Observational biases often infect analysis, and can convince businesses to make incorrect assessments of their website data. In 2010, David H. Freedman coined the term “Streetlight Effect” for an observational bias represented in an old humorous science anecdote. This Streetlight Effect can very easily affect the insight gained from Google Analytics, should you not be aware of it, and not take measures to properly shed light on your data.
What is the Streetlight Effect?
The “Streetlight Effect” is named after an old late night parable told by scientists gathered ’round the flickering bunsen burner’. It goes as such: Late one night, a rookie police officer came across a stumbling visibly drunk man in a parking lot, intently examining the ground. “What are you doing, drunk man,” asked the police officer. “I dropped my wallet,” the drunk man slurred back at the police officer, reeking of gin and stale cigarette smoke. (blog author note: I’m trying to give the story a little more character than it normally gets) The police officer looks around and seeing no wallet asks, “Are you sure you lost your wallet here?” The drunk looks at the officer and belches out “No, actually I dropped it over in that dark alley.” The rookie officer is stunned. “If you dropped your wallet in the alley, why are you looking for it here in the parking lot?” he asked incredulously. “The light here’s better,” hiccuped the drunk. (more…)
Recently, over the holidays with family, I participated in a conversation where I mentioned that the data supported my position. I made the mistake of saying “Statistically….” which as you might imagine, in a holiday family setting, is bound to generate at least one comment about how you can prove anything with statistics, and a dismissal of a scientific study, without reading it, because it contains *GASP* data.
“Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.'” – Mark Twain
Interestingly, there is no record of Benjamin Disraeli saying that, so the attribution of that quote by Twain to Disraeli itself was a lie. (more…)
Even Vulcans can be afflicted with dreaded cognitive biases.
The goal of content testing different variations of web pages during a conversion optimization experiment is to get as accurate a result as possible, in order to say which variation is the best possible one.
Just because we’re testing, however, doesn’t mean we can’t be wary of cognitive biases which can filter into our experiments. Just because you put up two pages, release them to the public, and then gauge reaction, doesn’t mean that you’re not subjecting those tests to your own or others’ perceptual distortions. Watch out for these cognitive biases when forming content experiments. (more…)
Part 2 in my unofficial “You’re Probably Doing It Wrong” series. Read the first part about using Google Analytics to gain Insights.
The Cruel Tutelage of Sayf Sharif
You are wrong. This is the first rule. You think you know what the right thing to do on your website is? You are wrong. You think you know what to do to properly improve your digital presence? You are wrong. You think you know what your users want? You are wrong. Before you can truly begin to optimize your website and your online digital presence and marketing you must come to this realization. You must be unmade before you can build yourself back up. You must know that everything you think you know to be good, and right, is wrong.
I emphasize this with a beard swish.
Gaining Insights With Google Analytics
There are four steps to really taking advantage of your data. Measurement, Reporting, Analysis, and Implementation. It’s too easy to fall into the trap of implementing your data tracking, and then measuring and reporting on that data, but never really analyzing or acting on it. The entire point of your implementation of tracking, and measuring it, should be to use that data to improve your website’s performance. (more…)
What does this sweater have to do with universal analytics? Read on…
Last week I had the pleasure of attending the Google Analytics Summit for 2012 in Mountain View, California. It really was a great two days, and the Google Analytics folks did a superb job of hosting a bunch of nerds interested in their product. There were tons of announcements, particularly on the first day, and the main one involved the biggest change to Google Analytics since it transitioned from the old Urchin.
You can read all about it on various blogs covering the event, particularly Google Analytics own blog. So what are the key points?
Easy and Fast Measurement
Google Analytics had up to this point stored a whole bunch of it’s information within the user cookies, which constantly would get passed to GA on every pageview. Use premium with 10-20 custom variables on a visitor? That was getting passed with every page hit.
