| 2:55 pm on Jan 7, 2005 (gmt 0)|
After about 3 months of trying a few affiliate programs here and there and losing about $200, I finally found a program where I'm making some good money (PPC -> merchant).
I think what I might do is make a crappy one page site and the only way out is back or clicking the desired link. This is my plan since PPC -> merchant is working so well I want them off as quickly as possible.
I was also thinking of doing some cookie stuffing just in case they hit back and go to the real site. However, I have never done this before and not sure about the ethical complications, maybe someone can explain how you feel about cookie stuffing and what your plans are if you used to go PPC -> merchant (assuming you aren't going to pay for the top spot).
| 4:27 pm on Jan 7, 2005 (gmt 0)|
Just be careful of the cookie stuffing method you use, looks like any using frames or an iframe might be caught as they will be checking for that already in the new TOS?
| 11:47 pm on Jan 7, 2005 (gmt 0)|
You think they are going to check every page for iframes that are used for cookie stuffing and not noticeable right away?
I doubt it, unless there is a system in place to display any iframe code while reviewing a page. Anyhow, not sure what I am going to do yet as I am not sure how successful I will be at converting traffic via a landing page.
| 12:05 am on Jan 8, 2005 (gmt 0)|
I setup a bunch redirects - but it doesn't look like those will be allowed... so landing pages it is... which will affect conversion somewhat but ohwell - the price of growth.
the other thought is to buy a couple small websites and use them for content
| 2:30 am on Jan 8, 2005 (gmt 0)|
Cookie stuffing is against the terms of many
affiliate programs, just be careful or you could
find your affiliate account getting terminated. I
know of people that are actively looking for
affiliates doing cookie stuffing and reporting them
to the merchants, and some affiliates have gotten
their affiliate accounts terminated for this.
| 2:34 am on Jan 8, 2005 (gmt 0)|
Also, for everyone that thinks using landing pages
will lower conversion instead of linking direct to
the merchant. I have been on the ClickbankSuccess
forum today, and some merchants have posted their
saying that most of their "top" affiliates have
already been using their own landing page before
this change, and seem to have a better ROI than
ones that just use the hoplink. I read all the time
where affiliates are complaining about keywords being
slowed or stopped. I've been sending my visitors to
my own site for a long time now, and very rarely
have any problems with my keywords. I think you may
be surprised with the results you will get with your
own landing page, compared to using the same domain
50 other people are also using.
| 2:18 am on Jan 9, 2005 (gmt 0)|
My experience is that direct to merchant is much better. I had landing pages and they clicked through to the merchant at 50-60%. Direct to merchant is 100%. It is not a contest. How can you beat 100%!
| 5:38 am on Jan 9, 2005 (gmt 0)|
If you set up a good landing page you can get that up to about 90+ percent.
I have set some up on overture that got high clickthroughs...
Dont have too much content on the landing page.
You want them to read words that they have to click on such as "Enter Now" in 60+ font lettering.
If you have too much pictures, flash movies, etc...they might hit back before they all load.
I would have the whole page in text format with some small graphics if needed.
| 7:39 am on Jan 9, 2005 (gmt 0)|
Interested in thoughts on this approach:
In each ad group have 2 ads, one going direct to merchant, one going to yourdomain.com landing page.
On a given search:
Would Google pick up the direct to merchant ad for more obscure keywords that neither the merchant or any other affiliates for that merchant bid on (or bid too low)?
Conversely, if I were the loser on my direct to merchant URL, is there still a chance Google would show the mydomain.com landing page for that search anyway?
Guess what I'm saying is would both ads in the adgroup be evaluated for a single given search?
We're working on the assumption that direct to merchant converts better which I have not always found to be the case.
| 8:44 am on Jan 9, 2005 (gmt 0)|
patient2all, this is an excellent question and one which I have been pondering. Your question is however slightly wrong (IMHO), you couldn't have "direct to merchant" and "landing page" in the same ad group as they would need different display URL's. Therefore you would need to create two sets of ad groups, one for "direct" and the other to "landing", each with the same set of keywords. "Direct" would have a higher max cpc.
Then we still have the $64 million question, if the "direct" ad failed as there was a higher bid elsewhere then would the "landing" ad still get a shot?
I'm guessing that it depends on the sequence that Google does things. Does Google generate the whole list of adverts (in which case each advertiser only gets one chance) then deduplicate at the end or does it calculate and fill each slot individually. I bet it's the former. If so then does this make a second adwords account essential? One for "direct" and the other to "landing"? Then you run the risk of having two ads diaplayed and potentially clicked on for the same term! Indeed, are you allowed to have two accounts or one in my name and one in my wife's?
