Web Analytics Memes

You may have read Ronny’s paper or heard Avinash’s talk about the HiPPO - (Highest Paid Person’s Opinion), but there is more to the story than a fancy acronym for Africa’s most dangerous animal - and your meeting’s biggest foe.

It has become analyst lore because of the great image that it imparts and because of the message it delivers. I was there when this term was created - December 2005 - in an office in San Diego.

The term came about when trying to make a decision, and the HiPPO said that in the absence of anything else, the highest paid opinion in the room got to make the decision.

This is a very powerful statement from the very person who is going to make the decision - “In the absence of data and analysis, HiPPOs rule”

Avinash points out in this post that:

“Very few people, HiPPO’s included, can argue with a customer’s voice, the customer afterall is the queen / king”

And he is absolutely correct, but it goes beyond that - HiPPOs thrive if there is no available information to help make a decision.

Think about what data you would need in order to make a decision - get the data, do the analysis, and figure out the story to tell to help everyone understand the customer.

Make it a point to deliver data with insights and opinions to help create action within your organization and help reduce the amount of HiPPO sightings.

Each week, I get to read the Monday Morning Memo from Roy H. Williams and think a bit outside of my box. Often the memo is interesting, and sometimes it is so relevant that I feel beat about the head neck and chest area with the insights.

Today’s memo was full of insight, but the extra nugget in the rabbit hole held something even more powerful. The new idea is that if you trigger adrenaline you impact the brain and enable learning to happen faster. Roy writes:

“The adrenaline of emotion is the key to teaching, training, and branding. Long-term, automatic recall can’t be created without it.

Give people joy when you can. Make them curious if possible. But the only thing worse than ads which anger the public are ads that bore them blind.”

The same can be said for analysis, stories, actionable recommendations, or messages that you want people to remember - trigger emotions to aid in procedural memory.

After I run, I am reading the book Plato and a Platypus walk into a bar…Understanding Philosophy Through Jokes. It reads like the Marx brothers teaching a class on Philosophy - really quite good if you like that sort of thing.

From the book:

“Inductive logic reasons from particular instances to general theories and is the method to confirm scientific theories.”

What follows is a great joke about Sherlock Holmes and Watson - it is on page 30. In the end it breaks inductive logic into the following steps:

1. Observe some event.
2. Using your intuition, create a hypothesis to explain the observation.
3. Test the hypothesis by ruling out alternative hypotheses.
4. If your original hypothesis holds up, then you might be right.

Plato and the Platypus is much more accessible than this Stanford article, but there is a great inductive reasoning article at Wikipedia that includes this sentence:

“Induction is sometimes framed as reasoning about the future from the past, but in its broadest sense it involves reaching conclusions about unobserved things on the basis of what has been observed.”

As analysts, we observe the data that is collected by our tools, make hypotheses about what happened, eliminate alternative hypotheses, and provide recommendations that will result in “optimal” website performance in the future.

I thought we were Web Analysts when it turns out we are Practicing Philosophers.

On my to read list is The Black Swan which is based on the idea that we place too much weight on the odds that past events will repeat, and that really important events are rare and unpredictable.

What are you reading?

Forrester recently released research that indicated that firms use Wikis mostly for knowledge management.

Other uses included collaboration and project management - which seem like other interesting uses. I was speaking with an MBA student recently and he mentioned that most of his group projects are done through a wiki like interface so that collaboration and editing could occur quicker - brilliant!!

Recently WikiMindMap was released that creates mind maps from wiki articles - here is the result for web analytics. This looks like a great way to “see” how information is grouped. I can’t wait until it can be applied to other wikis!

The beauty of the mind map, is that the diagram shows context of the central theme as it relates to other themes or ideas - it is this relationship with other themes that defines the concept.

Web analytics without the right amount of context makes it almost unusable - so think about web analytics in terms of the context that you are analyzing data. Mind maps are very useful for delivering the right insights with the right context.

On Tuesday, I launched www.wikiwebanalytics.com to community.

The wiki is meant to be an online resource for web analysts, where we can share best practices, details, basics/advanced tips in one place. There are lots of great people blogging about web analytics, and two wonderful new books, but we need an online resource where we can share, document our knowledge, and learn from others.

The goal for the wiki is 300 articles in 3 months.

We are on track to get there, and the existing articles keep getting better and better. Take a look at the article for Visit/Sessions.

There are lots of articles that still need to be written, and with the addition of each new article other ideas for articles are started. This page is a collection of all the latest ideas for articles that are needed or need editing.

The wiki is using Google Analytics - I am providing a public GA log-in to see the tool in action and provide data for articles. (I have planned to add open web analytics in the near future.)

