Support site improvements via scientific experimentation
One of the great things about the web is that you can try things in a limited fashion, see how they work, and quickly learn and react accordingly. When an experiment is a success, you can expand the approach; when it fails, you can revert quickly.
The Cisco.com support team recently put some of this scientific method to work on the support site.
We started with some observations: Users struggled to find forums and other community content on products. This pattern of frustration was evident in the metrics, user feedback and in user testing, and it led us to formulate a hypothesis which could be tested.
For this trial, we selected the top ten product support pages based on monthly visits. These pages already had links to the community section, but the links were rather generic and did not consistently present useful information at a glance.
We gathered the pre-experiment data for these pages as a point of comparison and then added links on these pages to the specific forums and discussions so users could easily see some discussions that might be of interest when browsing other product related content.
Our theory was that more specific links that lead directly to relevant discussions on a product would be more useful and, therefore, more frequently explored. This would be good for users, because they would find helpful information more easily, and it would be good for our communities as they would get greater exposure and use.
The metric for success would be visits to the community originating from these 10 pages.
This little experiment has been running for more than a month and the numbers are extremely positive. On these pages, the traffic to the community forums has increased several fold.
By our indications, it has been fantastic success and we are making plans to propagate this approach to other product pages in the near future.