AWA, any comments?
| 10:27 am on Jan 9, 2005 (gmt 0)|
I've been using 2 distinct display and 2 distinct landing URLs simultaneously for some time now in the same ad group. Google would have got me by now if that was a no-no.
Shown below is how it displays on the bottom of the ad group page. In this case, direct to merchant won out, but mydomain.com was far from a complete failure. In fact on one particular day, 13 clicks to mydomain.com resulted in 13 sales. I'm not expecting that to happen too often though!
The data gets skewed a little bit because Google will start favoring the ad with the higher CTR after a while. The ad texts are identical, the only difference is that the destination URLs are completely different and by default, the display URLs are different too.
I believe a lot of our fears are that people see FamousCompany.com in a display URL and are more likely to click that than MyNobodySite.com. However, we have to hope that we're all on more equal footing when the majority of the ads showing are from display URLs that are unfamiliar to the user. Our domain names must now be enticing, certainly not geocities pages (even though some users may be impressed by those)!
Try this test like I've been doing. Quick :)
226 Clicks ¦ 3.4% CTR ¦ $0.16 CPC
Served - 41.8% [more info]
438 Clicks ¦ 4.9% CTR ¦ $0.16 CPC
Served - 57.4% [more info]
I've got about a 100 questions like this!
| 11:05 am on Jan 9, 2005 (gmt 0)|
Thanks, there's obviously a gap in my knowledge! If you have more than one ad in an ad group does it rotate them automatically?
You have hit the nail on the head though. Myhumbledomain (quite catchy I think!) might have lower click thrus but if the monetisation on each click thru is higher (adsense/additional merchant links) then you could increase the max cpc and get the same or a better ROI.
I did wonder if a sub-domain on a free hosting account might add some repectability.
| 1:45 pm on Jan 9, 2005 (gmt 0)|
By default, Google will show the ad with better CTR
more often. However, you can change the settings so
that both ads will show the same amount of
impressions. Go into your adgroup, click on 'Edit
Campaign Settings', and then uncheck the box where it
says 'Automatically optimize ad serving for my ads'.
| 2:15 pm on Jan 9, 2005 (gmt 0)|
|If you have more than one ad in an ad group does it rotate them automatically? |
It rotates them automatically, but gives greater weight to the ad with the better proven CTR. I don't believe I ever started two out at the gate at the same time; the second always came into play a bit later. It would appear that the second ad then starts at a slight disadvantage since it has a 0.00 CTR where the first ad likely already had a proven CTR.
You can see in the example that I gave that the initial direct to merchant ad had more showings and a higher CTR. I'd be inclined to say that's because it had a head start gaining clicks and started out as the favoured ad. A true back to back test is needed to prove this however. Unless one has a truly commanding second ad, it becomes a bit harder to outpace an ad which has already garnered a respectable CTR which it gained partly because it had the advantage of a wider pool of impressions.
I'd say there is every reason to think if two ads start at once, the rotation would be exactly even until one outpaces the other. It appears to operate as a subset of the campaign feature "Automatically optimize ad serving for my ads."
I believe sub-domains, if not engaging in trademark infringements, can be an invaluable asset in the new reality. I asked my wife tonight to try to help us think of a short, catchy available domain to which subdomain prefices could be attached elegantly. That may help conversions. Suppose we can't call it googlesales.yoohoo.com however....
I'm determined to make this Draconian punishment thrust upon all affiliates based on the actions of a few work for us, unfair and anti-competitive though it may be. Ultimately, this may become an issue for the courts as it reeks of a blatant attempt to stifle competition.
However, to continue to eat we must think creatively and make campaigns work in spite of these obstacles. It's also a good time to give some of the more maligned smaller PPCs another chance which I've been doing on a small scale.
I only hope that upon implementation, the rules are laid out with greater detail than what the email offered. IMHO, it raised far more questions than it answered.
Let's also hope that AWA can come back with answers that don't overuse the word "probably" and this doesn't involved the frequent unresolved "pings" to the tech team that more detailed questions tend to get buried in.
Pardon my current skepticism, but I can't help associate the fiasco of a couple of days ago where long destination URLs with tracking information on the end were getting mangled and the fact that currently the "Ad Diagnostic Tool" is pretty much "lying" when queried about ad showings. Makes one wonder if the enforcement of these changes are currently being beta-tested in the live environment....
