* Sigh * As long as I can remember, spring and summer have been the time for (but not limited to) barbecues, swimming and the effects of semi-surprise search algorithm updates on our clients. Recently, we’ve seen Google tinker under the hood, announcing the Mayday and (long advertised) Caffeine algorithm changes. Yahoo has been making various mods as well.

To the benefit of our readers, we’ve put together a client case study that looks at measurement techniques for evaluating site performance through Google and Yahoo’s algo’ changes. Some precursor notes: for this particular study, we pulled conversion data using the surfing single analytics segments technique. The featured site is a small but very profitable trade publication in Google News.  Also, there was a major push to improve organic conversion in preparation for the Caffeine update. That adds for more intrigue to some cool reports further down the page.

Global Traffic April 1 – July
Doing well…wonder where that traffic’s coming from… It’s always important to have a look at overall traffic and where the visitors come from.

All Conversion
Here we have all of the conversion sources.  Remember we’re going to be looking at the effects of organic algorithm changes. Still, it’s cool to take note of how organic fares against PPC. Don’t worry, we know this is basic yet. Trust there are going to be some pretty sexy reports coming up in this post.

Traffic Effect On Yahoo Organic Through Yahoo Algorithm Change
Wow, we actually gained organic keyword traffic through the change. The red dot is where the algo’ shift took place.

Conversion Effect Yahoo Organic, Through Yahoo Algorithm Change.
How fun is this! In reality, we prescribed and the client executed a major organic conversion effort. It worked so well the organic change at Yahoo barely even mattered.

Bing Organic Traffic
The traffic trend is pretty similar that of Yahoo. We’re watching Bing because, well, sooner or later Bing will become Yahoo.

This is the same view by the week as opposed to by the month.

Bing Conversion
This data tells us straight-up that traffic from Bing does not convert. However this may be a red herring. Run the percentages. The percentage of overall organic search traffic for Bing is still low because it’s still Bing, although Bing is growing.

The Effect of MayDay & Caffeine on Organic Google Keyword Traffic
Interestingly enough, the downturn in traffic here was only a slight one for this client. The green line is the previous 30 day period of organic keyword traffic. The blue line is the current 30 day period.

Reduced PPC Expenditure
In the overall profile of this site, we look at the client’s 3o over 30 PPC investment. The client ratcheted down PPC spend because conversions now come from organic.  How cool is THAT!

Progression of PPC Conversion
Also, the reduced PPC expenditure has not hurt because PPC conversion is up. Hey, this stuff looks good.

Progression of Organic Conversion, Even Across Caffeine
Again, the organic conversion initiative success obscures the Mayday and Caffeine updates.

Something interesting to note is that the peaking, highly-seasonal search pattern for this client’s industry is in fall and early winter. This new data speaks to how the algorithm updates enabled site traffic and conversion performance, resulting in a stark difference from the normal ebb of seasonal traffic.

For further info on the three recent algorithm updates, check out some of the resources below:

Yahoo! Update
Yahoo! Search Algorithm Moves Towards Links & Authority Sites – SEObook, Aaron Wall
March Weather Report: Yahoo! Search Update – Yahoo! Blog, Dan Rampton
April Weather Report: Yahoo! Search Update – Yahoo! Blog, Dan Rampton

Mayday
Google Confirms “Mayday” Update Impacts Long-tail Traffic – SearchEngineLand, Vanessa Fox
Whiteboard Friday – Google’s May Day Update & What It Means for You – SEOmoz, Scott Willoughby

Caffeine
Our New Search Index: Caffeine – The Official Google Blog, Carrie Grimes
Matt Cutts Explains Caffeine Update – WebProNews

The main selling points of Omniture are engine-independence and paid support. Conceptually, there is something to be said about using a third-party tool to manage your spend across multiple engines. At the same time, the primary engines have formats (API and bulk loading) that make it easy to export and import campaigns.

In this regard, the gap that used to be there a couple years ago is now closed– and then there are cheap tools like SpeedPPC that quickly multiply and distribute campaigns across multiple engines. Omniture’s SearchCenter integration with the platform is in marketing only– true product integration is a long ways off and not likely, given that it’s already been a couple years in the making (data warehouse export is still not working), and that Adobe bought Omniture.