According to Eric T. Peterson’s March 2007 Summary, 82% of web analysts believe that web analytics is misunderstood in their organization.

WikiWebAnalytics.com is not only a great place to learn about web analytics, but you could use the articles to help explain what web analytics is to your organization.

If the topic you want to reference doesn’t exist yet, start an article today!

Dylan

You have spent hours on data collection, building reports, and analyzing data, but if you don’t answer the question “So What?” for your audience then you have wasted everyone’s time – including your own.

I had a great conversation with Bob Page during lunch the last day of Emetrics. We were talking about the presentations and what we could take back and put into action.

Bob challenged me with the “So What?” test.

Most analysts are challenged with the “So What?” test either by providing actionable recommendations, or delivering web insights. The “So What?” test is a great litmus test for your analysis. “So What?” puts the perspective of your analysis in the right place – in the minds of the people who are hearing your recommendations.

Why should your audience read the email, PowerPoint, or attend the meeting? Why should they prioritize your recommendations over all the other tasks they have to do? What is in it for them?

Future Now created the We We Calculator to analyze the perception that your website is creating for your visitors – as analysts we don’t have such a tool (yet).

But here are three tips to help your insights and recommendations pass the “So What?” test:

1. Motivate: Take a few minutes and think about what motivates your audience. Do they have specific goals they need to achieve? If so, how does your data, analysis, and insight relate to those goals? Lead with that and they might follow.

2. Activate: What action do you want to generate from the analysis? Does your data support the action that you are recommending? What are the three data points that suggest that this is the right action? Help your audience understand the action and why they should prioritize it over other tasks.

3. Complete: Your recommendation is only a suggested starting point. If you help your audience complete the recommendation in their mind, then they can understand the impact relative to their other priorities. Help your audience visualize what success will look like. Help them “see” the benefits of your recommendations.

For me, the Emetrics passed the “So What?” test with flying colors – I have more ideas for web tests, and more contact with sources of interesting thinking that helps push ideas around in my head.

Jim Sterne and Matt Finley did it again.  The Palace hotel in San Francisco was the site of the West Coast US Emetrics event.  Over 500 people were in attendance, including practitioners, vendors, consultants and even a cultural anthropologist named Joseph Carrabis.

Joseph is Jim Sterne’s web analyst.  He took a look at Jim’s site, made changes to it and came prepared to talk about the results.

Joseph’s presentation was one of the most interesting.  In other presentations, I look for the one thing that I can take back and apply right away.  Joseph shared insights about how people’s brains are wired – nothing about web analytics just optimization for brains.  His main point was that 85% of communication is non-verbal and we perceive things subconsciously that affect our consciousness.

Two of my favorite points:

1. Left Face Front – You are more likely to like/trust someone when they present the left side of their face.  This works both in-person and in a picture on a website.

2. Left vs. Right Navigation – Navigation on the left hand side forces the visitor to pick the least painful path.  Navigation on the right hand side enables the visitor to pick the most enjoyable path.  (If you test this, please comment on your results.)

Joseph ran out of time around slide 10. When he asked the audience what he should focus on, everyone yelled that we wanted it all….go over the time and continue.  Joseph went an additional 30 Minutes over for an enjoyable hour and a half. 

I felt like I had just seen Bryan Eisenberg for the very first time. 

Everyone had similar reactions about the content.  It was mind opening, new to me, crazy in a good way, and well received. If you get a chance to see Joseph speak, head for the room and pick a seat up front. 

He also has another (#23) book coming out that looks interesting – Reading Virtual Minds.  There are excerpts available here. 

So, here we go again. My previous blog fell to the wayside once work started picking up…but now I am back!

How will this blog be different than the previous attempt?

In preparation for version 2.0 I have:

• Chosen a (hopefully) better tool box to write with. Avinash provide a great primer of the tools he uses, and a conversation with Eric Peterson that confirmed my suspicions. Movable Type was not the right platform for me.
• Set a goal to write at least one interesting post per week. How will I measure this?
  • I can measure posts per week easily. The hard part is writing them.
  • Interesting is difficult to measure. I will do this with metrics and outcomes like:
    • Number of trackbacks per post
    • Technorati Rank change from today
• Thought about the various things I would like to write about. 1 post per week is still 52 posts for a year. However, I have hope because there are a lot of analytics memes to be written about.
• Found inspiration with my first efforts and the people that have linked to me, asked me for more, and said you should write about this or that.

So, if you are still reading, thanks for the encouragement, poking fun of, help and upcoming lunch to get the blog going again.

Dylan

Tags: Blogging, Goals