ZoomMetrix.com
The field of User Experience (UX) is a very interesting field incorporating various aspects of studies including psychology, anthropology, sociology, computer science, graphic design, industrial design, cognitive science, etc. It is far more deep and complicated than my brief description, but here are the typical outputs from UX practices (via Wikipedia).
Site Audit
Flows and Navigation Maps
User stories or Scenarios
Persona
Site Maps and Content Inventory
Wireframes
Prototypes
Written specifications
Graphic mockups
Some of these outputs seem overlapping with IA (Information Architect) role, but these are pretty important practice in digital world. I want to talk about how analytics and some of the measurement tools could help web site designer make better decisions around improving the consumer experience.
First things first. You need to ask what are you trying to achieve in the analysis. Typical one and probably the biggest one is:
Removing the barrier users are facing so they can complete their task (i.e. add to cart, check-out, form complete, finding information)
In my opinion it is all about the end users experience. You can use the tool in many you want to know or understand things, but at some point, you need to take action against those findings.
There are four types of tools that I’ve used that are particularly useful to measuring and addressing the UX.
- Web Analytics tools (Omniture SiteCatalyst, Webtrends Analytics, Google Analytics, Unica, etc.)
- Heatmap or interaction tracker (CrazyEgg, ClickTale, Mouse Flow, etc.)
- A/B testing tool (Google website optimizer,
- Qualitative measurement tool or Customer satisfaction measurement (Foresee, iPerceptions, NPS, etc.)
Here is an example flow of the usage of these tools. I’m sure there are more creative usage, but just as an example…
1. Web Analytics Tool shows very high bounce rate on a particular page that is getting a lot of traffic
2. Customer satisfaction tool tells the analyst something about the site and that segment of page/category with very bad satisfaction score. Diving into a tool like iPerception tells the analyst that the users can’t find something that they expected to find.
image source: iperceptions.com
3. Analyst goes into the Heatmap tool and looks at ever aspect of the user engagement from mouse over, clicks, browser scrolls, and recordings on a particular segments that engage with that page. Hypotheses are created out of this analytical practice.
image source: crazyegg.com
4. A/B test is performed on that page with better navigation, call to action that speaks to the majority of the page visitors.
Here is a video from Google Website Optimizer
After the A/B test was performed, marketers rolled out the new creative and reduce the bounce rate, and task completion rate increased by X. Satisfaction score went up, and everyone involved are heroes.
It may sound like an expensive practice, but a lot of these technology could be implemented at a relatively cheap cost. The biggest challenge is the process and the resources in place that enables such execution in a rapid fashion. For example…
- Google Analytics is free
- Mouse Flow is free up to certain number of recordings (you can apply the code to just the page you want to analyze). CrazyEgg starts at $9/month for 10,000 visits
- Google Website Optimizer is free doing A/B or MVT testing
- 4Q iperceptions is free and it could be a good start
In a lot of cases, it takes years to even get to this level of data practice. My hope is that many companies are able to take it step by step to the next level and start demonstrating the ROI of each disciplines in web analytics. Yes, there will be middle managements who’s mind sets are stuck in maintenance mode, budget/resource constraints, etc.
I’ve heard great stories from some folks where they were able evangelize the data practices by doing many rounds of POC (proof of concepts) using various applications. Then they’d take that lift in traffic, sales, outcome, downloads, etc. Basically, showing the effort relative to the lift in bottom line. People get excited when they’re involved in that journey and that is the fun part of analytics.
-
Enr