One last question for this session:
How will the advertiser know if their well planned campaigns have the chance of a snowball in Hades of actually getting an impression?
Is there any chance of current high bids being public knowledge a la Overture?
Will the historically shaky "Ads Diagnostic Tool" be capable of reporting the new reasons one's ad has no chance of showing?
Will outbid, never shown keywords still retain their active status indefinitely so they have a fighting chance if the winner fails to maintain a sufficient CTR despite their high bid or blows through their budget?
That was more than one last question. Somebody please, teach me how to write a "post" instead of a book!
| 2:18 pm on Jan 9, 2005 (gmt 0)|
That feature is campaign wide, not ad group wide to the best of my knowledge. If one had a campaign where they were selling multiple widgets with various adgroups for the different types of widgets, the widget adgroup with the best CTR is not always the one you want to favor. It might be a $1.00 widget where selling 5 of those is not as lucrative as selling one more expensive widget in a different ad group in that same campaign.
What we're talking about here is that additional subset that is allowed <b>within an adgroup</b> where multiple ads may be utilized. This is mostly used to gauge the success of ads for the same product since the keywords are shared.
I don't know of any feature which controls optimization at that level, but it's unlikely you would want one. In this case, you'd hope the best selling ad sold the most widgets that those shared keywords attempted to advertize.
| 3:54 pm on Jan 9, 2005 (gmt 0)|
I bet there will be alot of affiliates clicking on their own ads and having some other people click as well from different IPS. (the ones that don't want to purchase their own domain).
To improve their CTR.
That is funny.
| 4:03 pm on Jan 9, 2005 (gmt 0)|
Very good points raised by patient2all and the others. I just wish Google sent a more detailed email or set up a FAQ for these changes.
One thing that should be taken into consideration is branding. Fair enough DirectToFamousMerchant.com will get more click-throughs, but MyHumbleDomain.com will always remain MyHumbleDomain.com if it is not marketed. If you use FamousMerchant.com in your URL in Google Adwords, you are building their brand; if you use MyDomain.com in the URL slot, your website will gain brand recognition, even if people don't click on it.
We have always used landing pages and our own URLs. We have spent a significant amount of money building our brands. The result is that, in the long run, we are in a better position than those who just use somebody else's brand. Even though I work in my pyjamas from home (don't tell anyone), because of the massive efforts put into building a brand, I regularly get contacted my multinational companies asking to advertise on my sites.
So, at the end of the day, even though these changes are going to have a severe impact on the direct to merchant affiliates and I think they are grossly unfair, think of it as a blessing in disguise and maybe take the opportunity to build a brand that you can live off in the long run. Direct to Merchant business models put you at the whim of Google/Overture etc. and no business should be dependent on others to such a degree. What if Google were to increase their min. bids from 5 cents to 10 cents? And then 20 cents? Therefore better to focus on your own content sites.
| 4:12 pm on Jan 9, 2005 (gmt 0)|
PeteM said: you couldn't have "direct to merchant" and "landing page" in the same ad group as they would need different display URL's.
why is that? I see no reason why one ad group can't have two ads, one going to merchant's site and one going to your own website landing page? Each ad has same title, keywords and copy, but different URLs (obviously) AND different display URLs too.
what is the problem?
| 4:47 pm on Jan 9, 2005 (gmt 0)|
I was wrong and hadn't realised that you could have two ads in the same ad group.
| 4:53 pm on Jan 9, 2005 (gmt 0)|
whew! had me panicking then! :-)
I have just started switching a huge 240 ad-group campaign in this way.
now each ad group will have one ad going to merchant and one ad group going to mysite.com landing page.
| 9:38 pm on Jan 9, 2005 (gmt 0)|
|now each ad group will have one ad going to merchant and one ad group going to mysite.com landing page. |
Are you doing that on the theory that the merchant ad still has a chance of being shown on some keywords? And if not, then mysite.com will have the same chance it always did to be shown?
That's the strategy I suggest we pursue, I'm just wondering if the new Google algo will act that way upon encountering two ads. Is it willing to evaluate both during the same search toward the goal of allowing one in?
| 10:16 pm on Jan 9, 2005 (gmt 0)|
|now each ad group will have one ad going to merchant and one ad group going to mysite.com landing page. |
Sounds like a good plan to me. All we need is for AWA to confirm that it will work.
