Success with link units only

[farmboy]

Hopefully there are still a few people here who want to talk business.

I've been experimenting on some pages with removing all ad units and using only link units on those pages.

As a result, the CTR increased significantly which pulled up the overall revenue for the pages in question.

If feasible for your site, you might give it a try.

FarmBoy

[Leonard0]

CTR increased significantly

Ad Link CTR is reported differently from Text Ad CTR so probably should not be compared to Text Ad CTR.
From [google.com...]
* Link unit impressions - reported each time a user views a link unit, regardless of whether they click the links
* Link clicks - reported each time a user clicks a link in the link unit. Equivalent to page impressions for the resulting page of ads.
* Ad clicks - reported each time a user clicks an ad on the resulting page of ads.
* Ad CTR - clickthrough rate of ads (equal to ad clicks divided by link clicks)
* Link CTR - clickthrough rate of links (equal to link clicks divided by impressions)
* Link eCPM - cost per thousand link impressions (equal to earnings divided by link unit impressions, then multiplied by 1000)

It's better to compare the eCPM of Ad Links vs Text Ads, or total earnings.

I've removed some of my Ad Links because the EPC has recently dropped below that of Text Ads.
Strange, because I thought that Ad Links were part of the Search network so would pay better than Content ads.

[edited by: Leonard0 at 5:33 pm (utc) on Aug. 10, 2008]

[nrep]

An interesting observation farmboy - I have been wanting to do the same experiment, but I've been too cautious in making major changes like this.

Good job on increasing your overall revenue :)

I'd be interested to know what sort of ad unit you replaced with a link unit?

[farmboy]

I'd be interested to know what sort of ad unit you replaced with a link unit?

Typically a leaderboard or one of the larger rectangles.

FarmBoy

[netmeg]

I have a series of sites that are built around a carefully seeded custom search engine; the only ads I run on that page are a single link unit. Seems to work really well.

[stormshield]

That's interesting what you're saying. I would never thought about having link units alone. I'll give it a try (some time ;])

[Funtick]

Most probably top-ad (first one on the page) is a highest-payable (and most attractive) so... putting single Link Unit performs better than some other sequences of events (like as putting fist unit at bottom of page, second one on top, hird in center, and link unit on left)...
Fist unit (in a plain HTML sequence) has higher-value ads...

[Swanny007]

Thanks farmboy. I think I will try that. Did the eCPM go up though? A big increase in CTR will naturally mean more money, but I care more about CPM. Thx.

[Scurramunga]

I have tried it in the past and it hasn't worked for me at all.

[rannans]

i have tried as many combinations of link units along with standard ads ( both text and image) and whilst there has sometimes been a little spike in earnings, it never lasts and i always seem to settle back to my original earnings per day

this is why i am one of the belivers that there is a ceiling on earnings per account and/or site irrespective of what ad formats are used and how many you place on the pages

i only operate one major site (5 years now) so i can only base my assumptions on that 1 site

[Funtick]

When you test smth please allow exactly 1 week... earnings flow depending on weather, day-of-week, and something else...

[purplecape]

I've removed some of my Ad Links because the EPC has recently dropped below that of Text Ads.
Strange, because I thought that Ad Links were part of the Search network so would pay better than Content ads.

If that were the case, the advertisers who thought they had opted out of the Content network would be very unhappy to see their ads appearing there!

The ads you see on the Links pages are ads from advertisers who have chosen Content, just displayed differently.

I've found that link units perform very well on my home page (no regular ads there at all) and on some second-level pages. They perform less well on specific articles, where they are outperformed by regular ads.

[bumpski]

Just a thought given all of the above.

Perhaps Link Units work better on Exit pages and Ad Units work better on Entry pages. For my site(s) I would say most visitors enter at level 2, call it the primary content level, then they tend to exit from level 0 or 1, which in my case tend to be "index" or navigation pages, with less topical content. So perhaps Link Units, where visitors are navigating, and Ad Units where they concentrate/think/read (detailed content).

Perhaps the same could be said for perceived Entry and Exit points on a given page. Maybe Link Units would work better at the bottom of the page where many users may exit your site. I haven't experimented with any of this though.

It might be very nice if there was a "heat map" for Link Units versus Ad Units? Perhaps the other Ad types as well(image versus text, etc).If Google provided this additional information it might double all our earnings.
-- How about it ASA --

Adsense Heat Map [google.com]

[realmaverick]

I tried it just for fun and received about $0.01 per click. So that was that..

[purplecape]

I tried it just for fun and received about $0.01 per click. So that was that..

How long did you try it for? I've learned, by using link units over the long term, that it can take the system some time to figure out the best link phrases to present, especially if your site is not tightly focused on one topic (and I think link units work best on sites that aren't tightly focused, in fact).

Give the experiment time to run. You SHOULD end up getting a click value in line with your site as a whole, and a better CTR. If you don't, then your site isn't a good match for them.