The purpose of Link Control is to make it easy to find the best-performing link for whatever affiliate offer you’re running.  Obviously you can do that manually, but Link Control makes it a no-brainer — which mean a lot more people will do it a lot more often.  Check out the site for a full demo.

It’s currently closed to the public, but there’s a list you can sign up on to get in as soon as they open the doors to more people.  If you do any affiliate marketing, I would get on that list ASAP.

(The title of this post is from the opening of Jeremy’s Link Control presentation at Affiliate Summit West — a very catchy piece by Rob Hustle.  If you follow Shoemoney, you saw it on his blog yesterday.)

I just recently completed another redesign with outstanding results, so I thought it would be a good time not only to talk about how important good design is but to show some real-world stats on what kind of ROI I’m talking about.

Post-Redesign Metrics

The first stat that was important to me was Visits per Registration — that is, how many visits did it take for the site to get a user to register.  You can see from this graph that the ratio sometimes approached 400 visits for 1 registration before the redesign rolled out in late March.  Then it suddenly plummeted below 100 before leveling off around the 110 range — a huge improvement.

Conversions per Week was the other stat that was critical.  A conversion is worth quite a bit for this site, so I wanted to know just how much the redesign was able to move the needle.  You can see that the change was significant and immediate.  The site went from averaging less than 20 per week to more than 40.

Total ROI

I spent well less than $2,000 on 99designs and have so far reaped improvements approaching $100,000.  Nothing else changed other than the design, so that’s a 5,000% ROI — and it gets better every day.

There are few things that will give you a better ROI than giving an ugly site a good facelift.

I can’t tell you how many sites I see wearing a whore’s uniform — selling themselves out for a quick AdSense click, not realizing that they’re sacrificing greater long-term benefit.

Take a good, honest look at your site and ask yourself two questions:

1. What is the visitor most likely to do when they get there?
2. Is that what you want them to do?

Take a look at this page, for example. Hopefully you can see that it’s designed to maximize the number of clicks on the AdSense unit. His model to is to attract free traffic from search engines and then send that traffic elsewhere via AdSense. And that’s a model that works great for him.

It’s probably not the model that will work best for you, though. AdSense is a terrible revenue source. It’s great to get you going, but most long-term models need something much more than the slim return from AdSense. Yet, how many sites do you see every day that are so heavily laden with AdSense that they lose visitors before they’ve really had time to turn them into customers (or subscribers, or whatever the actual goal of your site is)?

Is the design of your site encouraging the behavior that you want to encourage, or are you sending all the wrong signals?

One of the easiest ways to do that is with logos, and there are two categories of logos that help your visitors to trust you.

The first type is seals from companies like VeriSign, TRUSTe, Better Business Bureau and Hacker Safe. These companies certify different aspects of your business to assure your visitors that you can be trusted.

The second type is logos of sites who have mentioned yours in some way. When you include these logos under a header like “Featured On” or “As Seen On,” visitors see these as endorsements of your site — and the more prominent the endorser, the more trust is conveyed.

And in most cases, the simple presence of the logo is all that matters. I’ve seen conversion rates jump 30% with just one good logo, and surprisingly virtually no one follows the link to see what they actually said about you.

Getting several certifications could cost you a few thousand every year, but if you’re making $5,000 a month and can increase your conversion rate by 30% or more, that investment returns itself many times over.

Over the next few weeks, though, I found myself repeatedly coming back to this statement:

I can see the logic behind someone wanting to use a more popular feed image. Showing a larger subscriber count than what your blog has naturally makes your site seem more popular which unfortunately makes new visitors more likely to stick.

That seemed like a given to me. I know the more subscribers a blog has, the more likely I am to both subscribe and to give its writings more weight. So I figured that the same must be true for most other people, too.

I decided to put it to the test, though. If I selected a blog with a feed count in the thousands and showed that count here, would that make a difference in the rate that people subscribed?

So, I started showing a FeedCount chicklet like this one in the Subscribe box in the sidebar (with a count from one of my other sites) and emailed Chris to let him know what I was up to so that someone independent would know that it was all just a test.

Then I waited.

What I found really surprised me. Over the next few days, my traffic was consistent with the days before it, but so was the subscription rate. Advertising a huge subscriber base apparently had absolutely zero effect on how likely new visitors were to subscribe.

Now, granted, the results could vary widely on your site, so take these results with a grain of salt. However, I did think they were interesting. I also think there were probably other positive effects of a high feed count that wouldn’t be apparent.

So should you fake it? Absolutely not. Your credibility would be absolutely destroyed if you ever got caught — and remember, they can usually tell if you’re faking it.

However, I definitely would recommend showing your count if it’s impressive. It lends great credibility and undoubtedly has benefits that aren’t readily apparent. I haven’t quite been around long enough to show mine off yet, though