This post was co-written by Spring Metrics and UserTesting.com
If you’re an e-commerce site, increasing your conversion rate is key to running your business. But making blind changes to your site or marketing campaigns can be counterproductive. You first need to understand your users:
- Why are they coming to your site?
- What are they doing on your site?
- Why are the buyers buying?
- Why are the non-buyers not buying?
In this post, I’ll discuss four methods you can use to better understand your users: Session Analysis, Aggregate Analytics, User Studies, and Surveys. Each method has different strengths and weaknesses, so you will likely want to try a few of these in concert to best understand your site and your users.
1) Session Analysis
What is session analysis?
Session analysis is when you analyze site data (i.e. “time on a page” and “navigation flow”) for a select number of users.
Session Analysis, at a Glance:
Pros: Quick and easy. Let’s you see actual user data.
Cons: Can’t ask users “why.” You’re only looking at a small sample size.
Example: You look at a dozen buyers and a dozen non-buyers. Buyers usually land on your home page; use the “most popular products” list to find a product, and then purchase. Non-buyers usually land on your home page; use the on-site search – sometimes multiple times, and then leave. One possible conclusion: search results do not seem to be matching the users’ expectations – there is work to be done here.
What’s a good tool to use for session analysis?
Spring Metrics, other analytics tools, or your server logs.
Why should I use session analysis?
This is a quick and easy first step that will get you started within just a few minutes. While you should be able to target individual user data, you won’t be able to ask “why” to your user. You’re also limited on how many users you can realistically analyze.
2) Aggregate Analytics
What is aggregate analytics?
Aggregate analytics is simply reviewing all of your site data and metrics to help identify user trends.
Aggregate Analytics, at a Glance
Pros: Can see trends to understand different segments of users; data view is 100% comprehensive rather than just a sampling.
Cons: Can’t ask users “why.” Your web visitors become statistics rather than actual people.
Example: You look at your data and notice that first time visitors rarely buy, but second and third time visitors seem to convert at a higher rate. You might take two actions at this point; you may create offers for only first time visitors to try to get them to buy sooner, or you may try to capture other information such as email address on the first visit, so you can get them to come back for a second visit when they are more likely to buy.
What are good tools to use for aggregate analytics?
Google Analytics and Spring Metrics.
Why should I use aggregate analytics?
Analyzing your global, overall web data can help you determine high-level user trends. For example, you should be able to pinpoint where most of your visitors come from, and which of your marketing efforts are most effective.
3) User Studies
What are user studies?
User studies are when you observe an actual person using your site. In most cases, the user is speaking out loud as they browse your site so you gain insight into their thought process.
User Studies, at a Glance
Pros: Can understand users’ intent and hear users’ why instead of just tracking their actions.
Cons: Cannot realistically test every user.
Example: You run a user study with five participants. Three of them get frustrated because they don’t see the shipping cost early enough in the checkout process. You can make that change immediately or possibly run an A/B test with potential changes. If you had studied just the website data, you would have seen them stopping the process, but you wouldn’t have known why they were leaving the process early.
What’s a good tool to use for user studies?
UserTesting.com is an affordable web-based solution where you can get results back in an hour.
Why should I use user studies?
Hearing users’ uncertainty and confusion when they hit one of your pages is invaluable. In addition, having them tell you what they don’t like, what they would prefer, and what they expected (compared to what they saw) is an eye-opening experience.
Pro tip: if you are showing results to your boss, show them the actual video footage (or if it’s an in-person user test, have them come and observe at least one session). Watching users struggle with the site is always more persuasive than a written report or PowerPoint deck.
4) Surveys
What are surveys?
Surveys are generally a collection of responses tied to specific questions or tasks that are asked to a large audience.
Surveys, at a glance
Pros: Can sample a large pool of users; receive non-site specific data such as general buying habits
Cons: What users say they want may not reflect actual behavior; sometimes difficult to quantify where users are getting confused or stuck on the site.
Example: You survey 1,000 people to understand what sites they shop on and what influences their buying behavior. You find that users dislike getting charged large amounts for shipping and handling, so you create offers for Free Shipping for some or all of your customers.
What’s a good tool to use for surveys?
SurveyMonkey and Zoomerang are both very popular solutions.
Why should I use surveys?
In some situations, a survey can be helpful because it can address topics that may not be directly site-related. Sometimes—the problem might be more product or policy related.
Conclusion
Each method of analysis helps you to better understand nuances related to user trends and behavior—and each brings additional feedback that helps you figure out how to serve your customers best. Generally, you will want to choose several methods to get a well-rounded perspective before you dive into changing your site and campaigns. If you do the analysis work ahead-of-time, you can save time in the long run and gain more satisfied customers.