One of the biggest innovations with the new setup, is that there will be only 1 simple cookie with very basic information in it. This will speed up page times of the entire internet with the sheer drop in how much data gets passed on every page hit. All the data now kept in cookies will now reside on the Google Analytics servers, allowing a whole bunch of other things that we couldn’t do before, including customizing our sessions server side. Right now sessions are 30 minutes long and are in cookies. Get up and go chat with someone at the water cooler for 30 minutes and come back to your computer? New session. Now you can set your own session lengths and define them yourself up to 4 hours long. You can also put time limits on your campaigns of up to 2 years. In addition they’ll be opening up more code libraries and tracking api’s for better interactions.
User ID Control
The whole new Universal Analytics is based on users, not visits. It’s a key distinction. We’re interested in user behavior overall, not behavior of disconnected visits or sessions. We do this by having an override ID on every device interacting with GA. Sure there are issues like we still need to identify people via a login or some other means, but once we can do so, we can now see specific users across devices very easily.
Updated as of January 23rd, 2014 with version 6
Recently I was tasked with tracking YouTube videos on a website using Events in Google Analytics, and I was not particularly thrilled with the options. Either old blog posts from even our site, or posts that didn’t lay out any specifics on how to do it, or posts that listed basically how the YouTube player API instructs people to track videos. Most from a year or even two ago. Which was all well and good, until I realized that I was dealing with possibly hundreds of pages, with a variable amount of videos on them.
So I wrote it. (more…)
The sheer number of blog posts out there on how to increase your website conversion rate are overwhelming.
10 Ways to Improve your Website Conversion Rate
9 Time Tested Ways to Increase Website Conversion
25 Ways to Increase Conversion Rates
7 Basic Ways to Improve Your Conversion
And you know what, they all make great points. There are some really talented conversion people out there working to make websites better, but if you’re reading this, you’re either a LunaMetrics regular, or you’re reading this post looking for SOMETHING SIMPLE you can do to start out. (more…)
Last month, Phil wrote a blog about five features of Google Analytics that you probably are not using, but there obviously are more than five, and here’s another one that you probably aren’t using, but you should. In fact, I think this is probably one of the KEY things you should do when setting up an account and a new website, and it’s my bet that the vast majority of people don’t do it at all.
Why use Custom Alerts? Because you don’t check your analytics every day. OK well some of you data geeks that read this blog do, and I do, but most people who are just regular people, they don’t. That guy who is wearing 20 hats, he doesn’t have time to go over his data every morning for an hour or two. You could argue he should, but maybe he doesn’t. Custom Alerts can let him know, in general, if there is something that needs his or her attention. Is something significant, or at least possibly significant happening? Should someone take a look? it’s like a check oil or check engine light on your car. If the light doesn’t come on you probably (I hope) change the oil on your car regularly, but if something is going wrong with your oil, then you sure are glad that light is coming on to let you know to check it. (more…)
This is not a “How to Install Custom Variables” post.
I’m not going to bore you with a long rambling introduction to Google Analytics Custom Variables. We’ve talked about Google Analytics Custom Variables a few times on this blog, as we should… They’re an amazingly powerful way to get more out of your site data. Jonathan Weber’s early series from 2010 is still mostly valid though it uses the older traditional tracking code. If you need a primer, they’re a good place to start:
Google Analytics Custom Variables, Part 1: Why?
Custom Variables, Part 2: The Code
Custom Variables, Part 3: Slots
Also don’t forget what Michael Harrison laid out at the end of last year. Custom variables need to be in your code BEFORE a pageview or event is tracked, or they won’t work at all. Also the code he uses on this page is the current asynchronous tracking code format for custom variables:
Google Analytics Custom Variables Not Working?
And hey since I’m linking things, the best resource is usually from the source. Here’s the Google guide for Custom Variables.
Hopefully you can discern the basics from those articles, and others across the web about how exactly to use custom variables, how to segment your reports, etc. This post though is about specific ways to use them. it’s funny how often I’ll talk to a client who seems to understand that they’re very powerful, but can’t figure out ways to use them on their site.
So here are twenty different ways you can use custom variables. Five for each of the four main website types. Content sites, Ecommerce Sites, Lead Generation Sites, and Self Service sites.