• Google Analytics doesn’t offer paid support. While there are a few GAC consultants that Google authorizes (they have to pass a test for certification– harder than the AdWords qualification), and there are some PPC/Analytics training sessions sponsored by Google– there is definitely not the kind of support/consulting you get from buying software. Many corporations prefer the idea of paid software and are comfortable with that model of having an account manager they can call up with questions, plus a block of services hours for implementation and report development. That said, you might ask Omniture about their support offering and ask a few clients that are peers of Global Sources to inquire about the support they get.
• Campaign reporting. GA allows for multiple dimensions– in particular, motion charts, advanced segments, and the various multi-dimensional views that are metric specific. I believe Omniture is inferior in this regard because of GA’s ability to visualize data in cross-tab (pivot) and related view formats. In other words, it’s easier to uncover trends in GA than by hunting through Omniture.
• Integration. This is usually a red herring across most Internet marketing companies. The most important integration in analytics and PPC software is that of CRM interaction (salesforce.com, Eloqua, Microsoft Dynamics, SugarCRM, etc..) and offline conversions. This almost always requires some custom work, since every company has a different underlying data model (which they should), as well as a different sales funnel and attribution scheme. The collection, integration, and weighting of this data is not an out-of-the-box software module, but an exercise of sophisticated marketing analytics. Online conversion tracking is relatively simple for all enterprise-level analytics tools, whether using a method like Google URL Builder or cookie tracking. Google has a significant advantage in tracking activity from and on Facebook, despite the marketing efforts put forth by Omniture.
• Funnel tracking. Omniture does allow for multiple paths. Our viewpoint is that the more sophisticated method is to measure event-level attribution (page or click), rather than force the analyst (you) to have to define each particular path to analyze. The traditional methods of slice-and-dice is a needle in a haystack approach — you should prefer your analytics tool to do the legwork to tell you what combinations of pages lead to a conversion or a poor user experience. We are not aware of any clickstream analytics tool that does this out of the box. With the number of combinations of attributes events, and pages possible, you need click level data and a correlation algorithm to pull out the right combination of trends to view. You cannot do this out of GA yourself, because you’ll need the raw data to calculate. That said, log file parsing is probably the most practical solution here if you want to go that far in analytics, given that Omniture doesn’t know how to do it (we’ve had multiple calls with their top people and they are stumped).
• User tagging. Omniture does allow for more variables to be stored. You’ll want to consider what use cases you have that cannot be solved via an advanced segment and parsing urls. If you’re interested in Omniture’s solution, please read the chapter in their implementation guide — it’s a confusing read, but they do allow collection of personal data. Google doesn’t allow collection of such data for privacy reasons. Not sure about auditing requirements — any certification of data accuracy would have to rely upon click-level data out of your logs, which Google can’t do (unless you have the old urchin, which is not recommended).
• Goal tracking. Google has recently expanded from 4 goals to 20 goals. Most companies misuse goal-setting, as they confuse segments and points within the funnel as goals. The more goals you have, the more complex the attribution. It’s hard enough to do attribution when you have only 1 goal and many events for which you have to allocate credit — now try matrix attribution with many goals and many events. To the best of our knowledge, almost nobody has single goal attribution down, so matrix attribution is not even in the vernacular of analytics yet.
• Page overlays. Cool tool with wow factors for both GA and Omniture– but usually not usable because of tracking problems and multiple links on a page that have the same URL. On the latter, let’s say that on a particular page, there are two links to get to another page (a topnav and a footer nav link)– if they have the same destination url, you won’t be able to tell which one drove the click. We have rarely found the visual overlays to offer accurate data.
• Data freshness. Generally a 2-10 hour delay on Google. Data freshness is most important is when you have events that require real-time optimization. Keep in mind that PPC data may be on a full day lag and you’re limited by your weakest link. Thus, if your web analytics is only 30 minutes behind, but your PPC and CRM are 4 hours behind, you’re really 4 hours behind (or you’re making inaccurate decisions). Further, the concept of statistical significance is such that you have to gather enough data to determine what’s going on. At Yahoo!, we decided that a 3 day reporting delay (because we needed 2.5 days to crunch attribution) was worth the trade-off in speed versus effective optimization. You’ll have to decide what data you really need at what frequency.
• Independence. Several of the government agencies we have talked to don’t use Google Analytics because open source is considered off-limits. Some major advertisers don’t use GA because of the potential conflict of interest in having your analytics being tracked with the place you spend your money. And there are the “tin foil hat” and anti-monopoly people that in general don’t believe you should have your analytics, PPC, landing page testing, mail, and so forth with the same company. Given practical realities, we don’t think this is an issue right now.
• Effectiveness. Google AdWords is going to have more effective (effective meaning increasing profits, as opposed to allowing you to create more reports) tools than 3rd party tools– they have to, because they have the advantage of more data. Case in point– the Conversion Optimizer of Google versus any bid management tools. With the exception of folks like ClickEquations (market leader who is good, but not great), in our opinion, nobody yet has a sophisticated method of bid management.

Omniture vs. Google Analytics comparison

Another potential plus for Omniture is that they’re releasing tools to manage Facebook PPC campaigns via their API.

Whether these tools are robust and take advantage of social targeting is yet to be seen, given that advertisers who blindly copy their PPC ads to Facebook don’t meet with success. They are also interacting with Facebook solely via the API, which is quite limited versus what’s available in the web interface or other third party tools that communicate by other methods.

Net-net, you can do mountains of comparison and research on web analytics tools. There is plenty to be said about the GA vs. Omniture debate. Our advice is to start from your goals and then decide, as opposed to what most people do, which is to create a wish list of every possible feature they might ever use and see who has the most checkboxes. That approach will lead you to the wrong solution, as most of the nice-to-have features you’ll never use.

The act of defining your goals, optimizations to directly improve such goals, and the reports to diagnose when you’ll take such optimizations is a re-statement of the process that BlitzLocal uses in running campaigns— that is metrics > analysis > action. If there are any metrics/reports that are not actionable, they are a waste of time. Metrics are about what has changed significantly since last time, analytics is about why, and action is what you’re going to do about it.

Dennis Yu is CEO of BlitzLocal, an Internet marketing firm specializing in Facebook and PPC campaigns.  He is an international speaker and author.

Opinions expressed in the article are those of the guest author and not necessarily those of aimClear LLC or aimClear Blog.

The luminous Anne Kennedy, Founder & MP of Beyond Ink, FP & CMO Joblr.net, not to mention SES Advisory Board member, moderated the panel of authorities on Eye Tracking Research consisting of Shari Thurow, Founder & SEO Director of Omni Marketing Interactive, Jeremi Karnell, Co-Founder and President of One to One Interactive and Susan Weinschenk, Chief of User Experience Strategy at Human Factors International.

First to speak was Shari Thurow. She addressed eye tracking in relation to search engines as well as its weaknesses.

Next was Karnell who started his presentation with a question: “What is this?”

It’s the fleur de lis, but as a symbol it represents many different things to different people:

• French monarchy
• The Saints football team (I don’t know about you, but that’s what I said… )
• Boy Scouts
• Chevy Corvette, etc…

Symbols mean a lot of things and people have instinctual reactions to them. This is exactly why brands use them.

One to One Insights developed a study for eye tracking, here Karnell revealed the objectives, study, insights and their conclusions.

Objectives

1. Determine user behavior and engagement with SERPs vs. text only.
2. Asses impact.
3. Identify potential impacts on SEM campaigns.
4. Gather preliminary data to form a larger study.

Study

1. Recruited 17 subjects who wanted to change their cable provider.
2. Established 10 keyword sets- six branded, four unbranded.
3. Created four SERPs for each keyword = 40 permutations.
4. Created natural, natural with paid, universal, universal with paid.
5. Collected eye tracking.
6. Distributed two surveys (Geneva Emotion Wheel & Likert Ranking Scale).
7. Asked: ?What link would you click first? Second?”

