Side A: Path Analysis to Better Understand Your Website and its Users
This is the first in a series we call “Side-by-Side”. In this series, we will craft a pair of postings around topics that can help you build a better website. Side A will appear in the “Concepts” category and will discuss the topic as a concept. Side B will appear in the “Hands On” category and will go hands on to teach you how to actually take the concept and apply it to your website. Typically, Side B will be posted within a few days of Side A.
In today’s world, it is no longer enough to simply craft a professional, sleek website. You might have the coolest features, and a carousel that shows visitors the most popular products on the website. Maybe you even have different themes to use at different times of the year to give your site depth. You even put the date on the homepage so it feels current to your users. It’s a masterpiece. Your client absolutely adores it – it’s everything he wanted, it represents his brand, and he knows it’ll be great.
But he doesn’t know how. He wants to know people are using it and taking advantage of your magical masterpiece of web-nificence. He wants you to use this thing called “web analytics”. He doesn’t know what that is or does either, but he knows that his drinking buddy raised his site conversion by using it, so he wants it, too. After a while, using analytics, you are able to tell him how the site is being trafficked with confidence, but he wants more, he doesn’t want to know which pages are most popular, or where his visitors are geographically located. When they click on one thing, he wants to know what they click next. And after that? And after that.
Path Analysis does just this – it analyzes the path that your visitor’s take on your website. It sounds great, but we need to be careful. We’ll go in-depth here as to what Path Analysis is, why some people will tell you it’s useless, and why, when applied correctly, it really isn’t.
Path Analysis and its Assumption
Path analysis, very simply, would be the same as someone following you around the mall, taking note of the order of the stores you entered, the order of the things you looked at, if you bought anything, and what you did next. It’s easy to think that this could be useful information. Websites are always adding new ways to get to their most important pages (as they’ve decided) and you would think that path analysis would show them if users were moving around the site the way they expected.
While this sounds glamorous, you have to keep in mind the assumption that Path Analysis must make. That assumption is that every person is deliberate in their movements. Take, again, our mall example. Is there really any rhyme or reason to the order of the stores you enter? Some days, that might be true as you may be short on time, or have something very important you’d like to pick up first, but, most times you’re meandering. You know that you have to hit the shoe store before you leave, but if something shiny catches your eye, you’ll enter any store.
It’s because of this that many people will tell you that path analysis is useless. People aren’t so precisely deliberate, and, on the web, you have to deal with things like the Back Button, which web analytics hasn’t quite mastered yet. They say that if you have data that shows that visitors move from one department to the other, that this path doesn’t mean anything. For the most part, they’re right. When you start tracking the path people take through your website, what you’ll learn is that there are so many possibilities that 10 people looking for the same item might find 100 ways to come upon it.
Path Analysis Does, Indeed, Have Value
While path analysis assumes that users take deliberate actions, those that will tell you to ignore path analysis are making a worse assumption – that there is nothing of value here. I know what happens when you assume, and, I don’t know about you, but I’m no ass. There’s still something here.
If path analysis requires deliberate user action to be relevant, all we need to do to find value is identify places where action is expected to be deliberate, and look at those. In doing so, we can still get some return on our investment (in doing the analysis), and find out some interesting things.
The first place we expect our users to be deliberate is during checkout. Your client wants to sell more products, and he knows a good number of people are coming to the cart, but are they completing the path from cart to confirmation? Maybe not.
In the past, I worked for a client where we noticed that there was an unusual number of people breaking the path in the cart. When we looked at where they deviated, we found a problem. This client had a “Free Shipping” department in their store. A great idea… but terribly misleading during checkout! Potential customers saw shipping charges, saw the bright blue button marked “Free Shipping” and started to click away. Short story even shorter… we had to remove the button from the checkout, so that it wouldn’t distract users from completing this deliberate path.
Where else could you use path analysis to learn about your website and its users? Think about the mall example. When would you do something that you would consider “deliberate”? For instance, maybe you are looking to purchase new shoes. The first store doesn’t have the shoes you want. The second store doesn’t have your size. The third store is just right – they have the kicks you want in the size that you need. Isn’t your path from store to store to store, in this case, a deliberate action?
By visiting the second store, you are acknowledging that the first store did not have what you wanted. On your website, you can use path analysis in conjunction with search. Search is the “second store”. If a user browses your store and doesn’t find what they were looking for, they’ll search for it. They might believe its simply in another department. Understanding when your users turn to search and what they search for can help you identify product miscategorizations, or even identify products you do not currently stock that would sell if you had them.
Can you think of any other deliberate actions where path analysis may have some insightful insights to share? Leave a comment below and tell us!
Next Steps
As we’ve seen, path analysis can give us some very insightful observations about how our users interact with our website… but keep in mind it’s very easy to over-engineer it and lose time. Only focus on those actions that your users are completing deliberately to see where something unexpected happens, and only take stock of unexpected paths that happen in large numbers. If one person can’t find the product they need, please don’t recategorize your entire store!
In Side B, we’ll take a look at Google Analytics (with which you are hopefully already familiar) and we’ll show you how to set up goals and funnels to do some path analysis of your own! Stay tuned.
What They Say