How Bounce Rate Optimization Can Improve Your Search Rankings
Straight from Google's Quality Guidelines comes the mantra for the white-hat SEO community: "Build your site for users, not search engines." And Google means it.
The search giant disclosed its latest algorithm update includes page layout as a ranking factor, with a focus on how much above-the-fold real estate is dedicated to visible content vs. advertising or other distracting, useless shiny things (like splash pages).
Google states that customer feedback influenced this change, as it's no surprise that sites with buried content make for a crummy user experience, as do pages that load slowly -- another user-experience related ranking factor.
Design for Users - No, Seriously.
Google is open about its consideration of page load speed, and now page layout -- but read between the lines, and you understand that it's bounce rate that matters. And I mean bounce rate in the context of users pogo-sticking back and forth between search results (web pages) and SERPs (search engine result pages). This behavior is a strong indicator of which pages are truly relevant to a given keyword phrase, and which fail to deliver.
For example, if your page's content and incoming links suggest high relevance for the search query "learn to speak french," but you consistently have a high bounce rate and very short average time on page before users return to search to try other results, Google's simply not going to want to keep ranking you highly. Whether the reason for your high bounce rate is too much non-content above the fold, painfully slow page loads, cluttered and confusing design or something else.
So, BRO (Bounce Rate Optimization -- come on, you like this acronym) should be a priority, not just for hanging on to the visitors that arrive on your site, but to help keep those visitors comin' through search engines.
How to Improve Bounce Rates
- Page load speed optimization
- User testing
- A/B and Multivariate testing
In that order.
Improving page load speed is low-hanging fruit. It can be performed without user involvement, and takes care of one of the biggest usability issues right off the bat.
User testing should be next, not that it's any quicker or easier than quantitative methods like A/B and multivariate testing, but the insights you get from real people can help you make better testing decisions. There's no point testing a whack of site elements that never were problematic for users in the first place.
Finally, A/B and multivariate testing is the process that helps you quantify the impact of changes against the status quo, rather than just making changes and hoping for the best. Use your web analytics to determine which pages to optimize first. Remember to segment traffic source to Google, and hone in on bounce rate by keyword. Look for pages that get high amounts of traffic, with higher than average bounce rate.
Designing for users is designing for search engines. Expect more algorithmic updates in the coming months that favor pages that keep users on-site.