Hero Shots and Sliders

In the web 2.0 craze, almost everyone put a giant slider or hero shot at the top of their homepage.  Why not?  They're big, sexy, and provide a way to show off your company.  

Conversion experts say you should kill your slider.  

Designers say you need something to anchor the page.  

The problem is that most companies don't do or get anything useful with their slider.  They put up a slider for the sake of having a slider.

If you really want to get some mileage out of your slider, see what your marketing team thinks.  A large canvas that moves, grabs attention, and is best suited for visual media?  Sounds like a marketing wet dream.  (Look how well SlideShare is doing.)

Not all sliders are created equal.

Your slider should be clickable, encourage a single action, and respond to your visitors.  (Think of it more like a presentation than a web slider.)

There is nothing worse than a wordy slide that moves before I'm done reading it.  So many times, I have clicked back to the previous slide to finish reading and it slides again!  You got my attention.  I'm reading your copy.  Why would you want to interrupt that?

A slider should have a single purpose.  If your goal is to get someone to buy something from your boutique, your slider may feature various items from different categories, but it shouldn't also ask them to sign up for the newsletter.  

Finally, once you have captured someone's attention and convinced them to take action, you need a clear path to success.  A big "CLICK HERE!" button might be overkill, but making the whole slide clickable so they can't miss, isn't.

Sliders suck... but they don't have to.  What's on your slider?

Confidence Graphing

The job of an analyst isn't to find stories in the data.  It's a given that they exist.  The real job of an analyst is to be able to effectively communicate and convey those stories in a way that can be understood without elaborate explanations.

One of the biggest problems when presenting data on a graph, is being able to eliminate statistical outliers without modifying the data.  If you plot them the same way that you plot all the other points, they will invariably raise eyebrows and require explanation.

I was playing around with a graph this morning and I came up with a way to graph the confidence of my results.

The graph above has two lines.  The blue line is the conversion rate, and the red line is the number of visitors who saw that variation (the variation is page load time in seconds).

The blue line makes it look like 26 is a clear winner, and 18-20 convert twice as well as all other options.  Unfortunately, the red line tells a completely different story.  Effectively, the higher the red line, the more confident we are that the blue line is correct.

It takes a well trained analyst to be able to ignore the outliers in a graph like this.

My final result is a bar graph where each bar is given the opacity that corresponds with our confidence level. The darker the bar, the more confident we are that it is correct.  The outliers are still there (and going off the chart), but at 1% opacity, they don't warrant any attention.

Complexity is the hobgoblin of time

The entropy, or disorder, of the universe is constantly increasing.  Entropy is the measure of possible arrangements within a system.  In an isolated system, entropy never decreases.  This means the older something gets, the more disordered it gets and ultimately less work it can do.  (This is why perpetual motion machines don't work and a pendulum will eventually stop swinging.)

To fight entropy, you must constantly reduce disorder.

Entropy is primarily used in thermodynamics and statistical mechanics, but it applies to businesses and software too.

As time wears on, complexity and disorder increase.  You have to fight to keep things simple and organized. 

We've all seen companies that grow too big too fast and lose control over their internal entropy.  This results in a less productive and ultimately less profitable company.  If the system is not corrected, it will eventually stop.

Websites and software are the same way.  They start out clean and organized, but as time passes, they get cluttered with content, widgets, bells and whistles.  This clutter isn't all bad.  It's basically the equivalent of adding weight to the end of a pendulum.  The extra weight makes the pendulum swing faster, but it also makes it stop faster, and makes it harder to get moving again.  

When creating or revamping a website, the trick is to find the weight (level of complexity) that produces the speed (business results) that you can handle, and that doesn't require too much work to keep it moving.

Networking on social sites

Lots of companies have social profiles on LinkedIn, Facebook, Twitter, Pinterest, etc.  The problem with social profiles is that they are a lot of work to maintain and often produce very little value to your bottom line.  

Most companies think they can use their social profiles to generate interest and traffic to their site.  While this is possible for some companies that have a vast reach or "share-able" product, it doesn't work for the large majority.  The primary reason it doesn't work is because specialized products need a specialized audience.  Sharing specialized content to my 800 friends isn't a natural use of Facebook.

Many other companies try to provide customer support.  The idea is that you monitor the social websites for any mention of you and when something good or bad shows up, you respond.  Customer support, especially the extremely transparent social sort, is never a bad thing, but it takes a lot of time and effort.  Some companies have found wild success through creatively solving their customers' problems with YouTube videos etc. but they are the exception.

If you don't really have the time to keep up with LinkedIn, Facebook, Twitter, et al. but think your business should have a presence, I suggest you use them as they were designed, to network.  Every company can benefit from being connected to leaders and passionate individuals in their industry.  Connect with rock star employees from other companies, avid bloggers, and anyone you meet at a conference.  

Having a network of industry specific "friends" is invaluable.  Your profiles become low maintenance because  your network doesn't expect or want you to spam them with 30 tweets a day or 10 pictures of new products.  Rather, it's the modern day Rolodex.  When you have an opportunity or infographic you want to share, you have the perfect network to distribute it to the right people.

