One of the most commonly asked affiliate marketing related questions I’ve been getting lately has been "Are you direct linking or using landing pages?".
As I’ve always answered, I have been direct linking all my campaigns. At least, I have been since I gave affiliate marketing another go about a year ago.
Several years back, I had been primarily using landing pages (well rather, mini landing sites), but switched over to direct linking mainly due to the incredible amount of time it took me to pump out new landing sites.
Direct linking has been working great for me. It certainly has allowed me to focus more of my time on optimizing my ads, bids, and traffic, and less time on creating and optimizing landing pages.
However, as I mentioned in my previous post, my campaigns haven’t been too stellar lately. I’m certainly a far ways off from how I was doing in June and July, when I was netting around $10,000 a month in profit (sorry, sorry, I tend to live in the past).
Well, I’ve been talking to Richard on AIM a lot lately, who has been basically mentoring me recently, and one of the things he keeps stressing to me is how critical landing pages are.
I really hated (with a passion) dealing with landing pages/sites before, but now that I’m using much better tracking (CPVLab – highly recommend it), coupled with my mantra of "test everything", I decided to finally take Richard’s advice and give landing pages another try.
Hey, This Ain’t So Bad!
When I used to deal with landing pages, I would literally spend dozens of hours creating them, and hours optimizing and split-testing them. Sometimes it would take me many days, up to a full week to launch a new landing site built around an offer.
Now it only takes me around 3-4 hours to create a new landing page, and minutes to set up tracking. Making split-test variations only takes 30-40 minutes or so.
Part of this dramatic difference in time is due to the much improved tracking system I’m using (CPVLab) as I mentioned earlier.
Another reason is because I am creating actual landing pages now as opposed to mini landing sites, which is much faster and easier.
Lastly, I’m also not dealing with setting up new domains and accounts on my server and all that crap. Instead, I just picked a generic domain and am using subdirectories and individual pages.
Part of the reason I used to opt for separate domains for each new landing site was for the added benefit of SEO, but Google has cracked down so hard on low quality affiliate-related sites now that I find it futile. I also used to primarily use AdWords, so quality score was an important factor, but I don’t use AdWords much anymore.
I’m actually enjoying working with landing pages now.
Multivariate Split Testing
Something I plan on trying very soon is multivariate split testing.
There’s a really cool web script I want to try – hopefully I’ll have a review for you on it soon.
Basically, I won’t have to manually create separate landing pages for each A/B split test I want to run. Instead, I can run just 1 page, and it will automatically generate dozens, hundreds, or however many different versions and variations of my landing page on the fly.
The more amount of traffic you have access to, the more effective I believe multivariate split-testing to be, so I can’t wait to try this.
I’ve only been working with landing pages over the past 2-3 days, so I don’t have very much data yet – definitely nothing even remotely close to statistically significant.
But this post would be a bit boring without some test data, so here you go folks!
Below is a crappy campaign I ran yesterday that did very poorly (that I’ll be ditching):
You’ll notice that Landing Page #2 has a better conversion rate than the direct link.
This is a fantastic start as it means that I have already matched the effectiveness of the direct link, and can only improve on it as I can split test and improve the landing pages. This is the major downside to direct linking – you’re stuck with whatever that landing page converts at.
Below is a different campaign with a different traffic source but the same offer and landing pages:
Again you can see that one of my landing pages is slightly outperforming the direct link. Definitely only a slight improvement, but again I can only increase that from here on out.
Lastly, here’s a campaign that I launched yesterday afternoon that has only 11 conversions, so very little test data, but has huge potential:
At first, it doesn’t look very impressive. Sure, there’s a 300% improvement in conversion rate on the second landing page compared to the direct link, but check this out:
This is a screenshot of the stats from the campaign’s landing page targets with conversions:
Yes, this is very premature to show with such a small amount of data, but the early numbers do show incredible promise for this particular campaign.
I have already scaled this campaign up slightly, and am just currently waiting for it to start running. I can’t wait to see how it does – I’m really crossing my fingers that I found something here.
And so, there you have it. I’m back into landing page land. Let’s see how things pan out over the next couple of weeks…