A data funnel is like a treasure map of your website. In order for users to accomplish the goals you’ve set out for them, they need to follow a trail of web pages before they eventually reach that goal.

If your goal is to sign people up for an email, they’ll need to visit a page with a button that takes them to a sign up page (dotted line), then the sign up page (dotted line), and then a thank you page (big, red X) that verifies that they’ve signed up.

I briefly touched on data funnels last week when I talked about the importance of goals. Funnels and goals go hand in hand because a goal means you want your users you wind up doing something specific on the site or winding up somewhere.

So think of funnels as a map of where your users are in the process of getting to that elusive goal. For those users that don’t make it to the X on the map, a funnel can show you where they are getting stuck and if enough people are getting stuck that you should do something about it.

Here is what a typical funnel looks like:

In it, you can see that 853 users hit the home page, 126 reached the RegisterLogin page, and 21 made it to the Account Success page. That means 21 of your 853 users managed to follow your map and reach the goal. That’s a 2% conversion rate and depending on what exactly your goal is, is not too bad.

What the funnel lets you do is attack each step in the process individually. Let’s say you change your homepage around and add blinking lights and flashing arrows that point to the sign-up box.

In theory, more people should notice your sign-up box now. But instead of just checking how many people have created an account and seeing if that number has gone up, you can check the funnel conversion for that first step to see if it’s made it better or worse.

Funnels give you a more nuanced look at how each part of your site is performing. Instead of just having one big gauge telling you how the overall site is doing, funnels give you a more detailed look at the different parts of the site.

Once you’ve nailed down where there are problems and opportunities for improvement, then you can tackle them exactly where users are having the toughest time and make adjustments.

Image by cameronparkins

Most people that don’t put too much thought into their websites do the following: they log into Google Analytics (GA), check and see how traffic was yesterday, maybe click into the Traffic Sources section to see if there is any pattern there to explain the drop or rise in traffic, and then go back to doing their thing. Selling widgets, blogging about cats—whatever.

But if you don’t have a defined goal for your business, then you’re not only missing out on one of the coolest parts of GA, you’re also selling your site/business short. So ask yourself the questions: what’s the goal of me site? What am I trying to accomplish?

Everything on the page (and everything you do) should be geared towards that one goal. Whether it’s selling more widgets, increasing pageviews, making more money, etc. This will also help you make tough decisions about your business by simply asking yourself “which makes the most sense to continue to achieve our goal?” You’ll know which decision to make if you let your goal be your guide.

But you have to have a goal.

And once you do, you can have GA track it for you by setting up goals. What you’ll be able to see once you’re all set up is the conversion rate for your users, which you can break down into any of the slices of traffic you want (users from Finland that use Chrome, for example) and see which users convert best. It’s the kind of insight that causes “aha” moments and can improve conversion across the board.

Not only that, you can see conversion funnels too, which are a nice graphic depiction of where in the process of achieving the goal they drop off. If you notice most users dropping off at the sign-up page, then you know there’s something you can fix there.

If you’re a blogger, the goal-setting can be a little tricky. What is your goal? It could be to have users join a mailing list, read one specific post, or something as simple as subscribing to your RSS feed. If you’re a blogger and you can’t figure out a goal for your site, use RSS subscriptions. It encompasses a lot of things: are you writing interesting content that people like and keeps them coming back for more? Are you networking and getting the word out there so that more people come to your site? Are you selling yourself enough?

Tracking RSS subscribers via Feedburner is great, it gives you a nice, overall count and a trend of how you’re doing over time. But placing it as a goal in GA means you’ll be able to find out a lot more about your users. If users from a guest post at Sidehustleblogging sign up at a much higher rate than users from any other site, maybe you should figure out what it was about that guest post or what it is about Sidehustleblogging and their readers that “clicked” so well.

Then you try to replicate it.

That’s what analytics is all about: noticing patterns and repeating them to achieve your goals. Just make sure you have a goal to begin with, or else none of this will matter.

Image by D-32

The Google Blog has a great guest post up today from Tom Critchlow of Distilled titled Using The Wrong Tracking Code Can Cost You $500k a Year. In it, Tom goes over a very simple error that can cause a huge drop in conversion for a very specific slice of users—those using Internet Explorer 8.

It seems like such a non-issue until you see how the conversion rate for those users sucks compared to everyone else. And all because of a simple little thing like not using Google’s latest code.

I think a lot about how this stuff (analytics and numbers) can help people and small businesses with their sites, and this post is a great example of something that’s small, easy to fix, and has a huge impact on the bottom line.

Analytics can do a lot of stuff to make you smarter about who you target, how to target them more efficiently, and how to expand your reach on the web as quickly as possible. But when it comes down to it, what we all want to do is make more money and spend less time doing it.

