FREEDyourMind

Guess Who's Coming to Ann Arbor?

Last night we kicked off our 2009 ForeSee Results User Summit, eVolve. 

Why the name eVolve?  Well, to be successful we need to evolve our use of web analytics.  Many things around us are evolving, and at a faster rate then we have seen in the past.  The "internet economy" is evolving, from a rising tide that was lifting all "internet" boats to a fiercely competitive environment that has to fight and win to survive.

Consumers have evolved.  Thanks to the internet, there are more choices.  We are no longer restricted by location. 

The switching costs are lower.  We can  be in multiple places at the same time, by simply opening up another browser tab/window.  Our costs to switch are near zero thanks to the internet.

And now consumers have become more knowledgeable.  And with knowledge comes power.  The consumer is now in charge. 

And our web analytics solutions need to evolve.  We need to integrate our data, and become smarter.

To win we must compete and win.  Those that satisfy their customers/prospects will win.

It's Almost Time . . .

I am here at eMetrics in San Jose for what will for sure be another great emetrics conference put on by Jim Sterne and his team.  And while I always look forward to eMetrics, this time will be even better.

Tomorrow morning at eMetrics we will be announcing a new product from ForeSee Results that builds on our exisitng product and the American Customer Satisfaction Index technology. 

A lot of great effort by our team at ForeSee Results has gone into the product, and it will take the web metrics ecosystem to the next level!  Details tomorrow.

What is happening with my conversion rate?

This is the question I received recently at an emetrics conference.  The web analyst was getting questioned by his management why their conversion rate was dropping.  After all, his boss kept reading about the importance of conversion rate.  I hear this question a fair amount.  Recently I heard this from a senior exec at a multi-channel retailer.  Their online sales were soaring, so was their traffic.  And customer satisfaction was also increasing. But conversion rate was dropping.  They wondered why?  Well, let's take a closer look. 

As we looked at the data we found a couple of interesting things going on.  Their traffic was soaring for two reasons.  A good percentage of the growth was the result of a very effective SEM campaign reaching out to new customers.  New visitors were converting at a lower rate then the normal site traffic.  Not a surprise, we expect to see new visitors convert at a lower rate then returning visitors.

Another strong contributing factor of the growth was a significant increase in visitors doing research for purchases in the store.  This segment of the audience will also convert at a lower rate then traffic that is  looking to purchase online.

Now, the reality is that everything this retailer was experiencing was positive, yet it was driving conversion rate down.  Conversion rate can be an important metric, but only when used properly.  I find that most sites are using it incorrectly.  Conversion rate at an aggregate level is a great example of a misleading metric.

Be careful that the metrics you use measure success, not simply measure what you can count!

Perception is Better Then Reality

I had an interesting conversation at an industry event a short while ago that is worth sharing.  To protect the innocent, let's use the name Sam.  Sam was questioning the value of voice of customer analytics.  His premise was "why do I care what they say, I care what they do.  Satisfaction, attitude and perception are soft and fluffy.  Give me hard data that I can rely on, clicks, hits, page views, conversion rates and sales."

My answer -- perception is reality, better yet, perception is better then reality.

Well, what do you think?  Which is correct?

Let's dive a little deeper.  I can understand where Sam is coming from.  Real numbers are good.  I know when I am looking at our financials I want real numbers, not opinions and perceptions.  I know when I am looking at the shrinking stock market, I want real numbers, not excuses and opinions.  So why do I want perceptions, intent, attitudes and opinions when talking about something as important as the success of my website. 

There is more to the visitor then today.  There is tomorrow - what the visitor will do in the future - will they return, recommend my site to others, will they be loyal?  There has been a lot of work done in years past on the lifetime value of a customer.  That is an important concept. 

What the visitor will do in the future is directly dependent on their perception of the experience today.

The visitor may have bought today.  But will they buy tomorrow?  If they were satisfied they will. If they were not satisfied they won't buy from you in the future - they will go elsewhere.  The visitor's perception of the experience (their satisfaction) will determine what they do in the future.  Their perception of the experience is far more important then the actual and factual experience.

