Don’t try to lower your bounce rate
Bounce Rate Optimization
When developing a conversion optimization strategy for new clients, determining website goals is obviously important. The Conversion Optimization goal should be selected based on how well it supports the website goals. This is often an area where there’s confusion about what are the priority metrics to improve.
Your Business Goals should determine your Website Goals, which should be prioritized to determine your top Conversion Optimization Goal.
Strategic Conversion Optimization goal development
Many of the metrics found in your standard Web Analytics setup are not appropriate goals for optimization, even though they may be relevant Key Performance Indicators (KPIs) for your website.
Bounce Rate, for example, is a common KPI and an indicator of the Relevance of a particular page to visitors entering the site. If your site-wide bounce rate is 43% and a popular entry page is 56%, for example, the page may not be meeting the expectations of incoming visitors.
A higher bounce rate isn’t necessarily a problem.
Example #1: In one case, the page in question may have a high natural search ranking for non-revenue-producing keywords, bringing in non-target visitors. Those non-target visitors aren’t hurting you and can be disregarded (unless you can figure out a new way of engaging them in a relevant offer). Your true bounce rate for your target visitors may be lower than you think.
Example #2: In another case, your page may actually be doing a better job of communicating your message, so more visitors that aren’t ready to convert can leave rather than navigating the site to try and figure out what you sell. This also isn’t a bad thing and may still allow for a relatively high lead generation or purchase conversion rate.
But, a high bounce rate can suggest a problem, too.
So, shouldn’t you try to lower a high bounce rate with testing if you can?
No.
If you set bounce rate reduction as your A/B test or Multivariate optimization goal, you could hurt your true revenue-producing goals.
Here’s how your bounce rate could be lowered without improving revenue:
- Changing a single landing page into a multi-page microsite experience
- Moving some information, like pricing, onto a second page
- Simply adding links to popular news or blog content
- Replacing productive offers with free offers that may have lower revenue-producing value
If you were to test these as test variations, you could easily lower your bounce rate and believe your test was a success. But that new page may have hurt revenue production while “improving” your bounce rate. Do you see how you could inadvertently hurt your results?
So, what should you measure?
Your conversion optimization goal should be as close to revenue as possible, ideally recorded after a purchase. Although the most relevant goal will depend on your website goals, here are some typical conversion optimization goals that lead to revenue improvement:
- Ecommerce purchase conversion rate
- Average order value
- Return on Ad Spend
- Revenue per visitor
- Lead generation inquiry conversion rate
- Software download conversion rate
- Trial registration conversion rate
- Click-to-call conversion rate
- Click-to-chat conversion rate
- Request a quote conversion rate
- Request a demo conversion rate
Start with your Business Goals, then determine how your website goals should support that. Only then will you be able to prioritize and select the right Conversion Optimization Goal for your situation.
March 8th, 2010 at 1:20 pm
I have a popular entry page on my website that has a high bounce rate. I am going to try and engage the user more and see what happens.
May 13th, 2010 at 4:36 pm
For some pages, such as store locators or hours of operation, calendars, and the like, visitors may bounce because they *did* find the information they wanted right away. For example, "what time is the meeting?" or "where is the nearest store?" might lead to people leaving the website, then taking action in real life.
July 2nd, 2010 at 3:01 pm
You're right, Jean! That's an example of why Conversion Optimization will never be automated. Analyzing the metrics with an eye on your visitor scenarios to understand how your site is being used will avoid jumping to conclusions. Your top bounce rate page may be fine just as it is.
Conversion Optimization is not about following rules or "best practices" tips but rather about gaining marketing insights that can be tested.
Thanks for your comment, Jean!
July 2nd, 2010 at 6:18 pm
Excellent points Raquel!
Some folks you don't mind letting out of the barn.
You've summarized very succinctly some key issues to consider when optimizing the funnel.
Another is whether the visitors close/buy. In typical B2B cycles, the purchase is not made onsite so, there's got to be a tie between your site analytics and your web based CRM.
July 2nd, 2010 at 6:36 pm
Jean,
Nicely written post and very topical for us at the moment as one of our customers has a "lower the bounce rate" target as part of the Sales & Marketing strategy for their website.
Nice to know I'm not the only one who thinks that a high bounce rate could be just what the Doctor ordered.
For example, I tend to drive lead generation on our own site. I want people to arrive, sign up and then leave. That means we have a high bounce rate but we also have a high conversion to list rate as well.
Great post.
Mike
July 3rd, 2010 at 6:16 pm
If I were you I would check whether an immediate signup is counted as a bounce, Mike. If the form and "thank you" message are embedded on the page, it might be. If the "thank you" message is on a separate page, it might not.
July 12th, 2010 at 5:13 am
It all depends on what type of site you´re running. You could of course cheat with the techniques you mention but then you are pretty stupid. If you are running a blog or a news site reducing bounce rates is a natural goal. And you might want to fight bounce rates on your home page for example.
July 12th, 2010 at 4:36 pm
On the contrary, it would be stupid to ignore the real business goals.
Reducing the bounce rate isn't a bad thing, but it doesn't necessarily lead to higher revenue conversions. It is often a side-effect of improving revenue conversions.
You mention a blog or news site, which we could generalize as publishers. If their revenue is generated from advertising, then their conversion goal should be to increase pageviews. This isn't the same thing as bounce rate either.
July 12th, 2010 at 7:38 pm
Hm, what I ment was that the examples you gave (such as "Replacing productive offers with free offers that may have lower revenue-producing value") is like you say will affect bounce rate but not business goals. I´m not calling you stupid.
July 12th, 2010 at 7:45 pm
Ha! I mis-read your comment, Jonas, sorry!