Case Study: Troubleshooting Erratic Conversion Rates

Posted by Ad Hustler | Posted in Affiliate Marketing, Case Studies | Posted on 27-07-2009

I run offers in a variety of niches.  Lately, one of the niches that i’ve been running took a serious decline in conversion rates.  This niche is a product sale requiring 2 pages of information.  Page 1 is contact information and page 2 is credit card information.  Offers that converted at 9-14% were all of a sudden converting at 2% if I was lucky.  Needless to say this irritated me.  I have a lot of contacts and reached out to quite a few to see if they were running this niche as well.  Some were and I heard mixed reviews about the kind of conversion rates they were still seeing.  Since some affiliates were claiming high conversion rates and some were claiming low conversion rates this exercise didn’t prove to be fruitful.  Here are a few thoughts/theorys that were going through my head at the time (here is a peak into the mind of Ad Hustler):

  • My traffic source was burning out, hence the lower conversion rates. However, this made no sense to me as the users were still clicking through the landing page to the offer.  There was really no decline in landing page to offer click through rate so this theory didn’t make a lot of sense to me.
  • Some of the other affiliates I spoke to were lying. In the affiliate industry, you really never know who is telling the truth and who is lying.  People sometimes have vested interests in leading you down a certain path so I don’t believe everything I hear.  One thing that stuck with me though is that certain people in this industry who I really trust, were telling me that there were still good conversion rates in this niche.
  • The advertisers are blindly scrubbing and shaving the offers. Since I saw such a rapid decrease in conversion rates I figured that the advertisers were scrubbing and shaving the profitability out of the offers.  This still didn’t exactly make sense to me as some people (that I trust) were still claiming high conversion rates.
  • The advertisers are targeting affiliate ID’s or refferer URL’s to shave/scrub CERTAIN affiliates. You see this in the email submit niche quite a bit, so I was thinking this could be possible as well.  It also could explain why some people say conversion rates are still strong and others say they are weak.

To start off, I tested the last theory (that scrubbing is occurring on the affiliate id/referrer level).  I put my landing page on a new domain and sent traffic to an offer that I knew converted a few days prior but at a fresh affiliate network.  This way, I knew that my affiliate id/referrer were not on record with the advertiser.  The results were no better then previously with the old referring URL/network/offer.  This helped me shy away from this theory being the cause of my problems.

———————————————————————

I then decided to use the You Can’t Hustle A Hustler – Testing Method.   I ripped the advertisers landing page and recreated the offer on a domain that I own and am in complete control of.  I took out the advertisers forms and replaced them with my own.  Here are 2 good form tools you can use to do the same and replicate any form:

Coffeecup Form Builder
Logiforms

Since I was down to a 2% conversion rate, I figured this would prove once and for all if I was being shaved/scrubbed by the advertisers.  If this test proved an insanely high conversion rate I would be relatively sure that the problem was not on my end, but rather on the advertisers end.  (Please Note: I know that there are variables unaccounted for here such as merchant account declines, prepaid credit cards and fraud prevention tactics.  That’s not the point.  The point is to get a baseline of how this offer converts when I control the page and the forms)

Next Step: Drive Traffic & See How It Converts

Some Stats

Landing Page —> “Offer” Clicks: 200
Completed “Sales”: 8
Conversion Rate: 4%

So, the conversion rate proved to be slightly higher when I owned all of the offer elements but still not a great conversion rate and fairly inconclusive due to the unaccounted for variables.

BUT WAIT…..THERE’S MORE!

I purposely left out a statistic above.  Since this offer is a Page 1/Page 2 offer aren’t you curious about how many people complete Page 1 vs. Page 2?  I was and this proved most interesting.

Landing Page —> “Offer” Clicks: 200
Page 1 Contact Info Leads: 55
Page 2 Completed “Sales”: 8
Page 1 Lead Conversion Rate: 27.5%
Page 2 Sale Conversion Rate: 4%

Do you notice a HUGE problem here?  I do and it caused me to take some action.

Only 14.5% of the users who filled out page 1 actually filled out page 2.  That’s pretty pathetic.  85% of the users are bailing on the conversion funnel on page 2 where they need to enter their credit card information.  In a live situation, the affiliate never controls the conversion/sales funnel so what’s a Hustler to do?

