How Much Traffic Does It Take To Establish a Conversion Rate

WillDL · 09-11-2009, 02:26 PM

I'm a longtime lurker on the boards and I'm working on my first product. Right now I'm doing a test run with PPC to try and establish a conversion rate and tweek my salespage.

I'm curious how much traffic you think a site needs before you can confidently average out a conversion rate?

Thanks in Advance.

blase40 · 09-11-2009, 02:28 PM

More important than traffic, is the number of actual conversions that are received. It really depends on the type of product. Wait until you have sold a handful of them and then figure it.

Alexa Smith · 09-11-2009, 02:34 PM

It depends on the conversion-rate, because of standard deviations from the mean ("variance"). If you have a 50% conversion-rate (as, for example, a few people have for their opt-in), you don't need very many to measure it fairly reliably. And the closer the two outcomes (conversion/lack of conversion) are to 50/50 in their frequencies, the smaller is the number you need to measure for the sample-size to be statistically valid and have some "predictive" value. But if you're talking about a conversion-rate of 1%, you need many, many thousands before it means anything worth talking about.

Kevin-VirtualProfitCenter · 09-11-2009, 02:36 PM

OFFICIAL ANSWER: Wait until each version has had about 100 conversions. i.e.: If you are testing 3 sales page variations, wait until you have about 100 sales from each.

RULE OF THUMB IF YOU QUOTE ME ON THIS I'LL DENY EVER SAYING IT VERSION: If the source of the traffic is the same for all versions you can often eliminate one or two versions from testing earlier to get good results from the remaining versions.

For example: Let's say you have 10 squeeze page versions you are testing. They have all received 50 visitors from the same source of traffic. 7 of them have 20 or more conversions--with the highest being 32. One has 5 conversions, one has 3 and one has 1 conversion.

While technically the lowest three could still pull it out and end up showing great improvement, they are so far below the other 7 I would probably pull them from the test if I were concerned about getting good results.

(By the way...we just happened to do a complete 17 page study on Squeeze Page Conversion Factors that you can get at my sig below...hint...hint...)

ItsDubC · 09-11-2009, 03:04 PM

Good info here. I had always wondered how people confidently measure their conversions. I would say that it does depend on the product though.

09-11-2009, 02:26 PM	#1
WillDL In House In a Fez War Room Member Join Date: Sep 2009 Posts: 267 Thanks: 39 Thanked 103 Times in 52 Posts Contact Info	How Much Traffic Does It Take To Establish a Conversion Rate I'm a longtime lurker on the boards and I'm working on my first product. Right now I'm doing a test run with PPC to try and establish a conversion rate and tweek my salespage. I'm curious how much traffic you think a site needs before you can confidently average out a conversion rate? Thanks in Advance.

09-11-2009, 02:28 PM	#2
blase40 www.justinblase.com War Room Member Join Date: Nov 2007 Location: South Dakota Posts: 511 Thanks: 16 Thanked 28 Times in 16 Posts Social Networking	Re: How Much Traffic Does It Take To Establish a Conversion Rate More important than traffic, is the number of actual conversions that are received. It really depends on the type of product. Wait until you have sold a handful of them and then figure it.

09-11-2009, 02:34 PM	#3
Alexa Smith Wordsmith (& Skepchick) War Room Member Join Date: Sep 2008 Posts: 13,652 Thanks: 7,514 Thanked 9,553 Times in 4,952 Posts	Re: How Much Traffic Does It Take To Establish a Conversion Rate It depends on the conversion-rate, because of standard deviations from the mean ("variance"). If you have a 50% conversion-rate (as, for example, a few people have for their opt-in), you don't need very many to measure it fairly reliably. And the closer the two outcomes (conversion/lack of conversion) are to 50/50 in their frequencies, the smaller is the number you need to measure for the sample-size to be statistically valid and have some "predictive" value. But if you're talking about a conversion-rate of 1%, you need many, many thousands before it means anything worth talking about.
	Alexa Smith ... ... writes stuff that snaps, crackles and pops - even if it's only about cauliflowers.

09-11-2009, 02:36 PM	#4
Kevin-VirtualProfitCenter Full Frontal Lobe Nudity War Room Member Join Date: Aug 2009 Location: Knoxville, TN Posts: 912 Thanks: 519 Thanked 492 Times in 205 Posts Social Networking	Re: How Much Traffic Does It Take To Establish a Conversion Rate OFFICIAL ANSWER: Wait until each version has had about 100 conversions. i.e.: If you are testing 3 sales page variations, wait until you have about 100 sales from each. RULE OF THUMB IF YOU QUOTE ME ON THIS I'LL DENY EVER SAYING IT VERSION: If the source of the traffic is the same for all versions you can often eliminate one or two versions from testing earlier to get good results from the remaining versions. For example: Let's say you have 10 squeeze page versions you are testing. They have all received 50 visitors from the same source of traffic. 7 of them have 20 or more conversions--with the highest being 32. One has 5 conversions, one has 3 and one has 1 conversion. While technically the lowest three could still pull it out and end up showing great improvement, they are so far below the other 7 I would probably pull them from the test if I were concerned about getting good results. (By the way...we just happened to do a complete 17 page study on Squeeze Page Conversion Factors that you can get at my sig below...hint...hint...)
	The Miracle Squeeze Page: Scientifically Proven to Convert 84.1% of Visitors into Subscribers Works for Anybody, Any niche - It's so simple you'll wonder why everyone isn't using it Click Here Now to Download It!

09-11-2009, 03:04 PM	#5
ItsDubC Active Warrior War Room Member Join Date: Aug 2009 Location: Houston, TX Posts: 70 Thanks: 7 Thanked 6 Times in 6 Posts	Re: How Much Traffic Does It Take To Establish a Conversion Rate Good info here. I had always wondered how people confidently measure their conversions. I would say that it does depend on the product though.