DJ and music producer friends to use this as a vehicle to get their songs noticed. If anybody out there knows how Mr. @alain_gilbert is pulling this info or better yet you have access to some real data from turntable,fm, I would love to hear from you and your ideas about what we could do with this amazing data. You can usually find me spinning some house music in an electronic music room, under my DJ alter ego Anton Loshe.

 

How to Measure the Effectiveness of Internal Campaigns

Published July 6th, 2011 Analytics 1 Comment

Part 1 – How to tag internal campaigns with Google Analytics

Nearly all websites have some type of internal campaign or banner ad on their homepage or other key pages.  Whether it is a homepage marquee or tout image boasting free shipping, 4th of July tent sale, free product demo, webcast signup, industry event, etc., all of these onsite banner ads or internal campaigns should be properly tagged, measured and analyzed to maximize the results and optimize your next promotion.  If SwellPath decided to promote Free Web Analytics Consulting services, we would certainly want to measure the effectiveness of this campaign to determine how many visitors want our services for free and how much revenue we were losing with each conversion.  After watching way too many episodes of Google Analytics TV with Avinash and Nick – you know you’re a nerd when… – I found that correctly tagging internal campaigns was a re-occurring topic and that lots of practitioners may be doing this incorrectly or are not sure of the best way to tag and track internal campaigns.

You may be thinking that you should use campaign tracking parameters to append your internal links and then you can view these in your campaign reports.  DO NOT DO THIS.  It is highly advised not to utilize this tracking method for internal campaigns.  You can cause problems (mixing) with your “true” campaign data by doing this and it is really not the proper use of campaign tracking parameters.  Here is an example of some bad advice in the Google Analytics help forum:


There are a few options for properly tagging your internal campaigns:  event tracking, custom variables or utilizing internal site search.

Using internal site search for campaign tracking works in a similar way as external campaign tracking.  The links would have to be appended with a query string (example: www.site-page.com/?itc=campaign-name&icd=campaign-details) and GA allows for 2 values to be passed through this method.  If you choose this option, it is ideal to setup a new profile specifically to capture this data.  This will help segment your actual site search data apart from this “fake” site search data.  All in all, this method takes a little back end setup and is using a feature in a way that is not its intended use.  Justin Cutroni has a good blog post on how to implement site search tracking if you want more details.

When tagging internal campaigns for our clients, we typically recommend event tracking, and in some cases, utilizing custom variables set at the visitor level.  Event tracking allows you to capture all clicks on your internal campaigns and offers more flexibility in your naming conventions by making use of the category, action and label values.  You can designate your internal campaigns with the category value, use the action value to differentiate banner location on the site and use the label for the creative details and in the example below, add in another positioning reference “R”.  One last element that you can place within the tag itself is the launch date of the internal campaign.  This is very helpful especially when you are making frequent changes.  Annotations in Google Analytics is another place you can note these changes.  Below is an example of an internal event tag for Free Web Analytics Consulting for SwellPath would be:

_gaq.push(['_trackEvent', 'Internal Campaigns', 'Homepage Tout', 'R - Free Web Analytics Consulting – May 2019 - 300x250']);

Depending on your unique business case, combining your internal campaigns with a visitor level custom variable can take this tracking a step further.  By relating a visitor to a specific campaign message and setting a custom variable at the visitor level, you can create a correlation with interest in your campaign to a purchase that may not occur within a single session.  The custom variable will reference a persistent cookie and thus will maintain that the visitor clicked on your campaign if they come back in a subsequent visit and convert.

When it comes time to implement the custom tagging for your internal campaigns, make sure you have your developer build the tags in a way that will dynamically pull in the campaign details into the label.  This will cut down on your development time and expenses if you have internal campaign placements that change frequently yet remain in consistent positions throughout your site.

Now that you have your internal campaigns tagged in the most ideal way, you can start collecting data and begin to analyze the effectiveness of your on-site marketing initiatives.  You will be capturing data on site position, page position, messaging and creative details.  What do you do with all this awesome data….?  My next post will shed some light on how to analyze the data and optimize your internal campaigns.

Landing Page Testing – One Page Does Not Fit All

Published June 14th, 2011 Analytics, Clients, Paid Search No Comments

Landing page optimization testing is an ideal way to improve your paid search campaigns.  Finding ways to rebuild your landing pages with stronger call to action links and better messaging can really improve your conversion rates but there may be other variables that can factor into your test results.  Recently, SwellPath developed a series of paid search landing page tests for Jive Software, a company that makes social networking software for the enterprise, with a focus on improving conversion rates for B2B leads.  I wanted to share some insight into how we built out these tests using Google Website Optimizer (GWO) and some of the high-level findings.

The first hurdle to overcome was the setup of the experiments themselves.  Traditionally in GWO, you have your original or control page you are testing against and then your variation page.  In this case, we only wanted very specific ad groups in the AdWords campaign to be directed to these experiments.  If we just used the original landing page as the control page in the experiment, all ad groups pointing to that page would be included in the experiment.  In order to only direct very specific ad groups into the experiment, we had to switch the order of the pages in the experiment and treat the variation page as the original, basically build out the experiment in reverse.  By doing that, we could then change the destination URL in AdWords for those specific ad groups to point to the variation page and no other ad groups that were pointing to the original page would be affected.

The ad groups that we used in the 3 experiments were a sampling of visitors that were most likely in different phases of the buying cycle and had different awareness levels of Jive’s products.

Experiment A: Targeted ad groups that contained keywords that were the most broad in terms of their association with Jive Software’s products.   It could also be stated that these visitors were potentially further out in the buying cycle and more in the research or exploration phase.

Experiment B: Targeted ad groups that contained keywords that were more in the middle in terms of knowledge of Jive’s products and knowledge of exactly what they were looking for.

Experiment C: Targeted ad groups that contained keywords that were the most specific of the 3 experiments, correlated more directly with Jive’s products and targeted visitors who were potentially further along in the buying cycle.

The variation page contained 3 highly targeted call to action links pointing visitors directly into a conversion point which was, in nearly all cases, a form page.  The original or control pages contained lots of body copy, 3 call to action buttons on a sidebar and some text links within the body copy.  The original pages allowed for more navigation outside of the conversion points and allowed visitors more freedom in their navigation paths.

The universal hypothesis was that pushing visitors directly toward a conversion point should ultimately increase the conversion rate.  This was only the case in Experiment C which targeted visitors that were further in the buying cycle and had the best knowledge of Jive’s products.  Experiment A generated exactly the opposite result and Experiment B was more or less an even split.  Looking at the click path data for Experiment A, it was clear that visitors were resistant to being forced towards a specific conversion point as a very high percentage clicked on the only link that pointed to the site which was simply the Jive logo.

While all 3 experiments were generally identical, the biggest influence in the resulting data was the personas of the visitors coming from the various ad groups.  The takeaway from this series of experiments is that you should consider your personas heavily as you develop your custom landing pages.  Pushing visitors who are not ready to convert into a narrow conversion path can lead to deflection.  One amazing landing page design does not always fit all.  Building out different designs for different personas and mapping those landing pages to the personas you are targeting within your AdWords campaigns can help to gain the most success from your PPC campaigns and drive up your conversion rates.

Email Updates

Signup for our newsletter to keep up with the latest from the convenience of your inbox.

Connect with Us

Find SwellPath on the socialsphere and keep up with the latest news in our world.

  • Facebook
  • Twitter
  • Flickr

411 NW Park Avenue, Suite 400
Portland, OR 97209
(503) 224-9204
info@swellpath.com