Improving PPC Experiment
June 1, 2006 – 11:14 am
After reading Empress's post on sharing information and Tom's post, and related comments I figured that I'd explain about an actual experiment I've got underway, regardless of whether the outcome is good or bad. I think with the blog I tend to think that I should only post up about stuff if it works, not if it fails, but of course something failing helps others realise that it may fail for them too.
So to start, the site in question is a Tattoo site online for 6-7 months. It gets around 400-500 visitors daily and makes very, very little in adsense. It could be pulling in $20-$30 a day but it averages around $4 per day. This has been a frustration (and a challenge) with me for months now, as I just can't figure out how to improve it. So the steps I've taken/tried
Single ads
One downside to using the ever popular large rectangle ad unit with Adsense is that there are 4 adverts on there, one of which is going to be the high payer, but one has to be the lowest. If you get more clicks on that lower paying advert then you're losing on on the high payer. So at first I decided that this could be the problem, and on the busiest page of the site I changed the large rectangle to a single half banner. I also channelled this single advert to track it. At the same time I added a 2 advert horizontal banner to the front page as the magazines that were on there have made no sales whatsoever. After a couple of days, the channel on the single ad told me that my income from that ad was producing a lower than average PPC. So I then figured I should have channelled the advert as it was before, first, to get a better idea. I changed this back to the large rectangle and continued to see a lower than average PPC.
Channelling
Besides using channels for whole sites I've never really gone further with them, however I'm now using them for individual ad units, so that I can track their CTR and PPC. It's definitely made me realise how popular certain pages are but also how quickly people are leaving the site when there's a high CTR. Of course these people may just click 'back' to the search engines too, and the way the page in question is laid out, it should be obvious that the links they're clicking are to external web sites, so I guess if they were interested in looking around the site they would click a link to stay in the site and not leave it.
Anyhow, the channelling told me that the page I was targetting for my experiment had a high CTR, a reasonable number of clicks (should be making a few dollars on its own) but a very low average PPC. So I set to work on the page. First I used AdWords tool to check on the possible payouts for my keyphrases. The one for this page came in at over £2 for the top spot, so even with the top sites blocking the content network (don't forget this can happen), surely I can get a higher PPC than what I'm getting?
I then researched the subject a little more to write a couple more paragraphs on the page and boost the keyword content. The page is a link directory which I don't think helps (I'm actually wondering if the more external links there are on a page the lower the PPC value becomes? Just a wondering). So I removed all of the links that were originally there and started doing some manual indepth searching and found a handful of really good sites on the subject, wrote a short paragraph about each and added those to the page.
Now it's a case of waiting. The value of the page reflects on the PPC value. The value of the page is down to content, incoming links, page impressions and more. When you also make changed to your content or value of the page, you have to wait for the media partners bot or Googlebot to respider your site and recache it. Until then your ads will continue to be give the same value as before. The page was last cached on the 22nd May, and with the slow speed of Google right now, it could be another couple of weeks before it's recached.
So that's what I've been doing recently, when I've had the time. Rewriting content to increase the PPC. I'll update you when the page is recached 
4 Responses to “Improving PPC Experiment”
Sarah,
Do a search on "google adsense smartprice", I think this will answer all your questions.
In a nutshell it means your lowest performing ads on any domain will affect the PPC of all of your other ads on all your domains.
By Peter on Jun 1, 2006
Cheers Sarah! I was wondering if your tattoo site offers products or is it a forum or does it even offer free things? That might affect the click thrus as well. If people are looking for free stuff - they might not be tempted to click on links/ads that offer products.
The site might end up leaving people blind to the ads as well - if it's a tattoo site you probably have good images - which might make over-take the links.
One idea that I've see floating around on some sites is to draw a person's attention to the ad by setting up a graphic beside the ad block. I know I don't really pay much attention to google ads on sites (my bad!) but when a graphic or two is placed along side - I see them.
Just an idea
I look forward to reading more about your site!
By Empress on Jun 1, 2006
Hey Peter, I'll take a further look on smartprice, but if your nutshell description is correct, how can another site on the adsense account can get $3+ clicks?
Empress - The tattoo site offers flashes for sale as well as a free sample from each one. The page I've been working on has a 30%+ CTR so I'm not concerned with the clicks. Trust me, if I was in double figures with PPC averages the site would earning a lot of money on its own! The clicks are there, more than there, it's just the amount we're getting per click.
I've now also added in a filter list using the site AdsBlackList.com to filter out low paying adverts. Whether it'll make a change or not who knows, but it's worth a try.
By Sarah on Jun 1, 2006