Nice work Tom—well done.

I’ve said it before and I’ll say it again—the best way to use web analytics is to answer specific questions. Otherwise you’ll spend hours sifting through numbers and at the end of the day all you’ll be able to say is, “Well, that was interesting.”

But even the most innocuous question isn’t as easy as it sounds. Say the boss comes to you and asks you a question like this:

So we changed the image on the homepage, did the bounce rate go down?

Easy enough, right? Just open up Google Analytics, check the report for the days leading up to the change and then see what it was afterwards. Should be a simple matter of yes or no.

Wrong.

As much as your boss would like it to be that easy, it’s not. You have to look at all the moving parts on your site if you want to answer the question as honestly (and effectively) as possible. Some of the things you’d have to consider:

• Did we make other changes to the site the day we changed the image?
• What happened overall on the site that day?
• Was traffic any different that day? Did any high-bounce traffic get a spike (or a dip) that day?
• Did we send any emails that day that might pollute the data?
• Was our bounce rate trending in a certain direction to begin with?

You have to take all this stuff into consideration—you can’t just look at the overall bounce and answer yes or no. You’d be selling the analytics (and yourself) short.

This is why analytics is part science, part art. It’s not just about looking at the numbers and reporting on them, you have to use your brain too—that’s the fun part. It’s like being an investigative reporter—the best ones don’t just rattle off quotes from witnesses and police reports, they tell an engrossing story with all the information they’ve collected. That’s what keeps us coming back.

This is why it’s important that whoever is running the analytics be intimately informed about the site he/she is covering. If the person in charge of the analytics doesn’t know that an email blast is sent out every Wednesday the analysis will be wrong.

Of course the bounce went down! We always send emails on Wednesday and those people never bounce! Duh!

Does that mean you shouldn’t hire a consultant? No—it means the consultant needs to get to know the site as much as possible and constantly talk with someone who is knowledgeable. Your company should make it a top priority to give the consultant all the information he/she needs and assign someone to soak all this stuff up so that after a few consulting projects that person can run most of the analytics themselves. It should be part of the process.

Once you can run this stuff yourself, you’ll be in a much better position to spot anomalies. And if a huge problem comes up that you can’t solve—you can always bring the consultant back for help.

So next time you’re ready to shoot off a quick email to answer what looks like a simple question, step away from the keyboard and think about all the parts that need to be taken into account before you give your answer.

Image by Tallkev

This post was included in the Home Business Blog Carnival

Driving a car is like running a website: you have to keep track of a lot of things at once in order to get to your destination without crashing and burning.

You have to pay attention to what’s going on outside the car: other cars, jaywalkers, insane cab drivers, the weather, and the occasional deer. All the while, you also have to keep track of what’s happening inside the car: how fast you’re going, how much gas you have left, your mirrors, your windshield wipers, and you have to absolutely make sure no Yanni accidentally plays on the radio (trust me—road rage will ensue).

You have a lot of things to process and not a lot of time to do it in. And that’s why the dashboard is so important—dashboards process a lot of essential information and spit it out in really easy-to-read gauges, lights, and buttons. Everything you need to know about your car is displayed right there on the dashboard from the comfort of your comfy, heated leather seat.

A GA dashboard report

Same goes with your website—and the Google Analytics dashboard works the same way. In Google Analytics, the dashboard report serves as your summary of all the important numbers that tell you if your site is going in the right direction. When you log into your Analytics account, it’s the first page you see with actual data on it. The great thing about the Google Analytics dashboard is that your dashboard could be (and should be) completely different from my dashboard because we have completely different websites.

Everyone’s Dashboard is Different

Different sites serve different purposes and that means you need to pay attention to different numbers. If your site requires users to log in to access certain information, then expect the bounce rate to be a lot lower. If you have a terrible interface on your site you may have a higher time on site, which in this case is a bad thing.

The point is that every site is different and so should every dashboard report.

Under the Hood

While dashboard reports are a great way to keep track of what’s going on under the hood on a day-to-day basis, it’s not enough. Every so often, you’ll have to get your hands dirty and take a look under the hood.

That means going deeper into the reports in GA and looking for potential problems or nuggets of wisdom. Sometimes you’ll discover something great and unexpected (like a book review getting massive SEO traffic) and sometimes you’ll find things that are broken and need fixing.

I don’t want to overwhelm you with stuff, so I will write a follow-up post on how to actually use the dashboard, what and how to add and remove things from it, etc.

I know this can be intimidating and scary, but you have to trust me on this—there are some fascinating things happening on your site. And digging into your analytics is one of the best ways of discovering them. Don’t worry, I’ll be here to help you along the way!

Image by wOOkie