Next week, I’ll be giving a much deserved blogging break for Manoj Jasra while he does something besides work.  Anyone in the world of Web Analytics is familiar with Manoj and knows his stuff is right on. He’s very active in the practice function and frequently contributes to the Yahoo! Web Analytics Forum. For me, this is an honor, and the first time I’m blogging away from my domain since moving full tilt to Wicked Sciences from both CableOrganizer and Web Analytics Demystified.

As of today, I have 8 posts in the stables. I’m currently trying to match up which of these is useful for the stylings here and which I can cohesively integrate into Manoj’s subject preference.  From here, it looks like some work on Google Analytics and maybe some search marketing goodies.  For those of you interested, the dates for my posts are September 8, 9, and 12.

Since I’m writing tonight, I figured it might not be a bad time to also state that Wicked Sciences is teamed up with AlexDesigns here in Florida for a Web Analytics Wednesday on September 10th.  We’re also going to be present at Florida Atlantic University for the Career day that day.  The following day, we’ll be attending the SFIMA Ad Exchange event. What could all this mean?

Recently, at the Search Engine Strategies event in San Jose, California Google’s Website Optimizer Team announced the introduction of ‘Pruning’ as part of the GWO multivariate testing tool.  Billy Shih, an expert in Multivariate Testing as well as a close watcher of the Google Website Optimizer tool took note of this and immediately twitted on a great question which is brought to light by making this option available.  What does this mean to the average or intro user of a MVT platform?

To the beginners this probably does not mean much.  If an ecommerce retailer came into the experiment space recently, chances are, they’ve been getting hit pretty hard by vendors and agencies (like Wicked Sciences) looking to sell them on the idea of validity, feature richness, and insight value (monetization/conversion optimization).  Chances are they’ve heard us talk about divergence in variables, element strength, statistical validity, and confidence..at least enough to get a sense of how these principles outline much of what the testing and comparative analysis is about.  So, the people making their first considerations to go into some type of Split, A/B testing, or Multivariate Testing probably (I hope) have a sense of the impact going forward.

The experts, or at least those of us building expertise in multivariate testing as a way to optimize websites to convert better, probably have the most to gain or lose in the matter of the new features available to us.  As statistics geeks, a moniker which I wear proudly since reading ‘Super Crunchers’ (thanks Toby).., there are those of us whom care deeply about the absolute validity of our experiments.  That being the case, ‘Pruning‘, as a practice, is not a tactic which we’d openly employ unless the pruning included all current variations which have met with confidence, statistical validity, and divergence within our relative margins.

For simple marketing multivariate experimentation, ‘Pruning’ can be a wonderful addition to the toolbox.  Not having to drearily endure the most poorly performing variations is a blessing for a marketing manager whom is awaiting word to report to the administration.

Then, there are those in the middle. How I hope they are ready for this. Wicked Sciences is a firm believer in the skills and comprehension of others when it comes to our services. However, its a harsh reality that the level of understanding needed to appropriately process and apply the information presented now by Google is probably out of range for most of the non-expert testers at this time.  Unfortunately, as well, this is also a huge group.

Multivariate testing as a craze in marketing is a scary thing on its own.  Associate level multi-taskers with limited understanding of the business and its goals can easily get carried away on a tangent, not realizing that they’re still yet too new to value the experience of those people with real know-how. These are the people who constantly remind management of things like validity but had no idea how to conceive of it or calculate it in the first place.  This is the group which should be sitting back for a bit before deciding to jump right into pruning.

WebSiteOptimization.com and Andrew B. King has recently published a book through O’Reilly and O’Reilly Safari called Website Optimization: Speed, Search Engine & Conversion Rate Secrets.  The book is available through Amazon.com as well as through many other venues. Website Optimization (Secrets) is a comprehensive guide to making your website load faster, perform better, convert more, and produce more value in whatever form it takes.  Andrew and the group of additional industry experts which he chose to co-author chapters cover everything from set up, to structure, marketing and even performance metrics (which I actually presented portions of my work for Andy and Dave Artz to use in Chapter 10). Here is a list of Chapter information for interested parties:

1. Natural Search Engine Optimization
2. SEO Case Study: PhillyDentistry.com
3. Pay-per-Click Optimization
4. PPC Case Study: BodyGlove.com
5. Conversion Rate Optimization
6. Web Page Optimization
7. CSS Optimization
8. AJAX Optimization
9. Advanced Web Performance Optimization
10. Website Optimization Metrics (which also includes information on Multivariate Testing)

Having read the work, I’m actually not only impressed by the mass of the work, but the quality which was produced.   Over the past 10 months, I’ve wondered exactly how much of this was going to come together and still maintain a certain degree of readability, and, I admit, from my perspective, this exceeds everything I’d anticipated. What I’m holding in my hands is an essential guide to taking a website and turning it into a sleek machine for marketing and information distribution.  Congratulations to Andy on a job well done.

