I’m also attending DIY SEM: Your Agency May Not Be the Best Resource - Gary Angel, Mark Ruzomberka, David Harrod and here’s my notes and insights on it - coming to you from my iPhone, while live blogging - but first, here’s the marketing notes:

A06 Wednesday, 1:05 - 1:55
DIY SEM: Your Agency May Not Be the Best Resource

Gary Angel, Semphonic
Mark Ruzomberka, Traffic.com
David Harrod, Traffic.com

Tracking the ROI on thousands of keywords seems overwhelming and managing them against business goals feels overwhelming. But outsourcing the task to an agency might not be your best move. Traffic.Com manages a significant ($1M annualized) search engine marketing program that is one of the primary drivers of traffic to their site. Facing competition from Google in their core business, Based on the optimization opportunities revealed by a Semphonic audit, Traffic opted to bring their program in-house and run it themselves. Mark Ruzomberka of Traffic.Com and Gary Angel of Semphonic will describe the process of an agency process review, outline the key decision-factors in bringing a PPC program in-house, and discuss Traffic.Com’s experience managing the program. You’ll get a better understanding of how to audit your existing SEM program, how to identify the optimization strategies that make sense for your business, and some of the benefits and barriers to managing your own SEM program.

By the way, as I noted in my earlier post a few days back, that Jim Sterne has made it really hard for me by putting an excellent session on Voice of the Customer Panel - John Fernandez, Lance Jones, Mark Brooks to be at the same time … but look, I can’t be in two places at one time - and sometimes, you just need to pick and choose what to focus on.

Here’s my notes and insights now:

Excellent session which pointed to a few conclusions:

1. It makes sense to take a paid search program, if only for being able to react much quickly and take advantage if buzz events.

2. Larger agencies maybe optimized for certain types of activities, they may not suit many businesses that operate with a hybrid model, or atypical business cycle (for example, if your business depends on natural disasters happening) you might be happy when events happen that are most people’s misery.

3. Certain optimizations are not worth doing, even if they are possible to do.

Semphonic just released it's Omniture Implementation Toolkit - a cookbook on how to setup and deploy Omniture SiteCatalyst - I wrote a bit about this and provided a little snippet from the report about the SiteCatalyst Data Warehouse in the Omniture Site Catalyst Implementation Toolkit from Semphonic post at The Analytics Guru.

BTW, you get 10%  off the base price with that link I provided (I could give you the link without the discount (this link), but I can't think of a good reason why you should pay 10% when you can pay 10% less.

Use sponder1008 as the promo code.

I don't think there has ever been a "cookbook" produced before to make setting up Omniture easier - pretty sure Semphonic's  Omniture Implementation Toolkit is the very instance of a set up guide for a enterprise Web Analytics platform.

Now the next question I have - when is Semphonic coming out with the Visual Sciences On-Demand Implementation Toolkit?   That's the one I need at this very moment. 

Forgot to mention that I had dinner a few nights ago with Gary Angel and Phil Kemelor of Semphonic; I invited my WAA Social Media Committee Co-chair, Elena Haliczer (who just moved here from Chicago to work for a non-profit called Games for Change).

I really enjoy meeting with Gary and we often have dinner when he comes into Manhattan, usually a couple of times a year.

One of things I wanted to bring up and ask Gary about was his amazing work with Analytics Reporting - probably the best I've ever seen - and I wanted to know how he did it, how Semphonic did it.    The post were the chart I'm showing below came from was titled More on Analytic Reporting Models - it was published a little more than a month ago:

My basic understanding of how Gary Angel created this amazing chart was this set of steps (and feel free, Gary, to comment and correct me if I left something out):

• need to meet with stakeholders, look at their data, collecting over a year's worth - find out all the possible variables that might affect any single measurement.
• All the data for each variable is pulled into a spreadsheet and a series of Visual Basic Macros are developed to take the data and create this chart of probabilities, based on the data.

We know this chart, for example, is about visits to a site - visits were going down in August and September but trending up in October … why?

Normally, an analyst would have to figure this out by looking at various reports - but with the chart above, the answers are clear, and are ranked by percentage of probability with "less" Direct Traffic causing the lions share of the traffic decline which was reversed in October by PPC Advertising buys.

Now….. the Analytics Modeling of data pretty much spelled out what someone would have had to spend hours and hours to have figured out - and may have missed details, or missed whole relationships of data.