KPI's Are Old School: Let's Focus on KSI's

A lot of smart, good, well-respected people talk about key performance indicators (KPI's), including Eric Peterson's excellent Big Book of KPI's. So I don't want to knock KPI's. And for a long time, KPI's have often been an excellent measure, but of performance, not necessarily success.

Maybe this is an issue of semantics, but I don't think so. I want to make a case that KPI's should be supplemented with KSI's (key SUCCESS indicators).

When you're measuring performance, you're looking at what has already happened. Things like how long it takes a page to load, how many visitors, bounce rates, search results . . . all those things reflect what has already happened.

The problem is, we measure what we know how to measure: performance. Measuring success is a much harder thing to get your mind wrapped around. Sure, you can measure sales as an indicator of success, but even that reflects what has already happened. Conversion rate is another good example that tends more towards being an indicator of success than just performance, but again is backward looking. And for that matter, conversion rate can often be a very misleading metric.  If you've ever heard me speak, you've probably heard me ask: how can you steer a car in the right direction if you're always looking in the rear-view mirror?

The bottom line is that behavioral data is backward looking. Important data, yes. Valuable data, yes. Insightful data, yes. But it doesn't help you as much as it could. The other issue is that behavioral data doesn't really measure the success of the experience and therefore the long-term value to the organization.

Satisfaction data, when measured correctly, is forward looking. Academic studies show that certain satisfaction methodologies can actually predict the future. They can tell you what your customers are likely to do tomorrow.

Some examples of KSI's: satisfaction, purchase intent, likelihood to recommend, likelihood to use the website as a primary channel, etc. In other words, KSI's = future behaviors.

The Web Analytics Association

I could remind you to vote today, but I figure you're getting hounded from enough fronts as it is.

Instead, I'll just say that we're thrilled to be joining the Web Analytics Association. And if you aren't a member, you should join too!

More Good News For Website Analytics

More evidence from Jim Sterne's eMetrics Marketing Optimization Summit that the down economy isn't negatively impacting members' online marketing budgets or interest and attention to web analytics.

65% of respondents said the state of the economy has no negative impact on their overall online marketing budgets. And while just over a third said it was negatively impacting budgets, only 15% plan to actually cut budgets.

More than 80% of respondents have seen increased interest in analytics from senior management within their companies. The report notes that this is usually a pain point for eMetrics attendees, and we certainly see it with out clients: getting management attention and buy-in is always a challenge.

But even more important that increased management attention: twice as many are increasing rather than decreasing their analytics budgets.

This is a powerful statement to the power of web analytics. As the report notes:

Web analytics helps us maximize the effectiveness of our shrinking marketing dollars by pointing out our strengths and weaknesses and providing an actionable roadmap to our most impactful ROI channels online. Tough times really do call for tough measures.

In a strong economy, anyone can succeed. When times are tight and competition is up, only the strong will survive.

What To Do Now (part 2).

About a month ago I shared some thoughts about what a VP of E-Commerce should be doing in these tough economic times.  In case you missed that post, the 5 items I listed were:

1. Be aggressive...and smart.
2. Focus on what matters most...your customers.
3. Measure what matters most...because you cannot manage what you cannot measure.
4. Accuracy, precision, reliability and validity of your measurements is critical to success.
5. Resist the urge to be penny wise and pound foolish.

Since then the markets have dropped over 20%, we have seen a huge number of companies announcing layoffs and losses.  So. lets continue the list.

6. Rely on metrics, not only on your expert opinions, your instincts and loud complainers.  In any times, but especially in tough times with limited budgets and resources, we need to make sure we are focused on improving the things that will have the most impact on our business.  Not on the things that have the loudest complainer or are an expert's favorite.

7. Make sure your metrics are valid.  They need to measure what is important.  Often measurements like conversion rate can be very misleading.  We need to understand what we are measuring and why it is important.

8. Make sure your metrics are sensitive.  They must be able to detect change.  Simple scales (such as 5 point scales) are often not sensitive enough to detect change.

