A fair question we often receive from Editors is ‘how do you ensure that your recommendations are in-step with what we would or wouldn’t do on a day-to-day basis’? Or put another way, how can Editors maintain control over the scope of the recommendations.

This question was put to Dennis at ONA’11 and here is his response:

We’ll also be LAUNCHING our new Collaborative Editorial Environment for our Front Page Recommendation Platform at the Conference, so do drop by and visit us at our booth (#15) to find out more:

My primary point is that, in any optimization activity, you must serve the overall purpose of the organization – and if so, can you honestly say that your optimization goals are KPI duplicates? and if NOT KPI duplicates, can you confirm them as KPI derivatives? And if NOT KPI derivatives, you must certainly be able to confirm them as KPI derivative proxies ?

If none of the above, then I believe it is fair to say that the further away your optimizations goals are from your organizational KPIs – the higher the risk of applying direct harm. (turning decent marketing initiatives into organizational de-optimization exercises).

My conclusion was (is) that you want to create marketing scenarios, where you can predict an organizational aligned outcome with a high level of confidence.

Find my slides below (which of course doesn’t do justice to my presentation, as the text is only pointers to the subject). Anywho, thanks to Andres, Luz, and team for a splendid day in Madrid (*even with the challenge of a power outage, mid-presentation)

Cheers :-)
/ Dennis (@dennismortensen)

So come and find me in this December version of “Where’s Waldo?” – I would love to meet up for a nerdy chat about Analytics or Media. Drinks are on me! You are most welcome to connect on Tripit (thus increasing your chance of getting that Mojito)

..
Dec. 01 – New York: Speaking and Socializing at Stratigent YWA Analytics event

Dec. 03 – New York: Speaking about Bootstrapping vs. Venture Funding at ULS

Dec. 07 – Chicago: Speaking on Analytics & Attribution at SES / (my comments)

Dec. 08 – Chicago: SES Authors Book Signings Event (12:00 in the exhibit hall)

Dec. 10 – San Francisco: Customer Meetings
Dec. 10 – Sunnyvale: back to Yahoo! home base

Dec. 11 – Mountain view:  Speaking on Metrics for Success at SDForum
..

Cheers :-)
/ Dennis (@dennismortensen)

Join Stratigent and Yahoo! Analytics for an evening of insights, cocktails and networking at the Yahoo offices in New York City. Learn how enterprise organizations are gaining new customer insights by leveraging Yahoo! Web Analytics as a highly customizable web analytics tool. Uncover hidden ROI as Bill Bruno of Stratigent shares best practices for deriving maximum value from your analytics program. This exclusive event is hosted by Stratigent, an official Yahoo! Web Analytics Consulting Network member in conjunction with Yahoo! Web Analytics.

Highlights include:

• The Future of Analytics: Discover how online marking measurement and optimization is evolving and learn what it takes to keep your business ahead of the curve
• Best Practices for Deriving Value from Analytics: Bill Bruno, VP of Business Development and Technology at Stratigent.
• Yahoo! Web Analytics Product Demo: Presented by Dennis R. Mortensen, Director of Data Insights at Yahoo! and author of Data-Driven Insights with Yahoo! Web Analytics
  
• Networking and Cocktails: Connect with analytics professionals in the area and speak with the Analytics experts of Stratigent and Yahoo!

Registration to this invite-only event is complimentary. To request an invitation (while seats remain) contact Jenny Kimbley via email: jenny.kimbley@stratigent.com.

Anywho..

Happy Thanksgiving. :-)
/ Dennis (@dennismortensen)

“Cut to the chase! Use analytics tools to get the specific answers you need about your search marketing campaign’s economic performance, your users’ on-site behaviors, and how to look for major red flags in traffic patterns. This slate of experts will keep you focused rather than poring through hundreds of pages of meaningless statistics.”

I know this is the usual promotional event pitch, and that’s OK. The easy response (presentation) would have been a focus around the magic, one can pull out of behavioral data in conjunction with search data. And that would be OK as well. I was, however, reading an article from Jacques Warren [1], who as a web analyst is as smart as they come, and during the post he concludes the following:

“Google organic accounted for 65% of all visits, 89,5% in December, 87% in January, and 84% so far in February. That’s only one search engine (and its various properties)! In a word, we LOVE Google.” [2]

I decided to use that as my outset and I in particular focused on what I bolded above (the capital letters are his). Before I move on, let me be clear that this is not a bashing of Mr. Warren, by no means. His post, and specifically the above comment, in combination with my talk in Chicago, just happened to trigger something in me, to finally utter my concern around the unfair relationship between content owners and search engines. This includes, what I believe to be a set of unhealthy search success metrics, or for some people, even a naïve belief in a search engine friendship.

