Over 100 people attended the November 30th AIMS event titled “Web analytics: Make the connection between your marketing spend and the bottom line”.
Two presenters showed how they used Web analytics to find areas in need of a boost in conversion and gauge the improvement of the changes they made.
In between these the two case studies was a session about why hits are meaningless, and how to extract meaningful insight from data in spite of inaccuracies.
Here are the highlights of each presentation:
In the first presentation, Steven Goldhar, Sundance Media, presented a case study about how redesign of a landing page increased search engine marketing (SEM) conversion by over 400%, without any change in the keyword basket. The client was a single hotel and the conversion goal was online reservations. SEM campaign ROI increased to 1281%. With Web analytics, this ROI increase was attributable to the SEM campaign.
Think of the power of this case study in supporting the business case for additional landing page redesign and the effectiveness of SEM. Here are Steven’s 6 conversion improvement guidelines:
- Use proper conversion based web analytics.
- Use web analytics to help understand your audience.
- Have a clear and prompt call to action.
- Eliminate distracting web page elements.
- Plan ahead to overcome standard obstacles to conversion.
- Use a landing page strategy.
Next up, Alan K’necht of K’nectology, gave a very energizing presentation of geeky topics such as 1st and 3rd party cookies, definitions of hits, visits and visitors, and the inaccuracy of the absolute numbers of data.
- Don’t get hung up on accuracy of absolute numbers. Work with trends. As Alan said, who cares if your visits of 100 per day are accurate? What matters is that if you make a change to increase traffic and visits rise to 500 per day, that’s good!
- Measuring hits are not useful. Each image or download on a page counts as a hit, in addition to the page itself. A page with 10 images and 2 downloads will trigger at least 13 hits. To illustrate why hits are meaningless, Alan told us of a time when a senior leader wanted more website hits because their competitor had issued a press release that stated a website hit number that was double the hits Alan’s company was getting. Alan tried to explain why this was a totally useless number. The manager didn’t care. He wanted more hits. So Alan made hits rise. In one week, he increased hits by 50%. Great, the manager said, what did you do? Alan had added more images to the home page.
In the third presentation, Sulemaan Ahmed, SearsTravel.ca and OneDegree contributor, shared with us his firsthand experiences at increasing revenue on SearsTravel.ca by remerchandising the site after a combination of usability tests and Web data analysis. Online merchandising has been improved and they have optimized their online traffic referral sources. How successful have they been? The results are impressive.
In February 2006, year over year sales were down 30%. Now, revenue is up 100%. Here are 2 examples of how Web analytics was used to contribute to this dramatic turnaround:
- Sulemaan’s team first created scenarios about how they expected people to navigate the site. Next, they looked for pages where visitors exited in large numbers. Then they looked for reasons for the exodus. On one particular page, visitors appeared to be exiting because they couldn’t get detailed hotel information. The problem? Hotel information was accessible through a graphic button, but visitors didn’t know that. A boring but clear text link was added “View detailed hotel information” and high page exits were a thing of the past.
- Web analytics flagged to SearsTravel that a small but fast-growing city in Northern Alberta appeared to very interested in travel. Email-based campaigns have been created to specifically target this market, with excellent results.