What’s more amazing is that my opinion is obviously shared by some of the top industry experts. While I’m not familiar with all the names on the list, I certainly know, or at least I am professionally acquianted with Jim Sterne and Bryan Eisenberg (who also recently authored another masterpiece in ‘Always Be Testing‘ - also available from Amazon).  Both Jim and Bryan concur with statements like ‘Website Optimization brings together the science, art, and the business of Internet marketing in a complete way’, and…’This is the first book to cover optimization from high-level concepts down to code-level details’…respectively.

Website Optimization Secrets was released last month along with a companion site and a new blog for WebsiteOptimization.com.  Each piece is a valuable source of information on the topics covered in the book. I’m sure, however, this will be a desk reference for the next phase of web developers and budding marketing optimization analysts.  Maybe, just maybe, someone will include this as a course text in the quickly multiplying offerings at top business schools. (Yes, Wharton, I mean you….)

Segmentation is an essential piece of the reporting suite for any analytics application. Finding quick and easy ways to filter information is the first step in creating a multitude of high return analysis.  A great deal of segmentation at the content level exists in being able to quickly differentiate pages relevant to analysis from the other content on a site which is less comparable .  A simple way to get ahead on segmentation of content is to install easily filtered naming conventions into your pagename structures. 

Wicked Sciences mainly provides support to companies marketing products online. As that is the case, many of our clients have  websites with a navigation system built from a very general brand identity into categories, subcategories to product detail pages and, from that point, into a checkout process.  In setting up and implementing Omniture SiteCataltyst, the idea of naming conventions is often built into how a particular site names individual pages at the title level.  Having seen quite a few implementations, we feel this is bad advice. 

Naming conventions for particular web pages should be tied into the function of the page.  For that reason, its advisable to either develop a unique meta element on which to base the page names, or, find a programatic function of your data structure to build the convention.  Some quick suggestions from this are things like breadcrumbs, or technologically identifying a stage of navigation and building naming based on that. 

For website with exactly this issue, we offer up the following helper suggestions to get you thinking about how to apply page names:

1. Any page with a buy-now button (add to cart) which is not a search results page should use the following convention:    Product Action: [PRODUCT NAME]
2. Any page which houses an exclusive brand should look something like this:   Brand Showcase: [BRAND NAME]
3. Any page which houses products from a single, broad category should say:  Category Top:  [CATEGORY NAME]
4. Any page which houses products from a specific category within a category should be:  Navigation: [CAT]:[SUBCAT]
5. Any page within the checkout process (which may have multiple stages) should be:  Checkout: [stage number][purpose]
6. Any page association with a search result should be noted as such using a universal [Search Results] filterable convention. With indexed search pages, these should also use rules to begin to establish grouping of search results pages.

 By using this method, drawing and comparing how pages are performing against each other in a fair way is much easier.  This, then, has application in normal Omniture SiteCatalyst reports, as well as in Data Warehouse, Discover, and, what is mostly valuable, the flexible Excel Client tool. 

It is important to remember that initial page naming, and how you build it, is attributed to the URL.  (Although I feel this is not necessarily a great method either) Omniture has a page naming tool in the full SC interface, but it can be problematic.  There are a few reasons for this.  First, the tool only allows one page change per server call and return. So, you can go on making changes ad nauseam one at a time, but if you have more than 20 pages on your website, you’ll be frustrated in an instant.  Second, the naming s.pagename tools inside of SiteCatalyst is not tied into the actual code or the URL, which is another piece which can go haywire. Its actually like an ‘alias’ page name that is attributed to the hard coded variable on record in the data set. 

Another tip to remember when naming pages it to choose filtering words which are not commonly used, or even better: absolutely not in use by any page name on the site.  The reason for this is that you want to be able to build filtering in one pass. In some circumstances this is going to be unavoidable, but, if you have the opportunity to make a good decision from the start, I highly suggest looking closely at how you want to dice the data when its being reported. 