My belief is this concept of mapping data by probabilities is where Web Analytics Reporting needs to go - and today, Semphonic is the only Web Analytics Company that can create the chart above - figured out a way to do it and the programming and insights behind it.

Now that Semphonic accomplished it - I'm sure a dozens of people could copy the concept - but I think Semphonic will always do this kind of reporting better than anyone else - as they're creative enough to have come up with it in the first place.

I've been at several Emetrics Summits where I heard mantras that Web Analysts should produce insight, not reporting.  Well …. for all that talk, few really good examples of Web Analytics Insights were ever shown ….. and this chart, above, is the best example I've seen, of Analysis, and not reporting.

If I ran a Web Analytics Group, I'd want to know how to create charts like this one for as many measurements as I could, and honestly, a data warehouse could be very useful here.  Also, I'd want to standardize, as much as possible, my reports to every part of the organization, to be prepared in the same way.

Anyway, to finish off the evening - which Gary Angel treated us all to dinner - we talked about the latest Eric Peterson - Avinash Kaushik wrestling match? and the next XChange conference (which I'll attend) sometime later this year.

I was excited as I read Gary Angel's post  on Modeling Traffic , last night, on my handheld; I was too exhausted from doing a clients' SEO work, painting and changing my hotel for the LeWeb3 trip later this week (I posted about why I might need to in My Room in Paris - is going to be clean or full of bed bugs?, earlier) to write anything much yesterday. 

Gary Angel showed reports that most of us pull - the traffic reporting stuff that often seems so pointless as it doesn't give insight or show why.  And then…he showed us some of the recent stuff (I'm sure he didn't show us everything) that Semphonic is doing to improve their reporting and came up with THIS:

This is exactly what we all need to start doing - yes, it mean having some Excel coding done, yes it is somewhat involved - but this chart is one of the most relevant things I've seen anyone in Web Analytics produce - ever - and this is what I want to start doing myself.  According to Gary:

To do this we dump all of the relevant data for the model into a working spreadsheet. Then, we create a (fairly complicated) Excel Macro that processes the data using the analytic model. It then populates the report spreadsheet with the core numbers being tracked along with the analysis of the key causal factors and counter-trends.

The beauty of this approach is two-fold. First, a good deal of analytic complexity and intelligence can be built into the model. This can prevent a decision-maker from misunderstanding or missing the contribution of key factors (like visitor loyalty) on traffic. Second, it allows the report to encapsulate all of the key information in one simple presentation that provides actual immediate and well thought-out answers to the inevitable questions.

 

What Semphonic came up with is a way to merge analysis and reporting …  I'm very excited about this - and this is, from my point of view, a very important post by Gary Angel, one everyone who does Web Analytics should read (and subscribe to his blog).

In short, this chart, encapsulates and surpasses everything else he put in the his post - and the funny thing is …. it looks simple - but it's not - really great ideas are often like that. 

xChange is getting even better - plus I'll be presenting on Second Life Metrics.

You know how things get better and better - well, every time I read about XChange - the Conference I'll be speaking at next month - it gets better and better. 

So I just read Gary Angel's last post on All-Star Game Comes to San Francisco and the XChange gets yet better, because he describes how it's going to be different than any other Conference you've ever been to.

And it' makes me glad I'll be there ….. and I decided to speak about … Virtual World Metrics … because I wanted to talk about something different than anyone else.  Besides, I have more experience with it, as a Web Analyst, and doing it for IBM's Virtual Business Center in Second Life, plus my industry contacts, than anyone else I know of.

Yes, I could have talked about Web Analytics, but everyone else is doing that …. and I want to do something no one else is doing.  I hope to have some additional help with Jared Freedman, creator of V-Tracker, a Virtual World Analytics and backend Advertising Platform - of Code4Software.com, who can describe not only what has been done both at CordWell Banker and IBM, but the coding that would be needed to do things like Geo-segment Visitors arriving in a Second Life Island, or any number of things that are now becoming possible.

To be sure, I work best with someone like Jared, who is one of the foremost developers in Virtual Worlds - who comes up with great ideas and who often listens to mine.  My strategy is more of a combination of several approaches, data gathering - and I think the blend of actual hands on - and what we've learned so far, is going to make the Second Life Presentation Unique.   I don't think you'll find anything like it anywhere else.

But your not going to hear any of this unless you at XChange - and the way to get to Xchange is with this link http://www.semphonic.com/conf.