Insights
Universal SERPs that included mixed results

• Video was typically fixated on before the first natural and right paid links (17%)
• Video links received more clicks than top pd link
• Image links receive more fist fixations

Universal SERPs and the Golden Triangle

• With text only: F pattern.
• Many scrolled below fold.
• Universal Search = golden triangle. Visual density stayed at the top.
• Video and image results increase engagement on SERPs.
• Universal only and universal with paid had significantly higher levels of engagement.
• People believe images and engage as a result.

Conclusions

1. Creating a holistic stat that considers all search interaction will create competitive advantage and drive more engaged visits through SEM activities.
2. Understanding your brand does have an impact on search environment.
3. Think about how your digital assets can be leveraged. Be sure to use appropriate tagging.

Weinschenk, the brain lady, spoke about the psychological implications one must take into consideration when using eye tracking studies.

Humans have an estimated 40 million sensory inputs coming into our brains every second! Wowza!

BUT, we are only consciously aware of up to 40. Most sensory processing is happening unconsciously.

Three Parts of Brain

• New brain (Conscious)
• Mid brain (Process emotions & visual info process)
• Old brain (Instincts. Our animal instincts are constantly scanning environment asking: Can I eat it? Can I have sex with it? Will it kill me?)

How can we get at the old mind?

7 Eye Tracking Traps to Avoid:

1. Underestimating the effect of what you ask people to do on where they look.

• Be careful what you’re asking them. Look for this/that/the other thing … or just let them do it on their own
• i.e.: the Yarbus Visitor Study

2. Assuming that where people are looking is what they are paying attention to.

• New research by Larson 2009: tested peripheral vision vs. central vision. Eye tracking uses central, but peripheral is very important (which is why anything blinking is annoying). The theory is that this is from our evolution
• The study showed deteriorated images of rooms. If it was deteriorated in the center, people could still identify the room. If it was deteriorated in the peripheral people could not identify the room
• Conclusion: periphery is important!

3. Underestimating the effects your protocol.

• If the test subject is sitting too close to screen or too far back, this will throw off data.
• These studies must be  exacting!

4. Underestimating time needed to analyze data. Analysis paralysis .

• You get a LOT of data (yuck!)

5. Underestimating time/cost to perform the study.

6. Overwhelming client with the data.

7. Neglecting to draw useful, meaningful decisions and actions from the study.

“You can use all the quantitative data you can get, but you still have to distrust it and use your own intelligence and judgment” –Alvin Toffler

There you have it, a nice piece of heat map meat, just take it with a grain of salt.

photo credit: jennaddenda

Bottom line? Utilizing and understanding the analytic data behind your website traffic is absolutely essential if you plan on making money. The Introduction To Analytics session at Search Engine Strategies New York left the audience with a depth of courage to explore Google Analytics (GA), five elements of a CEO-worthy report, the best KPIs to measure and incorporating other analtyic data and tools from diverse sources. 

Moderator, Jeff Ferguson, SES Advisory Board & Senior Director, Online Marketing, Local.com introduced the solo speaker John Marshall, SES Advisory Board & CTO of Market Motive. Marshall ended up in Online Marketing by accident and hated the primitive tools for the early web years so he started a little known company called ClickTracks and developed the Overlay Report to see where people click on your site. (Psst! You can find it now in Google Analytics)

I Feel Your Pain… (but quit whining!)

• If only we had more reports.
• If only the data was more accurate.
• If only IT could tag the pages.
• If only I didn’t NEED to tag the pages.

Sorry, there is no one-size-fits-all when it comes to web analytics. (But, that’s a good thing!)

5 Elements of a Great Analytics Report:

1. Key Performance Indicators (KPIs)
2. Segments
3. Context
4. Other sources of data
5. Bring it all together

These are not a prominent and for a reason… so you can custom tailor analytics to your own business!

Element #1 – KPIs

• Key Performance Indicator (fyi “hits” are pretty much useless)
• CEO should understand them
• 3-5 works best
• Traffic? (no, no, no)

Marshall’s Favorite KPIs

• Bounce Rate
• Conversion rate
• Average time on site
• Task completion rate
• Share of search
• Profit per visitor

The last 3 are a little more complicated… Not much insight in GA

Measure Task Rate- use a 4Q survey, they are quick, simple and very insightful.

• What is the purpose of your visit to our website?
• Were you able to complete your task?
• If you were not able to complete your task, why not?

Surveys tell you why things happen.

• Web analytics can only tell you what happend
• Give users the chance to enter open text

Share of Search

If traffic is rising slower than the rate of searches, um, well, you’re traffic is actually going down because it’s not rising as much as search. So, yeah, sorry… re-cork the wine and save it for another day.

Profit Per Visitor

• Beware of ROI (cost of goods! DOH! Google Analytics has no clue about your COG)

Element #2 – Segmentation

• What it’s not
• What it might be, depending how you look at things (especially if you come from traditional direct marketing)
• Segment based on intent, buying v. researching (Don’t disregard those who exit without converting)
• Segment your exits (ie hotels: people who are interested in weddings probably aren’t booking online…)

Google Analytics – Advanced Segments – Jackpot!

Analytic Segments Abound!

5 Segment Ideas:

• New v. Returning
• Keywords: brand v. product

Element #3 – Context

• Annotations (right under the graph)

• Persuasive copy with supporting facts (some people do actually like to read)

Element #4 – Multiple Sources of Data

• Not everything is in GA
• Social Media (Not easy to measure, commercial tools are best: TrackUr or Radian6)
• RSS (aka your blog – key metric = number of subscribers. You should be using Feedburner, like, now)
• Video Playback (If you host your own, Flowplayer has a GA plugin, but you need to call up IT to hook you up)
• Event Tracking
• YouTube Insights
• AB Testing

Element #5 – Bring It All Together

• Make reports brief and insightful
• Use facts not opinions
• Scannable text with details
• Data from multiple sources
• Stitched together in one single e-mail
• Significant human effort is required (get over it and allocate resources)
• Lovingly hand-tooled works best

Ok, so you’ve spent painstaking hours generating beautiful analytics reports, what happens to them?

