60 million page views per month and very low CPM

optimizing a photography site

[PhilipGreenspun]

I'm the CEO of a photography community with about 550,000 registered users and 60 million page views per month. I started the site as a hobby in 1993, turned it over to some business geniuses in 2000 (they went bankrupt after about a year), and took it over about six months ago to try to fix the site and what has now become a business.

We have a pretty good audience. The surveyed readers reveal themselves to be gearhead guys, age 40, with incomes of close to $100,000 per year and spending on photography of at least $2-3000 per year.

Looking at our site, most of our ads are garbage and they are repetitive, especially from Google. These are ads that our readers wouldn't want to see even once and certainly wouldn't be likely to click on after the 100th repetition. eCPM from google is very low.

I know that we are doing something very very wrong, but I can't figure out what it is. What should we be doing? is it possible that we have gotten into a vicious spiral where our clickthrough rate was low so Google is giving us crummy ads and that drives our clickthrough rate down even lower?

[edited by: eljefe3 at 11:41 pm (utc) on Dec. 28, 2006]
[edit reason] edited out URL's [/edit]

[potentialgeek]

The problem with a photo site is people are looking at the photos, not the text ads. Their eyes are absorbed with the pictures. If you have text content, people are actually reading and may read an ad that may interest them. I have a photo site and clicks are very, very low.

What is your ad placement?

p/g

[PhilipGreenspun]

Thanks to all who've responded so far for the interesting ideas. We struck a deal with Gorilla Nation at the end of the summer where they would have the exclusive right to market display ads and Google would be just text. Supposedly this would enable them to concentrate more of their sales people on selling the big camera companies and other logical advertisers for photo.net. It might be too soon to tell if this is working. At least for readers coming from foreign IP addresses, it might be better to serve them Google display ads since Google has sales efforts in a lot of non-U.S. countries. Of course, that would require us to invest in a database mapping IP address to country (or at least US/non-US) and I'm not sure where to get one.

I'd like to quiet the site down a bit and remove the ads that really aren't working.

A big problem with having ads in equipment reviews is that it distracts people from clicking through to Amazon. Suppose that someone reading our review of a camera clicks on an ad. We just made 25 cents. Great. Had they clicked through to Amazon and bought the camera, we could have made 25 dollars.

[RonS]

Many, many more will click on the ad than will buy the camera, and you can draw attention to your Amazon ads with phrases like "Click here to look at what price you can pay today" or "Get this camera today at a competitive price with great customer service here", AND you can open your Amazon ads in a new window.

So people who are looking to buy right now may click on Amazon, while lookyloos can click on AdSense.

Amazon's 24 hour cookie policy makes it imperative that you try to send only serious/immediate/impulse buyers to them, and send the rest on to look at sites through another affiliate or to AdSense.

Good luck.

[gregbo]

Of course, that would require us to invest in a database mapping IP address to country (or at least US/non-US) and I'm not sure where to get one.

Well, there are several (ToS prohibits me from mentioning names). However, geotargeting is a best-effort service; plan for mistargeted traffic. If you can live with a rough cut of traffic, you can just take address allocations from ARIN [arin.net], which will give you the US, Canada, and some Caribbean islands.

[BaseVinyl]

The website would need a major redesign...no offense intended but the layout and current position of adesense and cute dog pictures and text size is really cumbersome...

But this is all a very good and likely effective PR stunt!

;)

[RonS]

Yes, I have an IP-to-Country mapping DB in use on my site; it cost $0 and I had it set up and running in an hour or two.

I don't use it to dynamically change ads, but that's a great idea. I have a different use for it.

I suppose it could also be used to send people to the different country Amazon sites, too! But my Amazon business and foreign visitor quantity is too small to worry about capturing affiliate income from non-US visitors.

This is an idea I will keep in the back of my mind. Thanks!

[potentialgeek]

A big problem with having ads in equipment reviews is that it distracts people from clicking through to Amazon. Suppose that someone reading our review of a camera clicks on an ad. We just made 25 cents. Great. Had they clicked through to Amazon and bought the camera, we could have made 25 dollars.

You should be able to make more than $0.25 for a camera click via Adsense. I don't know how well your Amazon ads are converting (I never had much luck as an Amazon affiliate), but if you have not been SmartPriced, you should make much more per Adsense click.

Spotted this comment from Google which led me to revisit this thread. It was back when Smart Pricing was first introduced, but apparently the same principle stands today:

"We may determine that a click on an ad for digital cameras on a Web page about photography tips is worth less than a click on the same ad appearing next to a review of digital cameras," explained a Google spokesperson.

(Pamela Parker, "Google Unveils 'Smart Pricing' for Content Ads," ClickZ, April 1, 2004)

It's been said many times here, but Adsense is all about experiments. You've just got to keep testing (placement, colors, etc.), and never assume you can't do better. On one photo-heavy site I thought the page design and ad placement was doing very well. A year later I added graphic design. Instantly and now every day I see a 20% bonus.

Good luck,

p/g