Who said long posts are bad?  I think I did…ops!  Well.. some long posts deserve to be read all the way though - Gary Angel's post on Getting Setup to Measure Search Engine Marketing  has information that's similar to what I've come across but goes beyond what I've personally done with paid campaigns.

The main point - Web Analytics is going to report confusing information about your search traffic, if you run paid and organic at the same time, unless you can tag your paid campaign URLs with a parameter your analytics understands - so you can tell the difference between the paid and organic traffic.   I often did that with KeywordMax, which I used on several clients sites in the past.

But it gets harder than this - you also need to assign keywords to Ad Groups in a way that reflects meaning and structure to the campaign, plus… you need to pass the Ad Group in the URL to your Analytics … to get the most out of the information …and that's tough to set up if you don't know what your doing.

Not only that - but the "Long Tail" information that comes about by Broad Matching…. will be difficult to match against a specific campaign unless  you planned for it - since what you see in the keyword referral logs is the query someone actually typed in to the Search Engine …and not the Broad Matched Query that is running in your campaign(s).

So, to summarize - here's what Gary Angel says at the end of  his post (and mine):

"..

• Step #1: Insure that every target URL in your PPC Program contains a campaign identifier.
• Step #2: Make sure that the campaign identifiers are at the Ad Group Level (not higher) and that the Ad Groups are structured around meaningful concepts.
• Step #3: If your PPC program is using Broad Match, make sure that the Purchased Keyword is passed in the Target URL.
• Step #4: If your PPC program is using Content Networks, make sure they are isolated in their own Ad Groups and that you are extracting the true referrer from the URL.

I was having dinner with Gary Angel yesterday and would have asked him about the just announced partnership of Semphonic and WebAnalyticsDemystified - but I did not know about it till tonight - now. In the SEMphonic partnership and the kinds of problems you want to have … post Eric T. Peterson announces the partnership - which is a sure bet.

"…Under the terms of the partnership, clients can engage Peterson and his firm to develop a web analytics strategy and then seamlessly engage Semphonic to execute on that vision; while clients currently working with Semphonic can engage Web Analytics Demystified to help build better processes to support their web analytics efforts. For clients, this means a range of services from customized training in real-world analytics to the actual hands-on analysis of data by world-class experts. "

SemPhonic does not actually run Paid Campaigns - they do pure Analytics - teaching companies how to get the meaning behind the data; once companies come to WebAnalyticsDemystified to get the "map" of what to do next, Semphonic can take over and implement. 

I don't say any of this is easy…it's not..but implementation is probably the hardest part.  It's one thing to come up with strategy and another to actually fulfill it - often things don't go smoothly and you have to make adjustments and workarounds.

In my own experience - there's not enough people, positioned correctly, in corporations to implement Analytics as well as it could be.  For all the talk about Web Analytics - it's often incorporated after everything else is planned - and often too late.  Companies don't tend to listen that much to the Web Analysts they have.

On the other hand, when a company goes to an outside vendor, like Semphonic - they are often ready to listen - and that's the leverage that an analytics vendor, from the outside can have, that in-house web analytics often lack (along with the specific expertise needed to carry out the mission).

It's another sign the Web Analytics Industry is maturing, with businesses ready to listen to what the data says, platforms and tools in place mature enough to handle most of the data , along with the right people and processes in place to execute on all of this - that's my take on it.

Gary Angel and I had dinner last night at the Restivo Restaurant in Chelsea, Manhattan and discussed Semphonic's upcoming XChange Conference, Avanish Kaushik's new book Web Analytics: An Hour A Day.   We also talked about Compete.com's new Search Analytics package that I just got access to as a Beta Tester.

It's always a pleasure meeting with Gary, who's one of the most knowledgeable, hands on people in Web Analytics and Search today - he's able to both discuss high end topics having the first hand knowledge of how to implement. Some people are better at speaking (but don't actually practice Web Analytics that much), others, like Gary, really can do both well. 