The Agency

• Uses reports to explain economic impact of events
• And the consequences of inaction

Bonus Element: Read any book by Avinash Kaushik

Spend Money on Training, Augment with Free Resources

• Web Analytics Association
• Web Analtyics Wednesday real life social networking/commiserating for web analysts
• Twitter: #measure

Marshall’s Top Tops

• More reports wont help
• Decide which analytics work best for your business – for entertainment, decision making, job justification
• You need help from IT
• Install WASP to Firefox to view any website’s analytics tags (stealth!)
• CLICK AROUND! You can’t break anything in GA
• Offline Conversion? Use time on site, it’s a solid indicator

At first gape, these questions seem easy, yes?  “Of course, yes, socialize everything” you shout!”  “After all, isn’t the ‘net  all about social media these days?” How could it be wrong to ever consider omitting the all-powerful and ubiquitous social media click-me-now badges?

Caveat:  We understand that as social-trust augments and even partially supplants trust rank in how content is ranked by search engines, having influential tweeps and and wielding one’s own street cred’ is of paramount importance. This article is not about whether social media participation is important. Of course it’s important.

However when herding traffic to social KPIs, there can be implications in how other crucial KPIs are impacted,  serious analytic concerns and business decisions at hand.  There’s no one size fits all.  While it’s often a good decision to drive traffic from landing pages and email blasts with “follow me/friend me”  links, there a number of issues to consider–especially when socialization is not the primary KPI.  For the sake of clarity, let’s define secondary KPI:

“Secondary KPI” is an additional goal a) if the primary KPI is not reached or b) in addition to the primary goal.

Analytics & Outbound Links Exiting Directly to [Community]
(Third party social assets like Facebook, Twitter, YouTube, etc..)

There are a few analytic problems to sort out, so we’re all on the same page. So far as tracking socialization as either a primary or secondary KPI directly from an email, how do you measure success exactly?  What are viable metrics to track the outbound click (from email or landing page ) directly to [Community]

• First, we all know It’s fairly easy to track an outbound click directly to [Community] from an landing page or email. Google Analytics does not track exit clicks, in the form of destination URL, out of the box. It can be accomplished with a hack.  Most email platform analytics track outbound clicks and destinations by various methods. If technology is lacking, you can always tag outbound links to [Community] and use a script on a domain you track using redirects. Unfortunately tracking the outbound link is not enough data.
• Once users get to  [Community], unless you own [Community], you can’t directly associate clicks from that email (or landing page) with behavior in  [Community] like fanning, grouping, sharing, signing up for an app’, any desired action, etc…How do you measure the KPI?
• We must depend on less empirical, though somewhat effective correlation methods to track the ROI of direct links to [Community] from emails and landing pages.
• We like this formula: (Exits to [Community] Per Day) Graphed to (Signups/Follows/other actions Per Day). We need to see a correlation of escalating traffic and Signup/Follows/Actions. Good testing can make this, otherwise speculative metric, useful. So far as the report all I personally need to see is lines lines moving from left to right on a graph.

• Graphic email exit maps show how traffic bleed and spray effects conversion. Similarly analytic overlays illustrate how users click out from an HMTL page.  Show us on a chart  please. Does the traffic forsake the commercial funnel for the social click? Is that the desired action for the page? Are we bleeding conversions?

• 13 different exit options are nice in a pretty HTML email, but doesn’t one of the paths have more value unless the email is about general branding? If not, should one path  be designated as the primary KPI?  We market for a reason usually. Sometimes good marketing is about removing customers’ options. Herd them like sheep we say.
• If selling or hard-core marketing is the KPI, we often advise that client’s insure that “being social” is always the secondary KPI, so long you can prove it does not distract from the primary conversion. Sometimes putting the links to [Community] should be on the thank-you.php page or on the next page in sequence IF users don’t convert.
• Sometimes it makes sense to sacrifice conversions for happy social BFF pals, but make that decision intentional, data driven and based on real business priorities.
• Never underestimate the value of conversations and making friends. To this author’s mind, socialization should nearly always be, at minimum, a secondary KPI.

Our General Rules:

• I usually want to see traffic I pay for go to a landing on a site I control first, even if it’s from a little Facebook button in an email or landing page.  I don’t own Facebook and ultimately we can’t control it, only work in harmony. The exception is when socializing is the primary KPI. In that scenario it can make sense to cut down on click-count to primary conversion. All the normal caveats regarding the limitations of tracking behavior in third party sites apply.
• Should there be Facebook buttons in every email? Right! Of course there should be…That’s easy to say right? Not…I.M.H.O. sometimes we should place buttons, even much of the time… but not all the time.
• When we do place buttons, consider herding traffic to our feed or landing page, which prompt  FB savvy visitors to login using FacebookConnect.  Then they’re already logged in with a few clicks.
• Sometimes email blasts and landing pages are rather guerrilla in their intent, with immediate, even difficult financial objectives. Saying that we should “always” point them to Facebook like saying there should “always” be PPC,  “always” be Press Releases or always be in Forums, etc…That’s not a mature approach as these sorts of absolutes ignore other business considerations.
• Is the real question here is what are you doing with the Facebook friends you already have? Are they engaged? What’s do their offpage engagement metrics look like? What percentage of buzz surrounding certain content is on site, on page, off site, where, how often, when, who…and how influential is that “who” guy anyway?

Not many humans are social 24/7, nor should marketers be. We’ve even had times where we recommended a nice tight little email/landing page funnel that, instead of spraying traffic in general (albeit good) directions all over the Internet, we supply fewer options to focus.  Honestly dude, sometimes we just don’t have time, inclination or money to be social .

Can’t Squeeze Social Blood
This is especially true when a marketer is engaging in activities which barely have associated analogies in physical life.  For instance, there are no gigantic “Laundromat” analogy-applications on Facebook because real people actually tend to be disengaged and separate when washing clothes in public. Go figure…

Social Media Mirrors Physical life. As I think about it, there are very few analogies in physical life that line up with “Download $2 Walmart Coupon.”  Ok, maybe some guy wearing sandwich boards or a leaflet temp’ in Times Square. Sure their occupation is noble but we don’t ever end up…umm….hanging out, hooking up, you get it.

