Chinalytics Longmarch Blog

Landing pages / experiences in China suck. Its official. I challenge everyone who claims otherwise to a duel to the death. While I have seen one or two exceptions, a short visit to the Sina, China’s largest portal. Will prove the point. Click on any of the ads and tell me what you think. In general I see a couple of common mistakes

1. The link goes to general page (mostly the homepage) that only loosely connect with the banner message.
2. The landing page is too long and too confusing. Its is not clear what the advertiser wants the visitor to do.
3. The landing page is executed in heavy Flash file that take forever to load and is all fluff but no content
4. The key message / action is hidden somewhere is a complicated navigation structure and not clearly visible on the homepage.

These mistakes are not unique to China, nor are the solutions. (See the “Post Click Manifesto” from the folks at Ion Interactive for an American perspective). The most effective way to an optimized landing page is testing, but testing itself comes as one part of part of an effective optimization process.

1. Identify the optimization target. This answer should come from the owner of the campaign. The person that pays the bill, the person that grew up with newspapers and TV (some education might be necessary). You would be surprised by the number of advertiser that cannot answer the questions. “What do you want the visitor to do, after he / she clicked the banner?” Often the answer is “Read the content on my website.” That is not enough. In best case you can drill down to specific actions (download whitepaper, fill in form, play video, sign up for newsletter, click on store locator). In the worst case you need to settle for metrics like bounce rate or pages / visit.
2. Measure the performance of the current landing page based on the target.
3. Develop a hypothesis, why you conversion rate is bad. Trust me it is worse than it could be. If you start with your optimization, any of the common mistakes above will serve well as a starting point. For more details about a good hypothesis take a look at “Website Optimization Starts With a Hypothesis” over at Grokdotcom
4. Develop an alternative landing page that fixes the problem and run an A/B test. Google provides as free and easy to use tool, Google Website Optimizer, that lets you run these tests without any changes to your banners. The test will tell you which alternative page performed better.
5. Then choose the better alternative and start the process form the beginning.

Going through this process, I found Semphoric’s Web Analytics Functionalism idea thought provoking. Borrowing from the modernist design credo “form follows function”, Semphoric theorized that each page has a role (function) and the most effective design will always be the one whose design (form) allows it to fulfill the function most effectively. Read their whitepaper to get a better idea and some great idea for hypthesis why our page does not work as well as it could.