I decided to better presell the fact that after page 1, the user would be taken to a credit card form that accepts all major credit cards to pay.  It seems stupid to have to tell the people that to pay for the item, they will need to use a credit card but apparently spelling it out helped the conversion rate.  I made the changes and set them live on the real offer.  All of a sudden I was back to 8%-10% conversion rates.

This testing in no way figured out WHY my conversion rate decreased.  It did however provide me with information to get it back to where it needed to be.  With more tweaks/testing I expect to see even higher conversion rates.  Unfortunately affiliates are never provided with sales funnel information.  Without that information I would have never seen this problem and hence never realized how to fix it.

Case Study Testing Funds Contributed By Tatto Media

Ad Hustler | Subscribe To Ad Hustler

Comments

  1. This might be kinda a dumb question but what do you do with the leads from your own test form. Do you email them and say “hey er um sorry here is the right link to enter your cc from”

  2. Sorry, should have read the comments in your original post, I’ll repost your response here for anyone curious:

    When we say monetize privately it can really mean anything but here’s a few ways that data can be monetized privately (it all depends on the offer of course)

    1) Sell the data to companies who need it.
    2) Use the data to build double opt in email lists. You can then monetize the lists over time.
    3) Create your own very similar product and use that for monetization.

    Depending on the offer there could be some really creative ways to make a buck without actually pushing your traffic into the merchant (who if your using this testing method, may have been ripping you off.)

  3. Very interesting results. I should definitely check more into the merchants’ landing page sales funnels. Thanks for sharing!

  4. @Conv3rsion – In this particular test I forwarded the user to a page after the credit card form saying that we have a temporary problem with the site and that the card would not be charged. This test was more about trying to see what was wrong then monetizing data.

  5. Thanks for this. I hadn’t really thought about holding the customer’s hand this way. I’ll try it out on my next landing page rewrite.
    Have a good one!
    Danger

  6. Nicely done dude. Interesting case study.

  7. Awesome post. Did you see a significant decrease in clicks from the landing page to offer clicks after adding the credit card information?

  8. This case study is money! Good lookin out Ad Hustler!

  9. Thanks for posting such solidly complete case studies. Most others allude to stuff or skip over vast chunks.

  10. Nice identification and isolation of factors causing the fluctuation in conversion rates vs just attributing it to some weird slap/shave/bad karma.

    You do know that bigger affs go on to launch their own offers in the space you’re talking about, and the other niches with pretty fat front-end payouts too, right?

    Especially since you’ve got some baseline data, you could commission a product over elance/rentacoder, get a copywriter+designer to do up the landing page, stump up the listing fees at a CPA network, get a merchant account at wells fargo or somewhere else, then let it rip.

  11. Ad Hustler’s the man. Nuff said.

  12. Better test would be to have more than 1 account on an affiliate network as even the domain used on the CC submit page can effect sales. (Good way is to just send a traffic sample to a friend you trust who has an account and hasn’t run the offer). I’ve done this before when I’ve seen conversions drop off sharply and most of the time the original conversion rate magically returns to the second account. I do believe advertisers scrub/shave/whatever PER AFFILIATE. The networks generally have no control over the advertisers side of the equation so of course they will have no real answers regarding this. They will still see affiliates getting good conversion rates and will see no problem. The only thing you can really do is jump from offer to offer milking each for all they’re worth. Split test until you find a winner, run it until it dies, then split test again. Not going to make any progress any other way.

  13. Hey how do you track, the first page,
    and the second page? How do you get your conversion code there, or does Affiliate manager tells you?

  14. @Mike – The affiliate manager has nothing to do with it since you now own all the forms. I suggest you re-read this post.

  15. Thats cool article!
    The case study & details make it valuable. Tell me with something. Sudden change you saw & did a great deal of effort on it. Mine isn’t growing. Any special area you want me to focus on.

  16. “in the affiliate industry, you really never know who is telling the truth and who is lying.”

    wow i thought i was being paranoid, but i guess my instincts were right :)

Post a comment