Any other person, might just agree on how fantastic the traffic influx from Google is, in the above example, and continue to thrive on the euphoria of that – and perhaps even apply additional search engine optimization processes across multiple search engines and aggregators in regards to new content. But doesn’t this seem wrong to you, that we celebrate yet another winner of the Google organic search lottery ?

If I ran a business where most my revenue, if perhaps not all of it, depended on visitors to my site, I would be very dissatisfied to the extent of concerned by the fact that my life was in the hands of a random search engine.

To marginally illustrate my point, but mostly for us to continue the debate, let’s have a look at Figure 1 and Figure 2, and the apparent question that’s comes with those two data sets, such as; which of the below two trends would you describe as more successful (forget all about the usual ‘it depends’ and just assume that all the stars are aligned to your satisfaction).

My point is that, in the scenario of monetization being most successful around returning visitors (as we see with a lot of content owners), figure 2 is the more successful trend. AND the accompanying suggestion could very clearly be that by the introduction of Search Engine de-Optimization one might be able to force such a pattern. One action could actually be to remove some of your content from search engines all together! Drastic, yes, and this is certainly not for everybody, but think about it twice, before you conclude that I am completely crazy. :-)

Conclusion
It seems fair to debate whether or not search engines and other content aggregators extract too much the webs value, leaving less for the content creators (originators) – if this is true, one should introduce tactics such as Search Engine de-Optimization to destress the dependency!

I’ll prepare myself for the bashing and first hit in this blog post – it might mentally prepare me for Chicago. So there you go! Dear search engine, I don’t Love you anymore. – and it’s not you, it’s me that changed ;-)

Cheers :-)
/ Dennis (@dennismortensen)

[1] If you are coming to Chicago Jacques, Diet Cokes are on me! :-)
[2] You can replace Google with any other content aggregator and the above critique still rings true

You can go to the original post here: http://widgetanalytics.wordpress.com/2009/07/08/the-analytics-food-chain-panel-from-omma-metrics-measurement/

I embedded the video below; for your entertainment:

Panel: Monster, Yahoo!, Razorfish, Hitwise and Neo@Ogilvy.

Cheers :-)
/ Dennis Mortensen (@dennismortensen)

I suggest, as an alternative, that you try using a Page Revenue Participation metric for your next Conversion Optimization routine, as it might reveal opportunities away from your traditional funnel. When debating this, please note that, I am purely talking about Controlled On site Content, as defined by the Online Business Measurement Quadrant.

The reason for brining this up; I only just returned from Boston, where I attended the Internet Retailer conference, and probably even more exciting the Yahoo! Merchant Summit, which was delivered in conjunction with the show (On that note, the ystoreblog folks did a respectable pre-show interview about YWA 9.5 and my book). Anywho; the merchant owners had some of the most honest analytics questions I’ve heard in a long time. You simply cannot wow these folks with a posh looking four-dimensional bubble chart – they wanted to know exactly HOW we are going to help them make more money!!) – Refreshing!

A small disclaimer; all of the below screen-shots are from Yahoo! Web Analytics, but you have the same metric available in e.g. Omniture Site Catalyst and even though Google Analytics have a different term ($ index) and slightly different calculation, it is the same optimization attitude.

So how do I use a Page Revenue Participation metric for my next Conversion Optimization routine ?

1.

Go find your most popular pages report, which in most tools, means a report that shows a sorted list of all your pages, with the one receiving the most page views on top. By most standards, this report does not provide much insight, unless you derive your revenue from page views (such as advertising) – fair enough, then this might not be a bad proxy for $money.

2.

Customize this particular report in a way, so that you get your tools revenue participation metric appended to it. In Yahoo! Web Analytics the Revenue Participation metric resides under the sales group under metrics, when using the Custom Report wizard.

3.

The below figure, shows a standard most popular pages report, which have had the revenue participation metric added to it. This is almost per default, and with out much fingering, a better choice than just looking at raw page view numbers.

4.

Fingering or not! I am a big fan of visualizing data and the below screen shot is essentially just an advert for data visualization (but also, potentially, a  honest report export, that will create better value than a non colored bar chart). The below bar chart was extended to 100 bars and had its bars colored by the revenue participation metric. This alone brings instant insight, as we can spot easily spot “small green bars” – or more precisely, low page view volume combined with high revenue participation (a.k.a. important pages). We could of course expand this with more rows of data or just sift through the bars a hundred at a time.