Coupon clipping in physical life is cool but humans don’t tend to physically congregate around the theme. Does this mean that it’s not a good idea to connect with customers at any given time…like while purveying coupons? Nope. It’s usually a great idea to make friends.

Should marketers attempt to socialize every single impression in any channel? Or, are there sometimes compelling reasons to avoid impacting the conversion funnel? There are so many channels, it’s hard to imagine that it’s always appropriate to place those link to Facebook and Twitter.

Don’t get me wrong. I’m a huge advocate and participant in social media.  Our agency has a humbling roster of SMO (social media optimization) clients, who inspire the hell out of us. SMO is crucial, especially at the intersection SEO.

One thing for sure, there’s “no one-size-fits-all” policy.  Let’s test and push the social envelope. Let’s always remember to place our customers in a well laid funnel, ask for the money and be careful where we fragment the objectives, for any reason.

However, you can make money from your website when properly advised & optimized  by your web analytics, as Jim Sterne so correctly pointed out at SES Chicago 2009. Many of us have access to some sort of analytics for web properties we’re involved with, but how many of us really use the data effectively?  More importantly, are we making real money from analytics insight?  Is there more to life than just the conversion?

The session How To Turn Your Web Analytics Into a Money Making Machine was totally educational, not just because it provided answers, but because it proferred the types of questions we should all ask when pulling up our analytics dashboard over morning tea. 

Moderating this session was Richard Zwicky . Speaking first was Bryan Eisenberg, SES advisory board manager.

Oh My God! Bad Analysis Killed Kenny
Bryan says there aren’t enough people in the world that know what to do with analytics & on top of that, we have a data diarrhea problem. Before Google analytics, not everyone had access to good data.  This is no longer the case. The problem is what do you do with it?

The first extreme people fall into is looking at analytics (even daily reports) & then just filing them without action. The other extreme is from looking so deep into the data that they’re completely removed from revenue.

Bryan says that analytics works because of process. It’s about continuous improvement (kaizen maybe?) Most people can plan, most people can measure, but not everyone can improve. You can’t make money by just looking at reports.

Improvement comes down to:
Budget – Even it you dedicated 5 hours a month of your personal time on things you can to do optimize – you will do better.
People - You need talented people, creative resources to pool.
Culture – Adopt the policy of being a data driven company. For obvious reasons, this can be harder for large companies.

There are three steps for making real improvements based on analytics insight:
1. Every time you see reports, you need a to – do list. What marketing efforts or parts of your site have challenges? What do you think can improve these things? What sort of things do you want to test? Look at your competitor’s efforts and test. Decide what efforts you do more of & what efforts you do less of.

Then you prioritize your improvements & recommendations.  You have to prioritize based on resources and impact. If you only have 5 hours a month and you have something that will take 20 hours, don’t do it.

2. Segment your way out of sadness.  Average metrics product average results – they don’t given you any kind of valuable information. Why? If one of your goals is to improve conversion rate – break down the number of people in your audience – do you treat repeat customers different from new customers? How do you treat people different later vs sooner in the buying cycle?

What do you do with this information? Create personas to classify your potential customers. They have different preferences. If you need to, create little stories around them to relate to these people. Each one of these personas is going through a different journey through the conversion points you send them through. It’s unfair to think everyone who’s never heard of you before wants to come to your website and sleep with you.

Then, look at basic segments in Google. There’s different conversion rates based on different visitors. Maybe users need different information. After you realize this -  that each person is part of a different segment, then prioritize which personas are the most important.

3. Always be Testing! (not a plug, Bryan swears). A/B and multivariate testing is classic, but you have to do other tests. Usability tests are important. Explore new marketing efforts; play with new things to see if they have traction or not.

Consider low cost user testing tools like 5 Second Test & Feedback Army (among several others). You can have people do usability testing of your pages with attached audio commentary for peanuts basically.

If these don’t work for you – Pray!!!

Next up was Phil Mui from Google Analytics. Phil said he admired the other panelists when he was even in grammar school

Making Money with Google Analytics
Most of us have a 2% conversion rate, how do you measure the success of the rest of the 98%? Conversion is such an important KPI, that our world looks like keyword -> website -> conversion. Is optimizing for conversion the main thing you want to do?  Mui says it may not be the right KPI.

Imagine if  Sears only cared about conversion rate at their retail stores. They could give every sales person a knife & force people to buy by knifepoint. Their conversions will go up , but is that really what they want?

Look beyond keywords – do you also optimize creatives. What about landing pages? What are you doing to optimize your ROI at each of these customer touch points?

1st Suggestion: Optimize Holistically
You can use Google analytics at every touch point. Look at creative, using your favorite search tool to test different creatives, then apply unique tracking parameters.

A set of best practices to use when optimizing creatives: There needs to be a strong call to action. Special offers pique attention. Delivery details generally help clickthrough rates. Also, inserting actual prices. Dynamic keyword insertion is generally a best practice. The percentage of capitalization in creatives helps it feel more personal. In the URL word length – keep it 4-6. No exclamation points.

Landing Page & Website Optimizations: You don’t know how effective a landing page is until you test alternatives against it. Google Website optimizer allows you to do two kinds of web/landing page optimization The first type if A/B testing. Alternatively you can use multivariate testing. Break up the page into multiple variable parts and then test them in combination. You can view combination performance in Google Web Optimizer reports.

2nd Suggestion- Always test (just like Byan said)
Most sites have a few “existential KPI’s” – reasons why a site exists. These are your macro conversions.
For a video sharing website it could be # of uploads. Blogs it could be # of subscriptions. Ask “Why are people engaged with the content” “what influenced them”, & “why did they bounce?” Then start measuring successes with micro conversions – little wins at every part of the site.

3rd Suggestions – Analytics Intelligence
One new Google analytics feature is analytics “intelligence.”  These are  proactive insights & will say something like” On this particular day, please look at these particular metrics.”

4th Suggestion: Be vigilant – know the unknown unknowns

Closing the session was Jim Sterne from the Web Analytics association & the Metrics Summit.