9. Make sure your metrics are precise.  They must be exact.  When we are fighting for every visitor, every customer and every dollar, we need precision in our measurements.

10. Make sure your measurements are accurate.  Our measurements need to be correct.  Garbage in leads to garbage out.  Measurements that are not accurate can lead us down the wrong path with a false sense of security.

Oh, and I almost forgot.  Don't look at your 401k statement - it will make you sad.

Are Domain Names Not Intuitive Enough Already?

This is fascinating to me. Google has a tool in beta that allows you to see the top searches in the world (or by region) as well as which searches are gaining in popularity.

A few days ago I looked at the top 10 rising searches in the last 7 days? Eight of them are domain names:
1.       cnn          +50%
2.       you          +20%
3.       hi5           +20%
4.       google      +20%
5.       games      +20%
6.       firefox       +20%
7.       youtube    +10%
8.       you tube   +10%
9.       wikipedia  +10%
10.     orkut        +10%

Why would you go to www.google.com and then type in "google?" Why google "firefox" or "wikipedia" instead of just typing in "wikipedia" into your address bar?

I don't know what "you" is supposed to be. When I googled it, the sponsored result is YouTube. There's also a South African magazine called You. What caused the search term "you" to go up 20% in the last 7 days?

You can also search by region, so the top 10 news and current events search terms in the United States from the last 7 days are:

1.     news        
2.     msnbc    
3.     new york times    
4.     nbc    
5.     wall street journal    
6.     cnn    
7.     washington post        
8.     usa today        
9.     ny times        
10.     fox news

Again, all domain names that shouldn't need to be googled, except for the first one.

The top 10 searches in Michigan are:

1.     news    
2.     free press        
3.     detroit news        
4.     msnbc        
5.     free press detroit        
6.     freep        
7.     new york times        
8.     nbc        
9.     cnn        
10.     wall street journal

Whereas the too 10 searches from California are:

1.     news    
2.     msnbc    
3.     clarin    
4.     new york times    
5.     nbc    
6.     wall street journal    
7.     ole    
8.     noticias    
9.     la times    
10.     cnn    

What's interesting about the California list is that one (possibly two, I guess) of the top 10 search terms is in Spanish, which gives you a sense of just how many Spanish speakers there are in California.

I don't know what you actually DO with this tool other than play around with it. But it is interesting. I would have expected your average internet user to be savvy enough to type in msnbc.com instead of needing to do a google search for "msnbc." But I guess not.

Into the Wild: Web Analytics as a New Frontier

Into the Wild is a bestselling book by Jon Krakauer and a movie made last year by Sean Penn. It's about a young man named Chris McCandless who disappears into the wilderness of Alaska for a planned stay of several months. He had no compass and no map and his only food was a 10-lb bag of rice. McCandless died after a few months alone in a bus in the Alaskan bush; accounts differ on whether he died of starvation or because he was poisoned by some seeds he misidentified.

There are debates about whether McCandless was crazy or stupid or idealistic or heroic, but what seems clear is that he could have easily survived with a few more preparations and a better understanding of what he was getting into. Admirers and detractors alike agree that he didn't have the proper respect for the challenges he was about to face, and that he mistakenly trusted his own instincts above logic and facts.

Bear with me. I promise this is related to web analytics!

Just like hikers and mountain climbers talk about preparedness and proper tools and equipment, those of us working on making successful websites have our own tools and equipment. Using a website to foster online and offline sales, loyalty, and recommendations does not have to be a random attempt filled with educated guesses. There are proven, scientific tools to guide your choices and you can easily determine what kinds of metrics have the greatest ROI and which tools to use for which problems. As they say, when you have a hammer, every problem looks like a nail.

Somehow this all reminded me again of Eric Peterson's white paper about the ecosystem of web analytics and how to integrate various web metrics for maximum success. I guess the point is that you don't have to be going "into the wild" when you're running a website--you don't have to go out there without proper tools and protection and eqiupment. The right tools will equal success. McCandless would probably succeeded if he'd only had a map.

So while others debate about McCandless' mental health, you can question mine because everything makes me think of web analytics.