 In summary, the best method to apply when naming pages is a simple system whereby you can easily segment or filter a web analytics report by one or two words.  This can be useful for quick drafting of reports on a particular facet of your user experience. Since this is done in the code, this applies to every major tool employed by your web analytics solution.  Segmenting based on these attributes is very useful for getting a good sense of contextual comparisons of website content. 

So, after 18 months of working at CableOrganizer.com, and about 9 months of blogging at Web Analytics Demtystified, I’ve fully transitioned to Wicked Business Sciences.  I am heading up efforts here to build a business geared toward adding value to ecommerce businesses.  Paul Holstein,  COO of CableOrganizer, is now providing life where I left off at WAD.  It was probably for the best too.  The tasks of managing analytics and marketing, developing Wicked Sciences, having a new baby, writing Secrets of Website Optimization with Andy King, dealing with new solutions and their measurability and patentability, speaking at eMetrics in San Francisco impeded my ability to continue to provide necessary focus on the WAD blog. 

In a nutshell, this blog is here to provide insights into solutions to problems which just seem to happen in web analytics. From time to time, we come across one of those nuggets of information that helps shed some light on an issue in collecting certain parts of multichannel business.  Or, once in a while, we might have some information you’d find useful for implementation tips, development news, or other chatter. 

I expect that this will be much more regular publication that WAD.   

A quick message to Dustin Wallace; I attempted to return comment to your statement on Paul’s blog, but, apparently that was a no-no. So, very sorry to Paul and Eric in case they felt it was inappropriate. (Maybe it was just a mistaken deletion) It was nice of Dustin to write back and let me know he picked up on the link. I’ll keep reading and I’ll be sure you’re on our blogroll here as well.

For the 100 or so people whom were reading regularly at WAD, we do have a feed which is available.  We call this the Web Analytics Research Blog. Please visit again soon.

In a recent post about Calculated Metrics, Dan mentioned an idea that pages within a path or funnel to conversion have specific purposes which can be drawn out in metrics and measured against each other contextually. For an eCommerce website, making a distinction about success prior to a user completing a sale transaction can be a challenge.  How does an analyst explain to a stakeholder or their boss that a page is successful when there’s essentially nothing to show for it?

Based on web analytics research conducted on ecommerce websites over the past couple years, Wicked Sciences developed a system to install comparative analysis at pages of equal footing.  This way, when a specific page is being analyzed for value without having a clear success metric, there is some foundation which can exist as a comparative benchmark against the other pages in the site which have the same purpose.  These are proxy measurements.  From mid 2007 to now, we’ve been using and testing proxies and their value for comparative measurement and to develop weighting systems for conversion goal testing in multivariate testing scenarios.  Most of the work which we based this on was the result of conversations with Gary Angel and Paul Legutko from Semphonic. 

The first piece of understanding how proxies work is based on understanding how we’ve functionally described the behaviors inherent to objectives and conversion events.  We essentially see this as having multiple steps, each becoming more related to the event as the depth increases.  The steps, quoted from Daniel’s post on Web Analytics Demystified, are:

Appraisal - This is the behavior of seeking information on product queue for a potential purchase. These are broad strokes in navigation based on general term and phrase usage, non-transactional focus and

• Acceleration - The point where the brain begins to move from a general information processing state to a more focused channel. During acceleration, a subject should (ideally) only move toward action and curtail further lateral navigation.
• Impulse - Having collected and been presented with one or several points where a call-to-action or option to execute an objective occurs, the confidence and slightly adrenal motivation carries a subject through an action and transition to a state of risk assessment.
• Commitment - Acting on the information and the excited state of having executed a checkpoint in a system of obligation; this measurement should seek to imply reduced regressive states when continuing through to additional actionable areas and streamlined return to transactional navigation due to acquired trust and familiarity.
• Conviction - The completion of the desired final action. Conviction should be measured by the degree to which a subject does or does not participate in building trust by gathering information as to policies, security, examining financial options, and other potentially pertinent information associated with going forward with a purchase.
• Affirmation - Presumed to be existing in both a natural and provoked state, affirmation is post-transactional and is the return to the site with an abbreviated or non-existent appraisal process. The degree to which the initial experience was positive and powerful should, hypothetically shorten time to accelerate and kindle the impulse state.

