This is part 2. Part 1 is here: A4D Meetup 2010 Trip Recap Notes – Part 1
In my opinion, Mike gave the best presentation of all. It was interesting, well organized, and evident the guy really knows his stuff. His style is also fast and to the point which is great and kept my mind from wandering. The topic of his presentation was PPV and most of the time he focused on TrafficVance and regarded them as having the best traffic quality.
Mike explained how traffic gets divided up: the #1 bidder on a given URL always gets 100% of the impressions. However, lower bidders will see some traffic once the frequency cap for bidder #1 kicks in. That makes perfect sense and clears up questions many people have had about why lower bidders still see some traffic.
He went on to mention that PPV is not cheap like everyone seems to think. If you really look at the CPM’s, it’s not cheap at all. What it is, is targeted. The true value of PPV, is that you can target sites that don’t normally sell advertising. He also said that to get a higher ROI, you typically need a higher frequency cap. By higher frequency cap, I am pretty sure he means that 1/72 hours gives a better ROI than 1/24 hours.
Another tip: Many people think they need to bid on a site with a large number of visitors. This is wrong. Instead, you want to bid on sites with a large ratio of your target demographic. Again, makes perfect sense. It will lower your CPM’s and may make you more profitable.
For landing pages, he recommended doing site emulation. Model your LP after the target site’s design. Do this on sites with a loyal readership to leverage the authority and trust of that site. This is something I used to do in my software publishing days. Emulating (but not outright copying) the look and feel of packaging used by established brands allowed me to outsell all of my competition by a 3 to 1 margin. Even when people know it’s not really Brand X product, they are conditioned to associate the design with trust. An extension of this idea is what affiliates are already doing. You have probably seen the landers that pretend to be television news sites. These were pretty popular and used to promote rebills. Again, the idea is to create trust.
He also recommended targeting shopping cart URL’s. The idea being they are largely ignored by most marketers but can work well because people are literally in buy mode at the very moment they see your pop-up.
Mike said he always uses a landing page when doing PPV. He uses a big benefit, and a big call to action. He then uses the interaction with the LP to gage profitability. He showed a spreadsheet he normally uses to figure out CPC after he has sent some traffic to his landers. He doesn’t use Tracking202 and basically only uses pixel tracking from networks. He stressed how page speed is crucial in PPV. This is because if your page load is too long, people will have time to close the pop-up. There was some discussion about using a CDN (Content Delivery Network) and he agreed it was a good idea. Some of the attendees in the audience mentioned that Amazon S3 can work well. For those that don’t know, a CDN specializes in handling large volumes of traffic and uses mirrors around the world so that a site visitor gets pages served from servers that are geographically close to them. This speeds things up tremendously.
He mentioned LeadImpact bids are determined by the category you select. Bryn Youngblut made a post about this awhile ago. He said on MediaTraffic he likes using the RON feature. He said quality can be a bit lower than TrafficVance but he also finds that they are less caring about what you run (think: site emulation as discussed above). He also said they are OK with resizing the pop-up window.
Mike also touched on Adbrite. He agreed, overall the traffic quality is crap, but it offers a way for you to easily retarget users. This is done by putting a cookie on your pop under. Now you can retarget with the same pop under when a user goes to a different site. You can retarget with any offer you want. He suggested Adware Removal offers. At which time, the guy from TrafficVance sitting 2 rows in front of me shifted uncomfortably in his seat 😉
For LP content, he said he has had good results with using audio and animation. He recommended some sites: istockphoto.com has some good pre-made audio. He also recommended voices.com It’s like 99designs.com in that you can put post your job and a professional voice over artist will do the work for you. Using audio calls to action can increase profits.
His talk was technical but interesting. Bottom line, the current version of Prosper/Tracking202 employs this technique, so you should be using that if you are concerned about leaks. One of these days, I may get around to ripping that bit of code and posting it here so we can all use it outside of Prosper202.
The other interesting thing he found is that hashmarks get trunctated in the referrer string. For example:
Will show up as a referrer like this:
As you can see, everything after the hash mark (#) gets truncated. So you can effectively use this to hide data. He noted that currently works. But it is something that may change in future browser versions.
Jason Akatiff (Smaxor)
The final speaker was to be Jason Akatiff, the owner of Ads4Dough, but since we ran long, he decided to jump to the Q&A Panel. Someone in the audience did ask him to do a brief summary of what his presentation would have been and he obliged. He focused on helping everyone to understand how important data is. By data, he means the info used in list building. He said on offer LP’s, all the data on page 1 gets sold. That is a powerful idea if you think about it. The networks and advertisers use every scrap of data. And that is what affiliates should be doing as well. He said to think about this: “How do I put data collection into what I am doing?” He said to think about list building paths, like IQ offers for example. Lots of these types of offers are designed to capture leads and as affiliates we should be collecting as much data as possible too. Building your own list is key.
The rest of the Q&A panel was sort of uninteresting. Dr. Ngo joined the group of Jason, Nicky Cakes, and Mike Powell. Not many people asked questions, and for some reason I stopped taking notes. So that’s it.
Overall this was a really great meetup. If you were not there you really missed out. A huge thank you to Jason and A4D for hosting this. The fact that it was free was awesome. The venue was great, the speakers were top notch, and the content was valuable. Be there next year!