3 Tips for turning your website into a money making machine
Tip #1  Optimize that basket - Amazon has done a great job at testing their shopping cart process. Their magic is the 1-click order button. Jim says to focus on optimizing a specific process, what about just the cart process? Users have to add to cart, review, change quantity, shipping, billing, confirming etc. just to get to the all important thank you page.

Where are the problems here? This is what Web analytics are really good at. Are people leaving at shipping page before billing? We have to do classic testing – usability testing or the survey. Analytics will tell you where the problem is and which problem is most significant at the moment. Test and measure and oh by the way – occasionally look at your page.

Tip #2  Determine Visit Value - If I’m spending a lot of money on a keyword or a banner ad or PR – how do you determine the value of a visit? Jim likes to assign a point system:

Time on Site – interesting but not terribly useful. With time on site, it might be “I’m gonna go to your website but then go eat lunch or go on the phone etc. 0-5 points
Page views – cool you went other places on the site. 1-10 points
Events – did you do some activity to prove engagement 5-20 points
Personally identifiable information – Did I get your email address or phone number. 25-50 points
Purchase – did you buy something, worth a lot but not necessarily the most? – 100 points
What’s the Cart Value – Did you spend 5$ versus $500. 200-400 points
Profitability – did you buy the most profitable thing. 400 -600 points

Ultimately, you want to invest in the sources that yield the highest values.

Tip 3#: Determine attribution:
How are these things working in concert? Assist attribution. Compare your quantity of key phrase searches vs conversion. What if you only focus on your high conversion traffic. It could drop off. Your high search volume stuff might be assisting your high converting keywords (excellent food for thought).

Jim then gave us a taste of his upcoming Social Media Metrics book due in 2010:

• Reach – How many people could hear my message. If it’s on a city street, it’s a billboard.

• Frequency – How often are people talking about me? If discussion increases, that’s good. There’s no such thing as bad publicity

• Influence – It’s great to get a blog mention until your realize only the blog owner’s mom and maybe his dog read it.  Who has the most influence? They will help you with your reach.

• Sentiment – Having a lot of people talking about you a lot is great, but you still have to monitor that &  the tools aren’t there yet.

• Outcomes – The actual conversion; did they download the whitepaper, join the discussion, did they buy the product?

Q & A Session Selected Questions:
Q: Any chance in the future that Google analytics data will pass through the rankings.

Mui: We don’t comment on future features. I can say that every year Google analytics is doing better.

Q: What tools do you like to couple with analytics to complement and augment your understanding?

Eisenberg: Usertesting.com. I’m also  a screenshot-aholic, I like grabbing screenshots from competitor’s sites. Personal surveys are great tools, there’s also really cool search tools. Look up “69 free search tools to improve your website”.

Mui: How many of you use insights for search? How many of you are using ad planner? Search based keyword tools? If you have not used these, these will help you gain search intelligence, plan your campaign & understand your search traffic over time.

Hop into the YouTube analytics suite Insight and you’ll notice YouTube has added some new features to the platform. Obvious additions are pretty new color charts allowing us  to view discovery of videos over time. Being able to monitor view-count over the lifespan of a video and tracking the sources of those views is nothing new to Insight. However merging these two measurements is. The new feature allows us to toggle between stats in a cool stacked chart or line chart display.

Also new on the grid is a sweet Mobile Views feature. YouTube stated in recent months that viewership from mobile phones is in the tens of millions each day and phone uploads have shot up 1700% making this new feature highly substantiated. No doubt, this is a reflection of the dramatic increase of smart phone adoption. Be sure to check this metric out. You’ll be surprised the percentage of mobile views your videos are receiving.

Ever wonder how many of your subscribers are coming to check out your content once you hit upload? The Views from Subscribers feature allows account owners to see exactly what subscription modules are garnering views. Although  gathered stats will only show from September 09′ and forward, it’s helpful to see if subs are coming either from the homepage subscriptions box, subscription page and/or subscription emails.

YouTube also hinted at “more releases” soon to come, but made no specific statement regarding what we can expect. For now I’ll settle for crossing a few line items off the ol’ YouTube analytics feature wishlist.

When a video is tailored to a particular audience to reach a specified objective, measuring more granular demographic data about views and viewers can be incredibly powerful. Therefore understanding YouTube’s built in analytics, Insight, is imperative. Let’s go on an inside tour of YouTube Insight analytics must-know-fundamentals.

Deeper questions tend to arise at this point. “Who is my audience and how exactly are they discovering my videos?”  “Does my content keep viewers attention from beginning to end or are they bailing out midway through?”  “Which of my videos are receiving the most ratings, comments and favorites?”  “Which related videos is my content showing next to?” Pretty heady stuff…

Enter YouTube’s analytics tool, Insight. This free tool is available to any YouTube account holder who has uploaded video content to their channel. Take a look at what data is just there for the taking:

Date Range
Changing the date range allows you to view content performance across certain parts of video life in each Insight category.

Views
Find out exactly how many views your videos are getting as a whole or individually as a percentage of total views.

See view count of your videos in different regions across the globe.

Check the ‘Show unique users’ box in the ‘Views’ tab to see the number of unique views in relation to total views. The difference between total and unique views maybe a reflection of the content’s popularity among users. People’s tendency to watch a video multiple times in one sitting may mean they are sharing with friends or just plain can’t get enough of it.

Discovery
One of my favorite areas of Insight, the discovery section reveals the various sources of traffic to your content. Find the spectrum of keywords and phrases that are driving views as well as related videos your content is appearing next to. Also available is the percentage of total views received by embedded players, external links, Google searches, viral means and other YouTube pages.

Hot Spots
Audience attention is recorded to illustrate a video’s rate of retention in relation to videos of comparable length. High points in the graph may reflect parts of the video that viewers are rewinding to see again. Experiment with catchy annotations designed to preview a particular part of the video in order to keep the viewer engaged (ex. wait to see what happens at the next!). Then monitor any shifts in audience attention that may have come as a result of the message.