Now, having gone back into these metrics and trying to universalize them for ecommerce companies, I’ve come up with some fairly simplified versions of each so that people could begin to apply them and bring back results to discuss them.  A follow up post will decribe the functions, their application, and the formula which they are comprised of.  I’ll try to include an actual copy of the formula in Omniture so that people can copy and past them into their calculated metrics. 

Since leaving CableOrganizer, I’ve been working on trying to develop and sell the solutions which we came up with there.  I realize now how difficult a task that is.  

So, we have these patent-pending technologies.  Wicked Sciences is a reseller for the PreCognitive Search solution. This is a dynamic personalization element  available for your site which uses your current site search function to create dynamic elements related to the incoming search keyword of your referring engine. Right now, we’re looking for resellers, beta testers, and people willing to offer up data so we can refine the tool even more.  So, we have that, which presents users with clear calls-to-action in a customized include, module, zone…whathaveyou.  

Working at CableOrganizer this was a huge win.  The space that the PreCognitive Search took up was worth enormous value.  At times, it topped 10-15k per month.  So, moving to Wicked and having rights to sell this software to ecommerce companies, I thought, there’s no way to lose.  

Well, I moved on and got to thinking, there’s got to be a way to get this to people where it helps them the most and further justifies their investment in the tool. So, I started playing with numbers and looking at where marketing spend was, and I realized, AFFILIATES!!!!

So, now, with the patent-pending application, we’re going to also offer an affiliate module (like a widget or a piece for a CMS like Joomla or Wordpress) which an affiliate site can subscribe to through their sponsor.  Using that module, when a visit to a landing page on an affiliate happens from Google, MSN, or Yahoo!, an ecommerce site can serve up dynamic personalization for that affiliate.  In turn, that increases the chances a user will not only visit a retail site, but an add-to-cart call to action exists right in the object.  What’s even better, is that we can also build in web analytics so that it plugs into Omniture, Coremetrics, HitBox or Google Analytics…whatever, just to get some accountability out of the affiliate.

Now the tricky part.  I need to get this into the hands of the people who can translate what I’ve just said to the decision makers at major companies.  As of right now, we have no partnerships exclusive to this strategy….we do, however, have patent-pending on the focus technology and the business method.  I’d like anyone with an interest in testing this, measuring it, offering a trial to a client or more to contact me as soon as possible. 

Search Engine Optimization to Build Taxonomy for the Informational Organism

Getting good free advice on SEO has always been a challenge. There are a couple reasons this is true. First, this highly complex world of data, metadata, optimization, and function is tough on its own. Add to that the idea that the algorithms which guide the SERPs based on robotically collected information which is indexed on 130 million plus websites are constantly changing, and a highly-competitive canvas and you have a recipe for confusion and deception. The outcome is that when companies talk about search optimization they tend to embellish major facets of the strategy or omit key components completely.

In the interest of proving a helping hand to people who are really paying attention, here is our disclosure:

The information discussed in this post provides no guarantee as to results which you may experience by using any techniques which are described. The purpose of this post is to highlight our experiences as what SEEM to be best practices for preparing a website to be indexed by a major algorithmic search engine for inclusion in its search results. Most of what you imbibe here should be valuable information on how to THINK about SEO.

 Let it be said that Google dominates the top referring search engine visits for every web analytics report on that subject which I’ve seen to date.  Even start up companies build their SEO strategy around making their site ready for Google to pick it up in a link somewhere and structure every piece to expedite the crawl or, at the very least, include clusters of words hoping it makes a difference to the search engine.

Google, as an entity providing a service on the internet, is born from academia. This statement comes directly from the ‘Google’ page on Wikipedia:

“Google was co-founded by Larry Page and Sergey Brin while they were students at Stanford University and the company was first incorporated as a privately held company on September 7, 1998“

That being the case, and the world being the way it is, Google continues to look for relevance in our websites by promoting results with appropriate structure, cascading subjects, and a keen sense of awareness of high-authority resources for citation and interrelation.

At Wicked Sciences, we think of our approach to search engine optimization as each page is part of a large, world-wide, academic journal.  Each sentence in a paragraph is support for the thesis statement. Each paragraph is related to the current subject in the context of a larger categorical ultimately related to the great scope of the site as an organism.  This way, and while this may not be one hundred percent accurate, we are able to ensure that if nothing else, the content has subject relevance built in.

