Turn.com advertising experiment: week 1

Saturday, January 9, 2010 , Posted by Johnny Fuery at 1:22 AM

Originally Published 2007-03-12 07:30:00

I placed a single text ad on turn.com last week about this time. It actually took 48 hours or so to get approved, so it's technically only been up for the past 5 days or so. Here's a quick snapshot of my numbers thus far:

Impressions: 22,246
Clicks: 14
Click Through Rate (CTR): 0.06%
Actions: 0
Rank (ad placement): 1
Total Cost: $0

Turn's claim to fame is that, unlike Google AdSense, which would have charged me for each of the 14 click-throughs (CTs) noted above, Turn only charges for "Actions", i.e., an action that the clicking user eventually takes when they hit my target web page.

It's too early to tell whether or not this ad is effective, although there's certainly room for improvement on the 0.06% metric, telling me my ad copy could probably use a little adjustment. I'll probably draft a half dozen other text ads over the next few days in an attempt to do just that. I also definitely want to float other creative (that's advertising lingo for a graphic ad) on the network to see if that helps the otherwise dismal CTR.

Some positive things I've noticed so far about turn:


  • The site is very responsive, with the exception of the reports page. I understand that reporting probably entails a multitude of crazy table joins, but it still seemed painful to wait several minutes (I gave up and came back the next day to look at it) for a report when it actually didn't contain any data not already available in the snapshot views. Of course, this is probably because the kind of comprehensive reports available are irrelevant with only one ad running. :-)

  • The pay-on-action model rocks. It allows me to play with my ad copy to figure out what works, and I only pay for an actual consumer action. In this particular case, I'm offering a payout on a form submission, which involves the customer committing to an action in the future. I've set my payout low enough so that I'll actually turn (ohh the terrible puns that could be placed here) a profit on every payable action. Since my placement is currently #1, and, as I understand it, the payout is actually auction based like AdWords (meaning it shifts upward to maintain a #1 ad position until the per-action payment limit I've set is met), assuming I can actually make Turn some money here, I'll make money too.

  • It was really easy to sign up. I had the whole thing going in under 15 minutes. The biggest chunk of time was coming up with the ad copy, which I did on-the-fly during the signup process. The only negative was the 48 hour lag time before the campaign went live. 36 of those hours were my fault, because I had to put the Turn tracking beacon in place (the JavaScript they gave me to verify the consumer's "action" -- an inconvenience, yes, but I don't see any other way, technically-speaking, to accomplish this), so take this particular criticism with a grain of salt. Compare this to AdWords, which allows immediate (or near-immediate) activation of campaigns, but which forces the advertiser to punch in the (either search of contextual) keyword combinations to advertise using. I remember taking a half hour the first time I started an AdWords campaign a couple of years back, but went back into the system within a few days and spent a few hours punching in hundreds more keyword combinations. Turn (supposedly) does all of this for me.


And the criticisms:

  • There's no indication of where or how these ads are placed. I suppose this is true of AdWords, also, but at least with Google I can go out to the search engine and type in the key phrases I've paid for and see them displayed. Turn's rhetoric about the publisher selection process is a little vague (undoubtedly on purpose so they can tighten or loosen their publisher screening filters without announcing it to everyone), so I can't even take a shot-at-the-dartboard guess. (They do have a no porn/hate/etc policy, of course.)

  • The ad placement categories are really broad. This may be a signal of limited publishers -- it's hard to say, but that's my impression based on the short lifespan of Turn as a company. My ad, for example, is being published on sites that fit the "Computers & Consumer Electronics - Multi-category Computers & Consumer Electronics" category. And yet, somehow, that only means ~4000 impressions per day over the last 5 days. It's definitely a smaller pond than GOOG at this point. But we knew that.

  • AdWords takes longer to setup (when the keyword selection process is included), but it may be that the added complexity is worth it for the added control. I'll have a firmer opinion about this in the coming weeks.

  • The need for the aforementioned tracking beacon means that I can't use the Turn network for arbitrage. That is, I can't setup a mini amazon store (via the Amazon Associates program, which is the Bezos version of affiliate marketing), create a bunch of ads on the Turn network, and then pay out when the customer buys one of the books in my suggested reading list. I suppose this would be too good to be true as an affiliate marketer, but it's worth mentioning. I believe that arbitrage and other similar schemes make up a measurable portion of Google's revenue. I wouldn't be surprised if it's a fifth of all of their ad placements.


Next week I'll be reporting on how my additional ad campaigns (same product) fared. I'll also try signing up as a publisher -- we'll see if fuery.com makes the cut. :-)



Comments

On 2007-03-25 10:45:07 Agloco said:
I enjoyed reading your blog. I’ve bookmarked your blog.

On 2007-04-10 17:54:22 Turn.com experiment/review/test - week 3 » Really Smart Guy » GeekSpeak, Real Estate, Landlording, Technology, Business Ideas, Web Marketing » Blog Archive said:
[...] my initial analysis of Turn for some background [...]

Currently have 0 comments:

Leave a Reply

Post a Comment