What is a really great bounce rate for a product or service site? What does Good look like?

[RobertFisher]

I am really curious about a result I have never seen before. Our bounce rate went down a lot on a new site. So, what is good???

Recently, we took on a project with a company that offers a product they install for consumers and who had been in business for 15 plus years. The company is successful, has good customer base of those who have been made very happy, etc. It is not a repeat sale type of product, etc. One and done.

Their site when we began talking was roughly a year old and was not well constructed but not terrible. Most of the issues were around I frames, use of older coding, poor SEO, etc. There was not really a way to "redesign" and we built a new site.

This became a true collaboration in a B2B environment as the owner pushed us like crazy. Not the bad kind of push, the one that makes you say to your team, "Let's find a way!"

The result, IMO, was a gorgeous site. But, as you know, those are a dime a dozen. But, to get to the point, when we took over the account they had a bounce rate of around 45%. I did not see this as either good or bad, but a fact and for this industry probably not bad at all. In all honesty, I was not looking at that as a first metric I wanted to move, but it was obviously at or near the top for all the reasons we know.

So, this site is a local business, not an everyday product and gets about  2500 to 3000 uniques per month. If we compare to May of 2011/2012:

2011 2012

Total Visitors   1852         3,298

Uniques     1609   2,740

Pageviews       5,634       23,203

Pages/visit       3.04   7.04

Avg Duration     2.05  3.20

Yes, I am leaving off what we are getting, yes, I am leaving off the site. Please don't hate me. I am really wanting to see what others see with site changes and bounce rates first and will disclose. So, what's a great bounce rate? How do you know?

[Adam.Whittles]

It's hard to give a definitive answer to what is a 'really great' bounce rate. I find that bounce rates can vary greatly between industries, website types (blogs, e-commerce etc.) and even between pages.

Brian Clifton gives an indication of bounce rate levels in his book by use of a traffic light system. Less than 25% is considered green (good), 25-50% is amber (average) and above 50% is red (not good). I generally go by this rough guideline. So I guess really great could be considered less than 15% perhaps?

Hope this helps,

Adam.

[RobertFisher]

Thanks, Adam, Good viewpoint in my estimation. Before I give it all away, I want to see some more responses as I am highly intrigued by what we are seeing over time. When it first occurred I thought the result would diminish but it has not. So,.......

BTW - Crimson Penguin is one I haven't seen before! Have to elucidate some time!

[KeriMorgret]

If you suddenly got a really good bounce rate of 1% or 2%, I'd take a look at your code and make sure that there's not two analytics instances there that are both firing, or some other code that's firing twice. The doubling in pages a visit would make me want to look there. I had the same issue on my site -- had a WP site and put in a custom analytics script in the footer, but still had the regular analytics in there via a WP plugin.

[eunaneunan]

Yes same thing a happen to me with a site, tracking code was in twice, so bounce dropped to 1% 2%.

Back on topic, under 30% would be good/v.good I would say

[ToughTimesLawyer]

It really depends on your business sector.  I'm a consumer bankruptcy lawyer.  I've got a brief summary of my services on my home page and contact information.

A potential client can get all the info they need on the homepage.  So in my case, I have a 80% bounce rate, but get a ton of business from my website.

Interestingly, I have section pages, i.e. Chapter 7, Chapter 13, etc., that have much lower bounce rates.  Those pages contain much more detail and link to more in depth articles about those practice areas

[AWCthreads]

Hi Robert. I think of bounce rate as my site's pulse. Over time I establish what is a "healthy" pulse, which for us is about 50-55%. Every site will be different based on its content (products) and delivery.

Our site has a lot of products that are not shippable, but only available for store pick up. We also have a lot of products that are only able to be delivered in-state, which affects bounce rate.

After we started our product feed, our bounce rate jumped to 55-60% but our sales jumped significantly also. A 55-60% bounce rate is not great for us, but the sales have been great so it is an acceptable trade-off.

I would say we have a "good" bounce rate for what we do. Its not great, but good. I would suggest great is a rate that is trending down while sales are trending up.

[RobertFisher]

Thanks again Keri, 
I wanted to reply and thank all who have responded so far. I do not think the code is in twice and am checking it with three different sources as, frankly, Keri made me nervous.
If you look at May to May, the Total Visitors are up 78% (I see this as attributable to better CTR mostly and to some degree you would have to say it could be attributable to different market or marketing, but for May last year they did PPC all month and this year it started on the 25th with lower daily spend).

If you look at average visit duration it is up 56% and pages per visit have more than doubled - up 131%. With total pageviews, well it is 4 times. Now, when you look at percentage of uniques to total it is 87% for 2011 and 83% for 2012 so no real change.

When I add it up, I see: People are engaging the site more and deeper. They are finding something that takes them in and makes them want to look at more. Obviously, if this is true (I am open to it not being) you would assume the bounce rate would go down.

Or am I off here?

[RobertFisher]

All,

I have to admit, sometimes you guys are too good. My head of Dev was in this morning and I was speaking with him re Keri's assertion (eunaneunan posit) on two instances of GA code. Well......... He looked and said its here in header and footer. "Good bug catch!" I opened my email and Keri had followed up with screen shots earlier this morning. Well, that's why she is one of my favorites.

Does anyone know how to rebuild an ego that had gone way up with a way low bounce rate and which has now gone way down due to being, uh, well, uh brought to reality??

While the current stats are great anyway, I had allowed those improvements to cloud my judgement on this and therefore did not look closer. (Picture me wiping the rose coloring from my glasses!)

Thanks to each of you as usual. I cannot imagine a world without Mozzzzzzzzzzzers!

Keepin it real.

[eunaneunan]

So I'm guessing you had a 1-2% bounce rate? If so  that brings up another question.  If the code is in twice therefore bounces can't be record properly, then surely bounce rate would be 0% not 1%-2%.

The only thing I can think off is some users are bouncing before the page is fully loaded (eg header but not the footer).... just a thought

[RobertFisher]

Actually better! 0.7% WOW. ...if only...

Interesting thought on header/footer though our JS is at bottom so not a 'slow' load.