We also discussed V-Tracker - the Second Life Web Analytics platform from www.Code4Software.com I came across recently via Jared Freedman, it's creator. Gary Angel and I discussed the capabilities of V-Tracker and what the future might hold for Web Analytics in Second Life - it seems we're at the turning point here - things can go one way or the other with Virtual Worlds - we get the metrics and, either we get the measurement processes established soon, and move forward, or we don't (and the Web Analytics Association Social Media Committee, which I direct, can help by working with the rest of the WAA to propose Social Media Standards, and this fact is discussed in the video I took at our dinner below):

I was thinking about Gary Angel's last post on Evaluating Internal Search Performance; he mentioned that certain web analytics packages will tell you what keyphrases are searched on "after" an initial internal search:

"..Surprisingly, one of the most interesting types of Search Term analysis is what happens after a Search! Some web analytics packages will do this analysis for you – providing an affinity report of Search Terms searched AFTER any given term. It’s a very cool idea. This analysis can help you with several key search optimization tasks. It can help you identify Search Terms where customization of the output may be most desirable; it can help you understand possible customer market baskets; and it can be used to help identify weaknesses in the Search Results returned by your Internal Search Engine."

Unless Coremetrics does this kind of reporting (and I have not been working with a fully functional version of Coremetrics yet, at IBM - but hopefully soon will be) - then I have not been working with an analytics package that does this kind of after keyword analysis.

But that's not exactly true, because Microsoft AdLabs Research offers such a tool - it's called Search Funnels, but it's not based on internal search - it's coming off of Microsoft Live.

I just came back Paris, France a couple of days ago and if I did an outgoing funnel on "Paris" it looks like Paris Hilton would be expected as something connected with Paris (which is kinda obvious, but the wrong context in this case) - something with the Eiffel Tower (or the Paris Riots - which is again, the wrong context) might be expected - by the searcher.

It's as if a keyphrase generates an expected complement - just as a color yellow would generate …perhaps it's opposite - purple, or a sound generates it's complement (I don't know music - I bet Gary Angel could figure out the music part better than I).

Let's take it up another step, keeping in mind that adCenter Labs Search Funnel is not as richly populated, strange as that seems, as your internal search engine - but still extremely useful for generating what searches may be expecting to find when they use a search term (that's the easiest way to explain this).

Digital Cameras seems to suggest that the landing page(s) have links to Best Buy, WalMart, EBay or Circuit City on the landing page (probably targeted to the specific sections of each site).  To me, that suggests that people, or, the Search Engine itself, has formed an association with "digital cameras" and large electronics and computer outlet brands - and not specific camera brands like Nikon, Canon, etc - and that's important to know, especially if your generating dynamic landing pages using Paid Search - because that's how you leverage search.

And yet, if you search for "digital camcorder" you get something slightly different:

With Best Buy seen as a clear leader in having digital camcorders - but once people get to Best Buy - they expect to see "Sears"…..Weird….or maybe..inciteful - at least, that's how I look at it.

A search on my favorite painter, Paul Cezanne, generates these expectations, if Microsoft Live is to be believed:

So, people who would search on Paul Cezanne probably want to see his biography, or look at other artists who were contemporaries like Van Gogh or Renoir (Dali was not really a contemporary, having risen to fame after Cezanne died).

However, if people just search on the last name of Cezanne - a different expectation is being expressed:

People who searched n "Cezanne" are comparing him to other great masters who emulated him such as Picasso and Matisse - see how subtle this thing is!

If people really want to tackle why their paid and organic search conversion rates are so bloody lousy - consider what landing pages your site search is matching up with the keyphrases being entered and if those pages actually have what people are expecting to see.

I think a good argument can be made for dynamically generating landing pages for both Paid Search (AdWords and Panama to start) and internal search (as well) because … with intelligence on what is "expected" you can more likely answer the needs of your searchers - and that's what I got out of Gary Angel's post on Evaluating Internal Search Performance.

However, building customer segments based on keyword analysis might be difficult to pull off unless you start off with some preconception of what kind of segmentation your likely to have or need beforehand - as search term segmentation can be very complex - with almost infinite granularity. 

Just as artists make often make a decision about what is going to be the foreground, middleground and the background (and what coloration makes makes up all three) the analyst has to decide what the limits of segmentation are before they start segmenting -or else they will be overwhelmed with too many segments - which is simply not workable - creating more confusion without solving the fundamental problem of conversion rate. 

Here's what Gary Angel has to say about that:

"..When this isn’t an out of the box report, you can replicate it by building a visitor segment based on using a particular keyword and then studying the other search terms used. Note that while this analysis is useful for most of the tasks mentioned above, it doesn’t necessarily capture any time sequence. It’s possible that the associated keywords were used FIRST!"

Like anything else, the quality of results you get with Web Analytics tools depends largely on the insight of the Web Analyst using the tools.