Conversion Rates, The right way to calculate them Options

[alan]

Although the ratio of converting visitors on your site into paying customers will of course vary with the site, the industry, and a million other factors, I've read that the typical conversion rate is like that of direct mail: Somewhere between .005 and .02.

When the pros calculate this typical conversion rate, do they divide the number unique visitors to the entire site by the number of sales to get the rate?

Or is it more usual to divide the number of unique visitors to a product's page by the number of sales of that unique product?

[nethy]

In my books, it depends onwhy you are doing the calculation.

If you are trying to benchmark against other industries/businesses, I suppose this would require using aggregate figures (total UV/Total conversions or total conversions from ppc or email marketing etc.)

But benchmarking this way canbe extremely deceptive outside of context. I don't mean this in the general 'there isnopointlooking at metrics out of context.' I think that many metrics & indicators that most on this forum seem to scoff at (rankings, time on site,bounce rate...) can be very useful when used rationally.

Let me give an example of a context:
Say you're measuring conversion rates from ppc and you are trying to generate leads from the campaign. Say carpet cleaning leads, you are a carpet steem cleaner

You initially target only extremely specific keywords:

[carpet steem cleaning Brighton]
[carpet steem cleaners brighton]

-Users are prompted to fill in a quick contact form on landing
-any keyword that does not convert at more then 5%, you eliminate.

In this way you manage to have an 8% conversion rate. That is high compared to the average for lead generation websites.

Now, in this scenario the keyword bids are probably quite high. Other carpet cleaner IMs will be on tothis and also bid high. Also you have limited yourself to the best keywords only.

Say you decide to switch strategy you add along list of keywords
carpet stain removal
rental renovation
flat maitanence
etc.

You bid low (on tenth of what you bid for the others) and get lots of cheap & nasty,low converting clicks.
Your site wide conversion rate has plummeted to >1%, low compared toother websites.
But, the number of leads you get has gotten bigger and your google bill stayed about the same.

Conversion rate outside of context is not a very useful metric. That's why it is not agood idea (it think) to benchmark too seriously. Except maybe to calm a client that just realised that 98.6% of visitors to his site left without paying.

[alan]

Conversion rate outside of context is not a very useful metric. That's why it is not agood idea (it think) to benchmark too seriously. Except maybe to calm a client that just realised that 98.6% of visitors to his site left without paying.

Thanks for the very complete answer. But maybe I should have given more information as to why I was asking the question.

I own my own site, and I don't do much PPC. Almost all of my site referrals come from organic listings. So I'm not very concerned about click costs.

I'm in the process of tweaking some of the pages on my site to try to get them to convert to sales better. I know what my site's overall conversion to sale rate is. And I also know what each page's conversion to sale rate is.

If I knew if the commonly quoted conversion rate of .005 to .02 is usually based on an entire site's conversion rate, or if those numbers (.005 to .02 ) are usually based on a specific page/product's conversion rate, I would have a better idea whether my specific pages were converting well, or not.

It is quite possible that my product pages are all ready converting well the way that they are, and tweaking them could ruin them. I would have a better idea about this if I knew the answer to the previous paragraph.

[Randy]

As everyone has mentioned above, when you start talking about conversion rates it quickly becomes difficult to apply them across multiple sites or multiple markets.

As a general rule though, I'd say your 00.5 - 0.2 numbers are probably talking about overall site conversions. I would personally expect CR for individual pages or even a series of pages (funnel) that lead to a sale to be considerably higher. And I'm giving the benefit of the doubt that 0.05 is one-half of one percent and 0.2 is a 2 percent conversion ratio.

When I get into optimizing individual pages or series of pages for conversion I generally want to get my conversion rates for those specific sales funnels up into at least the 8-10% range. Or .08 to .10. I've found that's usually a very reachable goal with a bit of tweaking and testing.

If you get a really good one (well qualified traffic, good offer, something people want/need, reasonably priced) it's not unusual for me to see conversion rates for these pages or funnels jump up into the 15-20% range.

Does that help?

[alan]

If you get a really good one (well qualified traffic, good offer, something people want/need, reasonably priced) it's not unusual for me to see conversion rates for these pages or funnels jump up into the 15-20% range.

Does that help?

Randy, yes, that helps a great deal - thank you!