Online banking has become commonplace in the United States. In fact, online banking has become one of the fastest-growing Internet activities. Approximately 43% of those who use the Internet, or about 63 million Americans, conduct some sort of online banking, according to a 2006 survey by the Pew Internet & American Life Project (Salazar, 2009). With this in mind, I decided to look at a large, national bank’s Web site to see if it could be enhanced for navigation and overall ease of use.
PNC is one of the largest bank’s in the country. This was further solidified after its acquisition of National City Bank in 2008. When logging onto the PNC site, one is not greeted with numerous graphics. Instead, the user sees some Flash animation, and then numerous links to various banking-related topics under the headings of products and services; solutions and PNC’s security assurance. In the top left-hand corner of the page is the user log-in. The site also has a toolbar directing users to personal banking, small business and corporate and institutional areas.
The clean, streamlined look of the site is very appealing. It seems to almost simplify what can be a difficult topic to navigate: banking. You can easily find navigation on it, and within a few clicks I seem to be able to obtain the information for which I am looking. Yet, I do wonder if any type of testing was conducted on the site.
If I was on the team working to design and launch the PNC site, I would have conducted a multivariate test since you can test many different variations of a site or changes at once (Alt & Usborne, 2005). In addition this type of testing offers a reliable, scientifically-grounded method for understanding visitors’ preferences by using algorithms.
Multivariate testing is suggested because of the few things that stand out on the site to me: numerous links on the home page and minimal graphics. Within these tests, I would change the graphics. I may add more Flash to the site to make a bit more appealing graphically speaking.
In terms of the links, I would have pages that had fewer links and ones that had even more links. If you scroll to the bottom of the page, you will see even more links in a faint gray. There seems to be a plethora of links on the site, and I would like to see which ones are used the most frequently and which ones are not used at all. This could help determine the layout of the site and minimize the overall number of links on the home page and throughout the site.
In addition, I would also make a somewhat stronger call to action on the Web site with different page headings and titles. I would want to assure that the proper page titles have been selected, and they are terms that the average user will understand. I would want the language simplified and not simply banking terminology that only bankers would understand. The toolbar could also be changed to reflect different navigation and if any other headings should be added to the main navigation on the site.
I would then combine these different link, graphics and headline options into different pages to test.
From the above, you can see that I believe multivariate testing as well as A/B testing is a viable option for companies looking to either launch or re-launch its site. But, it is imperative to first select a testing option that best suits the need of your company. Essentially these two different tools that can help a company with its Web presence and success. Learn more about the differences between the two methods here.
The very reason I believe this is because multivariate testing is one of the most frequently used marketing tactics that has achieved positive results in terms of increased sales and services (MultvariateTesting.com, 2009). Some may say because you can test numerous options at once, you are getting the biggest bang for your buck since the algorithms used in this type of testing are accurate in determining which option or combination of options would be the most effective (Verster.com, 2010) Yet, one needs to ensure the testing is being done right.
I would advise companies to utilize professional marketers who have extensive experience in this line of work to conduct these tests. Jonathan Mendez outlines a strong, strategic approach to this type of testing in his blog. Within it, he discusses, step-by-step, how to implement such a test from designing, creating the content and implementing the code. This is a not-to-be-missed post as it points to the positives of multivariate testing and the results that can be had if done properly.
Web testing is not something to take lightly. It requires time commitment, research and creativity. But, if done properly, a company will see numerous benefits including increase in Web traffic, sales and any other goal it may have for its site.
References
Salazar, C. (2009, Jan. 28). How dangerous is online banking? MSN Money. Retrieved March 8, 2010, from http://articles.moneycentral.msn.com/Investing/StockInvestingTrading/how-dangerous-is-online-banking.aspx#pageTopAchor
Alt, B. & Usborne, N (2005). Multivariable Testing. Marketing Experiments Blog. Retrieved March 8, 2010, from http://www.marketingexperiments.com/improving-website-conversion/multivariable-testing.html
MultivariateTesting.com (2009). Multivariate Testing. Retrieved March 8, 2010, from http://multivariatetesting.com/
Verster.com (2010). Multivariate Optimization vs. A/B Testing. Retrieved March 8, 2010, from http://www.vertster.com/multivariate-testing/multivariate-optimization-vs-a-b-testing
It's the end of the road for me and Web metrics
16 years ago
No comments:
Post a Comment