In terms of taxonomy, we try to build a site to mirror the way in which biologists classify forms of life.  The classification of a specific is based on its ability to be grouped into a collection of items with similar traits.  For example, we can take the widely read work on economics by Adam Smith called ‘The Wealth of Nations”.  We can create a web page about this book. This page may have reviews and discussion, or maybe you can buy the book.  This page is specific to the information pertaining only to that book. Other qualities of this book exist:

• Scottish Literature
• Considered Non-Fiction
• Written in 18th Century
• Hardcover
• Subjects in Economics

When all of these qualities are taken together they comprise a relatively narrow set. If we search Google for “Scottish 18th Century Non-Fiction Works on Economics” we’re met with 104,000 results of which our example comprises 20% of the top 10 in title and descriptions on the SERP.  Certainly, a portion of the 8 remaining results will have a mention of if not a page associated with this work.

 Take any one of the combining phrase words away from this search and the search suddenly becomes much more general. By simply taking away the words “on economics”, the results widen to 163,000.  Less specific = more results.  By taking away “Scottish” the number of results jumps by almost 400% to 671,000. 

Understanding Your Website as an Informational Organism

Our websites are comprised of many pages; sometimes hundreds of thousands or millions depending on the purpose, function and scope.  Those pages house content, applications, elements, and action areas which should relate to the level of specificity desired by the user.  Each individual page is, on its own, a valid source of a type of information based on particular qualities of the means in which it has been classified among its group members, and in relation to its value when compared to similar pages in other catalogs.

When sufficient value of a subject is identified and structured into the system of navigable information and meta data, it becomes visible to more users that they might find it, process the information, and eventually link back to it as a source of informaton. This, again like publishing a professional or academic journal, is akin to a APA style citation. This process should occur naturally, and quickly, with powerful content developed under the appropriate conditions. 

From time to time, slight shifts in how pages are indexed should be necessary and expected. Retailers and companies who rely heavily on search engine results page ranks love and fear these shifts. This fear should be unreasonable. 

If COMPANY A builds a website and takes all the SEO items discussed here into consideration while COMPANY B over analyzes and creates a site using some ’strategy’ they read about, the outcome of their index value as well as their future ‘Google Dance’ anxiety should be predictable.  COMPANY A is potentially slow to get going. Structure takes time to build and good content takes thinking hard about subjects and making sure you highlight the right qualities.  Money, talent, and technology can shorten this time. COMPANY B  is scraping content to build pages, stuffing keywords, and buying links.

Over three months time Google, Yahoo!, and MSN have had a chance to build out the index sitemap for producing in results pages for both companies. Where COMPANY A looks like a primary resource for information (also like academic publication index) COMPANY B looks like, at best, a secondary source of information with cognitively disrupted or disconnected data, and less than benevolent intent.  When the index is juggled, even slightly, this is going to have an impact on B; because weight of ideas might not be what they were when B was created,  while A still remains a primary source.  Granted, new developments since A might make the subject or topic more competitive, it should still be valid and valuable to a searching party. 

So, there you go. A nickel worth of free advice on helping build a site with good taxonomy for indexing by the major search engines. This should be useful to people who are going to try to build a website or are considering an overhaul of their current strategy for search engine optimization.  If you found this information valuable, consider leaving a comment below or feel free to use it as a cited source….;)

This Installation Guide is Provided For the Implementation of eVars in Omniture SiteCatalyst which are Based On Javascript On-Click Events

One of the benefits of constantly creating new approaches to web analytics work is that you get to push through the nitty-gritty details of some of the more complicated issues within commercial tool implementation.  With Omniture SiteCatalyst, the most useful items in terms of generating insightful data with regard to an eCommerce site is the eVar (the custom commerce variable employed by this report interface).  Unfortunately for many of us, the installation and use of the eVar variables and their reporting is neglected for a lack of documentation on how to appropriately apply their technology. 