How do you use social?

The 3 stages of data

The web is a glorious place that lets us collect data on almost everything a visitor does.  All this data is fun to look at and analyze with various dashboards and graphs, but it doesn't stop there.  Once you have found meaning in the data, you still have to use that knowledge.

These are the 3 stages of data: Collect, Analyze, and Implement.

Some data in easy to interpret and use.  If, in a split test, your visitors purchase more when exposed to version A vs B, the implementation is easy.  Show version A to all your visitors.

Most data insights are not as straight forward.  Consider the insight that people don't feel comfortable giving out their date of birth on your signup page.  Does this mean they don't trust your site enough to share personal info?  Should you not ask for a birth date or should you explain why you want to know?  Maybe the issue is unrelated and this is just a symptom.

When you don't have a clear path forward, you have to collect more data either through advanced collection techniques or testing, analyze the results, and try again to implement a solution.  It becomes apparent that even valuable data insights can take months to fully implement and realize.

It seems that while there are 3 stages of data, there is only 1 overarching goal:  Minimize the delay from collection to implementation.

If you're just collecting data and viewing it in a dashboard, your data isn't able to achieve its goal.

What do you do with your data?

The downside of automation

It seems that everyone reveres automation as the holy grail.  You can automate your marketing, your email campaigns, and even your coffee maker.  Automation is great because it ensures consistency and frees up valuable man hours to do more mentally intense tasks.  But, automation isn't always good.

Automation reduces personalization and requires extensive maintenance to keep up with variable change.

Personalization is important because it not only adds a layer of attention, but re-enforces social ties to your customers.  I'm not talking about the kind of personalization that your bulk email service provider offers, Hi [First Name].  I'm talking about true personalization that would take terabytes of data to be able to program into an automated system.  The digital equivalent of my coffee pot being able to recognize that I'm dragging more than normal this morning and responding by making me stronger coffee.

I'm not saying that personalization of this level is impossible to program, just that it isn't practical for most applications.  You could easily blend a little automation with a likable account rep and still benefit from automated processes while paying personalized attention to your customers.

Maintenance is another concern when looking to automate anything.  As humans, when we learn something new, we automatically integrate it into our process.  For a computer program, a change in the outside world may require a minor adjustment or a complete rewrite of the original automation.  

My personal belief is that if you can't afford to replace something, it's outside of your reach.  If you can't afford to replace your automation (based on the time and monetary investment), you're not getting enough out of having the process automated.  

Start manually and document each step and decision point of your process.  Just having the process scripted on paper should increase the consistency of your results and will help you determine if automating it is worth the effort.

I know a guy who took a 4 month process and made it a 1 day process through automation.  He didn't completely remove the human element from the decision points, just automated the basic steps (reason why it still takes a whole day).  The program would take weeks to re-write or make a major update, but that's still a huge savings over doing it manually, even once.

Is your automation working for you or just creating extra work?

Second Chance: Appeal the Bounce

A bounce is when a visitor lands on your website and promptly leaves without doing anything.  You work hard to get visitors to your website.  Search Engine Optimization (SEO), Social Media Marketing (SMM), Content Marketing, Email Campaigns, and Pay Per Click (PPC) ads all cost time and money to maintain.  If it costs an average of $1 to get each visitor to your site, each bounce is a wasted buck.

They came, they saw, they left... you lost a dollar.  What can you do?

Why not make one last appeal to your bouncing visitor to get them to hang around?  These "Are you sure you want to leave?" messages can be super annoying, but if you only show it to visitors that you've already lost, what will it hurt?  Even if you get just one of those bouncers to stay, you're ahead of the game.

I suggest limiting the message to those who are not engaged.  

Depending on the site, engagement could be as simple as scrolling or clicking (blogs tend to get lots of bounces because they are just 1 page.)  You could also limit it new visitors.  

Here is some sample javascript code.

<script type="text/javascript">
    //attach the exit confirmation to the onBeforeUnload that fires just before the page closes.
    window.onbeforeunload = confirmExit;

    //set a global variable to be able to turn on/off the exit confirmation
    var confirmLeave = true;

    //if the user clicks anywhere on the page, don't show the exit confirmation
    document.onclick = function() {
        confirmLeave = false;
    }

    //if the user scrolls at all, don't show the exit confirmation
    window.onscroll = function() {
        confirmLeave = false;
    }

    //function that displays your exit confirmation
    function confirmExit()
    {
        //only if our global variable is true, Appeal to the Bouncer to get them to hang around.

        if(confirmLeave===true)        {
            return "96% of our clients would recommend ContourThis to a friend! Are you sure you don't want to check us out?";
        }
    }
</script>

You have to use the browser's built in popup box to display your message.  Not all browsers will let you display a custom message (FireFox), and some browsers won't keep the page from closing (Opera), but the point is, in the case of a would be bouncer, anything is better than nothing.