Call-to-Action
Formerly only available to non-profit organizations, call-to-action overlays allow you to place an external link over your video thereby driving traffic to your channel profile or website. Find the number of clicks your call-to-action overlays have received for a particular video in the call-to-action tab.

Demographics
Insight relays age range and gender of your viewers either for a specific video or across all published content.

Community Engagement
The community engagement tab shows the number of comments, ratings and favorites your videos have received as well as the percentage of engagement in different regions of the world.

Pulling Data
Download all or segments of the gathered data into a .csv reports for an even deeper and more detailed look into video performance.

Ensuring  your video reaches the right audience is a demanding and ongoing process. Luckily YouTube has provided several different metrics with which to gain a better understanding of how your content is being viewed. It’s important to understand that the metrics gathered are just that, metrics. The true benefit comes from using the data to understand changes and opportunities that may be inferred within the pretty graphs and percentages.

Web analytics data provides an abundance of insight that can ultimately increase revenue for a company. The problem is many aren’t up to the task of deciphering the metrics to really understand making associated reports pay. Today’s  SES Toronto session titled “Analytics for Search: ROI, Engagement, Attribution, and More” was just loaded, with a great panel which shared many actionable recommendations to attendees.

Moderating the session was SearchEngineStrategies Advisory Board & Principal Andrew Goodman (Page Zero Media).  Speaking on the panel were Richard Zwicky (Enquisite), Anne Marie Lorriman (Outrider), June Li (ClickInsight) & Bryan Eisenberg (Future Now, Inc.).

First to take the podium was Bryan Eisenberg. His presentation will cover major issues in web analytics at a high level. He reminisces of at time when tracking analytics meant using log files and parsing them in excel. It was a very painful time in the late 90′s. The main goal of analytics has always been the same, to gather an idea of if we have achieved success? Are we winning the game?

These days a lot of people are talking about attribution. Attribution is a tricky thing; you’re probably getting bias answers. Google has been found to be over-reporting search visits. Attribution is a critical thing. What keyword was actually responsible for the conversion? So many time a user will begin their purchase process by looking for information on the product or service with a particular keyword or phrase. Further down the buying process they may enter a completely different phrase to get back to the site and complete the purchase.

Bryan gives the example of being the parent of a kid on a basket ball team. The first player checks the ball in, the second passes down the lane, the third sets up a pass to your child who dishes the ball to center where the player dunks. As the parent of the child, who do you want to give credit to? Who deserves credit for the score? In the game, the guy who dunks gets the credit. The goal of attribution is to identify the assists in the buying process. He shares an example of a client who found a keyword that wasn’t converting and cut it off. After that sales dropped 30%, because it was a crucial keyword in converting further back in the buying process.

What do we need to know?
Google likes to report on the last click or the person that last touched the ball. There has to be a solution that will show you the all time conversion process.  The objective is to get a better idea of the visitors needs and addressing it. Once you do that you can begin correlating what works and what doesn’t.

Problem with People Using Analytics
Astonishingly, out of 79% of people gathering information on their site, only 30% make the changes needed to garner better results. So many people suffer from “data diarrhea,” they find it easier to discard the information then to make the changes. People are forgetting to do the basics of analytics, hoping they can toss the ball all the way down the court and get that 360 dunk right away. You need to figure out how to make changes on the turn of a dime, every single day. Companies that can do this are the ones that are most successful.

What is the Secret?
Have a continuous improvement process. You need to have the ongoing process. The companies that have had success were the ones that made small steps to improve the site continuously (Dell, Intuit, Walmart, Xerox). Ask what do  you need in order to have an organization set up for marketing optimization

3 Key resources
1st – People
Marketing (the ones in charge of making business decisions), analysts (can point you to challenging areas that need to be addressed and areas of opportunity), graphic design, copywriter, creative resources, & a little bit of technical

2^nd – A Process
Identify who you’re trying to talk to, what action you want them to take & what they need to take that action?

3^rd – Tools
The analytics information you used to have to pay for in the past is free today. Get good at free before you go out and purchase a big analytics package.

Increasing conversion is like compounding interest in a bank.  If you improve conversion every month by 3% over the course of a year you will have improved your conversion rate by 42%.

Next up was June Li. She begins by stating that converting data to dollars is what we want. If you don’t have money in the bank, you’re missing the bottom line. She breaks her presentation into suggestions for those that extract analytic reports themselves and those that will need to ask a team for reports and changes.

Data and reports don’t make you money you have to take that information and analyze it.

Data to dollars Conversion Funnel
Collect Data & Reports -> Ask Questions -> Segmentation Analysis -> Take Action -> $$$

It’s important to realize you are the person that makes the data meaningful, not the tools. Hammers and nails don’t build houses people do. The same is true with web analytics tools. Get in on the web design early in the process. All too often the campaign is set up and running and then the team will call for SEO and SEM. If you’re able to be there from the design conception, address what extra tagging or scripting will be necessary and define goals for the future performance of the site.

Once you have the data, now you need to segment. Ask yourself what is different between those that are converting and those that aren’t.

Drive segmentation with questions as simple as what’s working & what’s not working? Do more of what is working and do less of what’s not working. Look at how visitors are converting and compare that to those that aren’t. That sounds easy enough.

Analyze the data and fix the right problem. What you want is a de-bottle necked funnel with good persuasion and good conversion. Sub-optimial conversion can result from – un-targeted promotion attracting the wrong people – good targeting persuasion ineffective conversion – good persuasion through the purchase funnel, but people are leaving right before conversion.

Drive Action with Analysis
Work backwards from the “conversion” or “value event”
- Are there sources that drive more conversions
- What are the keyword or phrases that drive more traffic, what does that tell you about audience you are tracking.
- What days of the week have higher conversions
- What content did those that converted access (pages, downloads, etc.)

If there’s NO ACTION, there’s NO DOLLARS. You have to take action. Get in on the design process!

Andrew goodman – do what you can with the free tools. Custom advanced segments on each page, Google analytics is getting better.

Next to bat was Anne Marie Lorriman. Her presentation focused on what happens when paid targeting fails and how to diagnose and prescribe a remedy. It all starts with targeting the right keywords.