 This guide to installing an eVar will explain how to (I’ve set anchors in each of the links to whisk you directly to the section if you’re stuck in a particular area):

1.  Edit Omniture SiteCatalyst Report Suites to Activate eVar Reporting
2. Apply Link Tracking Code to Collect and Report on eVars with Omniture SC H Code
3. Make Changes to the Delicate s_code.js which Apply the Data to Omniture Reports
4. Build Correlations to the Variable and Events
5. Provide some Examples of Valuable Points to Measure in eCommerce with eVars

Step 1: Edit Omniture SiteCatalyst Report Conversion Variables

The thought of editing the way in which reports can be viewed and conversion variables be allocated can scare even the most tested analyst. In that regard, you are not alone. So let’s make this short simple, sweet and to the point.  Follow these sub-steps to choose, name and configure an eVar for use in reporting:

• Login to Omniture SiteCatalyst
• Click on ‘Admin’ tab
• Choose “Select and Edit Report Suites”
• Drop down from ‘Edit Settings’ to Conversion Reports and fly-out to Conversion Variables….CLICK
• Choose an empty eVar Slot or Amend the list by selecting ‘Add New’ at the bottom right of the list and then name. 
• Make a careful decision on how you want to aggregate and report your conversion data. (Please see definitions for linear measurement, participation metrics and recency as well as expiration in our analytics dictionary). 

Step 2:  Install Link Tracking Code to Recognize onClick event for Web Analytics Tracking

Unlike Step 1, this is a little trickier than an average user understanding of HTML code, javascript and their place in the world of web analytics. This might be a poor explanation but we’re trying. So, here goes:

 Here is how the code looks:

onClick=”var s=s_gi(’reportsuiteid’); s.linkTrackVars=’eVar#’; s.eVar#=’variable name’; s.tl(this,’o',’custom link name’);”

What does this mean, let’s take it one piece at a time

• “onClick=” is the action which executes the tally
• “var s=s_gi(’reportsuiteid’);” is the reporting identification
• “s.linkTrackVars=’eVar#’;” is the allocation of the event to the eVar you specify
• “s.eVar#=’variable name’;” is the name of the eVar reported in the suite
• “s.tl(this,’o',’custom link name’);” is the allocation of the event into custom link tracking for traffic segmentation purposes

These pieces ALL must go into the <a href=”> which is on the front side of the link being tracked.

Wicked Sciences Best Practices note:  For best segmentation results in reporting, WBS suggests using general naming conventions for custom links and specific for eVar naming with regard to the type of link, its specific purpose, and possibly if it is an image or other non-text content link.

Step 3: Enable the Omniture SiteCatalyst S_CODE to Pick Up and Report the eVar

Now that you have the tracking code for the link in place, its time to make sure that the tally has someplace to go.  This requires an edit to the volatile and frightening code of the master collection javascript provided by Omniture for its SiteCatalyst interface. This is called the s_code. It should exist somewhere in your structure or on every page for internal or external calls.  The important part, for this lesson, should look like this:

 /* SiteCatalyst code version: H.6.
Copyright 1997-2006 Omniture, Inc.
More info available at http://www.omniture.com */
/* Specify the Report Suite ID(s) to track here */
var s_account=”reportsuiteid”
var s=s_gi(s_account)
/************************** CONFIG SECTION **************************/
s.trackDownloadLinks=true
s.linkDownloadFileTypes=”exe,zip,wav,mp3,mov,mpg,avi,wmv,doc,pdf,xls”
s.trackExternalLinks=true
s.linkInternalFilters=”javascript:,yourdomain.com”
s.trackInlineStats=true
s.linkLeaveQueryString=false
s.linkTrackVars=”eVar#, eVar#2″
s.linkTrackEvents=”eventX”
s.cookieDomainPeriods=”"
s.fpCookieDomainPeriods=”"
s.currencyCode=”"
s.charSet=”"

 I’ve highlighted the areas which should concern anyone installing a custom variable based on the onclick event.  To allocate tracking of the custom links as eVars, you have to mention that within the s_code.js configuration point.  Simply change this on the master javascript or on the pages where the custom track link is being presented as an option. This can be complicated and delicate, but not impossible. BP NOTE: Note that the ‘reportsuiteid’ should be reflect the same name as the one mentioned above in step 2.

With that, so long as you paid close attention and supplied values for the variables which were presented, you should have no problem gathering and reporting eVar data in your conversion reports in Omniture SiteCatalyst.  Steps 4 and 5 will be discussed in an upcoming post which can lend more clear behavioral data as well as impulses, intent, and how valuable solutions and applications are in terms of their relative return on expense. In coming weeks, I’ll come back and add images to make this extra clear. Until then, I hope this was a useful entry.