5.

What I really like for you to do, is to change the sorting from page views to revenue participation. This changes the picture completely and it is the essence of my suggestion for you trying out a Page Revenue Participation metric for your next Conversion Optimization routine. This should in a best case scenario bring forth a set of pages, that you might not see as active drivers for conversion today.

The last X steps in your funnels, which are required for people to check out, will of course always be on top, and if you are anal (like me), I suggest you go filter them away, so you only have pure content in this report. Especially if you have to communicate this beyond yourself or your close team.

The below shows a astonishing set of facts:

• The “About us” Page Touched 23% of revenue in < 1% of visits
• Site Search participates in about 50% of all my revenue.

Knowing these two facts (which will be completely different from site to site) I hope one would set out to work on the “About us” pages AND closely look at those internal site search result pages. For these folks, their business clearly depend on it.

6.

As a closing comment; not that the upper commentary need further input, it is obvious that neither “About us” or “Search” is part of your traditional funnel analysis as below – and as such we sometimes forget to look outside our funnel.

Doing a better “About us” page, in this example, even though it is not part of the funnel, is likely to widen the funnel – and if that is not exciting I don’t know what it.

This is not the first time I talked about revenue participation per say and if you are still listening, the post about how you can Use Web Analytics to determine the width of your Internal Search Query box might be worth a read.

Cheers :-)
/ Dennis Mortensen (@dennismortensen)

What is the current state of web analytics tools and their usage today? – and to what extent should we expect today to provide us comfort for the future.

What is the current state of web analytics tools and their usage today? – and to what extent should we expect today to provide us comfort for the future.

That is a grand questions and I am not suggesting that I hold all the answers. I do however believe that this is something we, the Web Analytics software vendors, need to ask ourselves on behalf of the entire community from time to time.

I suggest and advocate excellence on today’s current 3 steps of Web Analytics usage:

Today

1. Collect Data
2. Report on Data
3. Insight from Data

I don’t think any of these primary tasks have been solved to the complete satisfaction of our users yet, and I am confident in speaking on behalf of all web analytics software vendors on this. Something as basic as collecting data, is a huge challenge, especially if you think of every one of the Online Business Measurement Quadrants being important to your online endeavors. Reporting on data is far from being solved as well, and providing people a login and password to the analytics interface, is in most cases more harmful than good. Finally, I don’t think we can honestly say that we get all the insight we expect from our data just yet. However; I do believe that all of these point are well understood and that we are making great progress.

That said; we must look towards tomorrow and apply new (or evolutionary) thinking to what we do today. I personally advocate the next two steps clustered around the following:

Future

4. Recommendation
5. Automation

Don’t get to hung up on the words themselves, that’s not the important part. When I talk about recommendation, I want you to envision us, the analytics vendors (and this might not be your traditional vendor), to apply our own modeling on your data, coming up with recommendations, for what you should do next. When I talk about Automation, I believe that you should be able to have your analytics technology (or API like offspring) act upon recommendations without your direct interference.

This is all part of moving you towards the true data driven organization, where we input the objective and let data speak for how we get there – which is where I supposed you are headed? We do have players that do some parts of the above, but not to the degree that I can confidently say that we started these steps.

I am the International Keynote at the Web Analytics Congress in Amsterdam on 28th May 2009. Where I will present and debate the above, which is something I am really looking forward to. I’ll be using the below set of slides to get to my point:

..So if you happen to be in Europe>Amsterdam, please drop by and say hello, and perhaps get the one hour version of the above slide set, which without me, as you can see, doesn’t say much.

AND finally; what are your step 4 and 5?   – I’m sure we can’t all agree on this. :-)

Cheers
dennis (@dennismortensen)

n.b.
If this subject is of your interest, I strongly suggest that you go read: http://www.nettakeaway.com/tp/article/377/what-web-analytics-is-missing

I am walking up to the Hilton now, so If you are in town for Search Engines Strategies New York 2009, do come by the Yahoo! booth and say hi! – I’ll be around most of Tuesday and Wednesday.. And I am of course super eager to talk about anything to do with Analytics and online marketing.

I’ll be presenting YWA at the Yahoo! Agency day on Thursday, including some sneak peeks of  the upcoming version. So if you are a SEM agency and have not signed up, go have a chat with your account Manager.

I also noticed that Wiley advertised my new analytics book in the SES programme, so perhaps you can go pre-order it. AND THATS something we would want, right!? eh :-)

Throw me an email, twitter message or reply to this post if I am not around.

Cheers
dennis.