Issues to Diagnose
1 – Offside search result
2 – Hidden defect search result

Offside Search Result
She asks the audience, what does “sonic” mean to you? The problem is this word can have several different products associated with it – DVD publishing company, brand of tooth brush, a restaurant or a video game character. Keywords will inevitably mean different things to different people. You have to regularly go back and review your how keywords are being targeted in your campaign.

Say your trying to target the keyword “car” in the context of a purchase (ie. “buying a car”). Broad match is not a bad strategy but it requires appropriate attention.  The keyword “car” has several different contexts in which people search for it, buying, repairing, selling, etc.  Without filtering out these different irrelevant contexts you’re campaigns may fair poorly. Incorporate negative keywords into your campaigns to drive down high impressions and low click through, thereby increasing your quality score. Negative keywords may reduce the amount of ad impressions, but ultimately your click through rate (CTR) will increase as you hone in on what your targeted audience is looking for.

Prescription
Step 1: run a search query report
Step 2: analyze the report and create negative keywords for those that are not applicable

Hidden Defect Search Result
After filling your offside prescription, look back at your search query report for each campaign to discover the hidden defect search result. Although you may have honed in on the intention of the search, the spider doesn’t realize “car” to be the same thing as “automobile,” this can drive down quality score because of low query relevance. Separate each keyword into their own campaigns: vehicle, car, automobile and auto into their own campaign. Fill these different campaigns with negative keywords of the existing campaign keyword names.

Last to take the podium was Richard Zwicky. He makes note that although we as search marketers know what the marketplace looks like as to paid and organic, it’s astonishing how many customers out there do not. Looking at how most individuals spend time in the search engine result pages (SERPs), you find that there is very strong focus on organic results. How strong? It’s been reported that 88% of traffic from SERPs is through organic listings. Is an organic refferal worth as much as paid referral?  The problem is that it’s much easier to assign a value to paid rather than organic because paid can be directly tied to keywords and other metrics.

Advertising spend in the past year on organic was $1.4 billion, yet accounted for 88% of traffic. Paid advertising on the other hand had a $39 billion piece of total ad spend and only accounted for 12% of traffic. Organic provides so much value but because it’s difficult to track, it’s often overlooked.

There are countless actions to take in order to gain your share of the organic pie. Richard lays out some very specific things to consider. If you want to be recognized for the specific geographical location your company serves, you need to focus on links that reflect that location. If you don’t segment out value, you’ll never see your opportunities. Keep thinking about how you can move placement for keywords. Keep striving to identify what you’ve missed and what else is out there that can drive value to your business.

Analytic data can seem complex and daunting at times, but it’s ability to effect the bottom line makes it an area of search that can’t be overlooked.

All over the world, SEM industry bellwether Twitter’s rocking with chatter surrounding the long awaited public beta release of the Google Analytics API. That’s not surprising given implications for the search industry.

Classically associated with paid search (PPC) applications and keyword research, get ready for more mainstream search marketing API tools to impact methodology and applications by which we publish content and measure organic SERPs (SEO).

Throngs of drooling developers can now mash data from Google’s ubiquitous free analytics package with AdWords, Trellian, LinkScape, Twitter and other public and pay APIs. The practice of blended data by API has already become the norm. Things are only going to get more exciting.

It’s a brave new world of “ToolStrings.” Tech team Witch doctors mashup custom chains of term extractors, crawlers, keyword inventory reports, non-personalized rank checkers, search analytics and other metrics, all delivered by mainstream and boutique APIs.

In coming days and weeks, webmasters will certainly be wiring Google Analytics data to enterprise content management systems to “advise” various tagging and linking schemas by keyword, page, and many other performance metrics.

The implications of the burgeoning API universe, to the world of undertaking  actionable user behavior measurements will make SEO, paid search and social media applications much smarter. Bravo Google.

Here are resources packed with technical information, industry news and  bloggers’ reactions regarding the new Google Analytics Beta API.

From Google

• Attention Developers: Google Analytics now has an API! GoogleCodeBlog
• Google Analytics (Labs) Developers Guide
• Google Analytics Data API-JavaScript Library Google Analytics Labs
• Google Analytics API code.google.com
• Google Analytics Data API-Java Library Google Analytics Labs
• Attention Developers: Google Analytics API Launched! GoogleAnalyticsBlog
• google-analytics-api This official group for all Google Analytics related APIs.
• google-analytics-api-notify This is the official group for future Google Analytics API announcement

News and Blogs

• Google Analytics Opens API SearchEngineLand
• Attention Developers: google analytics api Launched! Webmaster world
• Google Analytics API Now In Public Beta, Desktop Reporting Takes Stats Offline TechCrunch
  SEOs Excited For Google Analytics API SEO RoundTable
• New Google Analytics API / DataSource! Debuggable
• Juiced Google Analytics Python API Juice Analytics
• Google Analytics API Released – Now Get Your Web Site Metrics via Code ProgramableWeb
• Google Analytics Gets an API ReadWriteWeb
• Google feeds Analytics API to world+dog The Register
• Google Analytics Launches API SearchEngineWatch
• Business impact of Google Analytics API Web1Marketing
• Google Launches API for Google Analytics WebProNews
• New API: Google Analytics Your Way HTML Goodies
• New API: Google Analytics Your Way Internet News
• Google launches API for Analytics platform Hudson Horizons
• Google Analytics API, Create Standalone Apps Website Magazine
• Introducing Garb: Export the Google Analytics Data API with Ruby Extend
• Google Analytics API Goes Public Slashdot
• Google Releases Google Analytics API Kikabink News
• Business impact of Google Analytics API Internet Marketing Blog
• Web Analytics Facts » Google Analytics API launched Web Analytics Facts
• Why the Google Analytics API is good for the industry and Webtrends
• Take your analytics anywhere using the Google Analytics API Econsultancy
• New API: Google Analytics Your Way Digital Media Buzz
• Log into the Google Analytics API using PHP and CURL using Username/Password Authentication
• Google Analytics API Launched The Eponymous Pickle
• Google Analytics API Launched MediaPost