Can't help thinking that Google was better before all this. All they had to do was can the spammy "dead pets" ebay sites and ban dynamic keyword insertion.
| 11:19 pm on Jan 9, 2005 (gmt 0)|
I agree that google could have enhanced the user experience more effectively by banning dynamic keyword insertion and stopping ridiculous ebay ads. Unfortunately, dynamic keyword insertion is extremely profitable for Google and so will having multiple landing urls that will compel the user to click on each of the paid search results.
They can boost their profits tremendously over time by taking the actions they took. I for one could think of a multitude of better ways to enhance the user experience.
| 11:37 pm on Jan 9, 2005 (gmt 0)|
To the original question: What will you do?
I'll be occupied for the following week in making some pages with hands on tutorials for the products I'm promoting.
I'll also be buying some new domains, not myhumbledomain.com, but WhatAWonderfullWidget.com types of domains.
Then I'll be searching for merchants that allow me to send traffic directly to their "buy" page.
| 12:42 am on Jan 10, 2005 (gmt 0)|
It wasn't only ebay ads. Like I said in another
post, I searched for 'dog training' and on the
first 5 pages there were 19 ads for the
example.com domain. I could find many more
examples like this with people promoting Clickbank
[edited by: eWhisper at 1:52 am (utc) on Jan. 10, 2005]
[edit reason] Please no URL drops. See TOS. [/edit]
| 1:02 am on Jan 10, 2005 (gmt 0)|
Certainly I think the E-bay ads did this in, but the dynamic keyword insertion does have its legitimate defenders. However, in useless retrospect, I think a cap of perhaps 1,000 on the dynamic keywords still would have let the sophisticated users of dynamic insertion continue using scripts that pulled from their databases, but would have stoppped all this uploading the dictionary jazz.
In fact, several of us pointed that out in the posts leading up to this decision, but it seems that idea was never considered.
Putting modesty aside for a moment, I like many of us, considered myself an honest and valueable affiliate who contributed positively to the user experience. Also, contrary to the popular belief, many of us used a combo of landing pages and direct to merchant depending on the product.
The few I've converted so far to landing pages have the advantage of featuring additional merchants who bring in some extra, in effect, "free" sales.
I have to disagree with the notion that affiliates at large had plenty of time to prepare:
1) It's not always wise to jump into action based on message board "rumors" as though they are fact.
2) Much contradictory speculation came through this board during the entire "rumor" phase.
3) Far, far less than 100% of affiliates read WebmasterWorld.
4) A few month window would have been far more decent of Google once they made the announcement. Many of us have hundreds of ads, if not thousands.
I won't possibly have all my ads converted by this date of the 12th. (which is still not official best as I can tell), so I have no choice but to develop a triage system where the most lucrative ads get priority.
I hope that dormant keywords in campaigns that have no chance of showing based on their affiliate status don't suffer a change of status during the early days post-conversion.
If they have no impressions and 3 months haven't gone by, my understanding always has been that they can remain "live". Google needs to abandon that recently introduced strategy where if a keyword didn't historically perform well for other advertisers, you were given scant opportunity to prove otherwise. That history is worthless now that all the rules have changed.
One way or the other, I'll survive, hoping the same for everyone else!
| 1:14 am on Jan 10, 2005 (gmt 0)|
BTW, I'll be disillusioned considerably if AWA does not respond to this post Monday. Many, many excellent questions and issues have been raised.
Another suggestion for Google:
They should now allow for one to land on a bookmark within a page, so that affiliates with large inventories who don't have the time nor inclination to build php scripts for each campaign can more easily direct the user to the requested product. I never understood the ban on URLs with bookmarks myself.
| 3:18 am on Jan 10, 2005 (gmt 0)|
Sorry about that, I won't post any url's again.
| 7:30 am on Jan 10, 2005 (gmt 0)|
This has been covered in some other topics but if AWA does reply to this thread then perhaps he/she could also clarify the situation concerning framing. We all understand how framing (i.e. iframe) works.
However, are multiple domains pointing to the same web space allowed? Could Amazon for example create a domain called AmazonUS.com (pointed at the regular amazon web space) for associated to use? This would prevent associates and amazon themselves competing for a single slot. As an altenative to AmazonAss.com are sub domains regarded as unique? Could Amazon create sale.amazon.com? I also use PHP includes. Are these classed a frames?
I must say that it's nice to see some collaboration in what can often be a very secretive world!
| This 57 message thread spans 2 pages: 57 (  2 ) > > |