Jolina Pettice

6 Ways to Dive In to Analytics & User Experience

Jolina Pettice     MIMA Summit, Online Marketing, Web Analytics

This post is one of a series of liveblogs from the 2010 MIMA Summit.

Web Analytics is an excellent way to ensure your marketing efforts are not in vain. However, looking at Analytics without understanding user behavior may leave you treading water instead of creating a wake your competitors will fear.Web Analytics

One way to take your methods of measurement to the next level is to combine Analytics with User Experience (UX) information. What you stand to gain can be great whether you dip a toe or cannon ball into blended analytic waters.

To make sure you don’t sink, first understand that Analytics will tell you what’s happening on the site. User Experience Research then helps you understand why it’s happening by gaining insight into the behaviors and motivation of visitors.

When you combine both, you can start to understand the What and the Why of what’s happening on your website. This also allows any potential problems to float to the top more quickly, allowing you to see the problem you need to solve and what changes might do just that.

The following case study illustrates what you can find when you combine the power of Analytics and user behavior:

An insurance quoting website, which had a 14 page quote and purchase process was going through a website redesign and decided it was the perfect time to double check their form process.

The goal was to identify where in the funnel visitors were getting stuck and abandoning the process.

After understanding user experience and their reactions to what you are testing, the data should be compared to Analytics. Does Analytics show visitor ‘drop-off’ points in similar places as what was identified as confusing/irritating by the users?

To understand drop-off points and not spend time in lower-priority places, be sure to identify logical places for drop-off. In this example a make-sense drop off would be after the quote is delivered because you can expect that some visitors won’t like the quote they received and abandon to either get a different quote or purchase elsewhere.

Where you want to spend time is identifying the unexpected drop-offs. One such drop-off point for the health insurance quote form was the request for a social security number. Almost all users are cautious when giving such information online, but the problem in this particular case was that the information was asked for too early in the process before the person is invested.

As such, they identified an item to test. After placing the request for the social security number further into the funnel where the user was more invested and felt like it was OK to give their personal information, the abandonment rate decreased.

The next drop-off point was the question of where they attended college, which some – via the user experience testing – found offensive and/or didn’t understand how it was relevant to an insurance quote. The Analytics confirmed this and showed a higher than normal drop-off rate at the point that question was asked in the funnel.  Ask yourself, why is this question part of the form? Is it necessary? If not, remove it and improve the percent of visitors who get further in the funnel and closer to a sale.

Start getting more out of your website by diving deeper into the following 6 areas using both Analytics and User Experience Data:

1. Landing Page Optimization
Analytics: Bounce Rate, Conversion Rate
UX: Why people convert.

2. Site Navigation
Analytics: Top Content
UX: How they get there

3. Form Completion
Analytics: Abandonment, Page reloads
UX: Specific Objections

4. Content
5. Analytics: Time Spent on page
UX: Is it engaging?

5. Testing
Analytics: A/B Testing
UX: What to Test

6. Terminology
Analytics: Search Logs
UX: How people use language – what is your target marking using to search for you

In the case that you need to dip your toes, go old-school and simply sit behind someone and watch them navigate your website. Where do they go first, second, third, what do they avoid? Ask them why. Some data is better than no data.

Thank you to Fred Beecher and Andrew Janis of Evantage Consulting for presenting the above case study and tips on UX and Analytics during the 2010 MIMA Summit.

PoorSo SoOKGoodAwesome (1 votes, average: 5.00 out of 5)
Loading...

  • SUBSCRIBE TO TOPRANK'S TIPS NEWSLETTER
  • This field is for validation purposes and should be left unchanged.

Related Posts You May Enjoy Reading:

Please read the Online Marketing Blog comment policy

Comments

  1. Your advice makes sense… I think you’re talking about CRO ( conversion rate optimization )? Metrics can mislead any marketer and it’s important to learn how to convert all those traffic into sales. It’s easier said than done and I say it takes building your brand from scratch into a brand people can trust.

  2. This is a good post but I think you’ve only identified the tip of a very very large iceberg. What tools do you suggest for the UX side of the equation? Have you used 4Q, Kiss Insights or even Foresee with any of your clients to get the “Why?” you describe in your post?

    We tend to infer too much from our web analytics but often neglect altogether the user’s experience. Thanks for your post.

    • The best tool to answer UX questions is a UX researcher. : ) There are a number of analytics tools like Tea Leaf and Click Tale that attempt to help analysts visualize the user experience, but you still can’t understand *why* people are doing things until you actually observe them interacting with your system while they articulate their thought process.

  3. The first thing I check when i look at analytics are the traffic sources. This way I know what needs to be worked on

  4. Good reminders that data alone without understanding the “why” can lead to misleading assumptions. Better to talk to the customer directly to figure out the why then guess.

    I will be interested in a future post on automated methods to check the 6 factors mentioned above.

    • That’s just the point. Until people start using their computers via neural uplinks that send their thought processes along with commands, we still have to talk with actual people to understand the UX side of the equation. I think the closest you can get is to have UX researchers collaborate with analysts on developing metrics that monitor a site’s UX.

      Some friends of mine have developed something called UX Health Check, which you should check out: http://uxhealthcheck.com/

  5. Great article on how to discover where your users are abandoning their shopping cart, you have also taken into account CRO (conversion rate optimisation) which is very important.
    One great way to analyse this data is to create a funnel process in Google Analytics, this funnel process will identify where users are abandoning the cart, you can then make changes to improve conversions. I find that this is usually due to not wanting to login, or not wanting to provide certain information, design is also important.

    • But it still won’t tell you *why* they’re abandoning. This is why having an analysis team consisting of a marketing/web analyst and a UX researcher is so beneficial. If you notice these unexpected dropoffs in your funnel, a UX researcher has the skills and resources to find out why that’s happening. If you left it at analytics, your option would be to make changes you *think* might help, but might have no effect and might even actually make the situation worse. Analytics will tell you where the problem is. Observing and interviewing actual users will tell you why it’s happening. This combined approach will help you quickly identify what needs to change and *how* it needs to change.

  6. First off, thanks to Jolina and TopRank for live blogging our session at the MIMA Summit, and thanks to everyone for their thoughtful comments. A couple thoughts: first, yes, we consider this presentation to be the tip of the iceberg, too! There are literally limitless ways of UX and analytics working together; we really just scratched the surface. That said, our presentation did contain three case studies of how we have combined these disciplines at Evantage: improving form interaction, focusing user research, and improving conversion rates.

    If you are interested in learning more, we have posted our full presentation to Slideshare, and it can be found here: http://slidesha.re/cKKbiJ

    I also recently wrote a blog post on web analytics metrics that can be helpful to User Experience professionals, which you can find here: http://bit.ly/dcR8Q4

  7. I am a strong believer in testing – You have to test test and test some more when you are trying to build a successful blog/website. Most of us don’t get it right the very first time.

    Thanks,

    Brian M. Connole

  8. Analytics are very important to the page’s performance. At SeeVolution we think the user experience is so important that we developed a web analytics service that creates heat maps of where the users are clicking and scrolling on in real time. So now you can not only know where people are coming from and when they are leaving but you can also know where on the page they are looking and clicking on. Using this technology will allow you to place all the most important components of your site in the most trafficked areas of the screen.