Trash Can

The Great Affiliate Experiment

Okay, maybe the title of this post is a bit grandiose, but whatever. Here we are.

The Backstory

I participate in a number of affiliate programs and am quite opinionated about how affiliate links and banners are used.

Almost three months ago, I made a decision to remove affiliate banners from my site – not because they were bad, but because I wondered if they hurt my business credibility. For instance, would a potential lead come to my site and think about hiring me, but then see a bunch of affiliate banners and think “Wow, she’s clearly not making enough from client work if she needs to have advertisements all over the place. She must not be very good at what she does.”

I don’t know if that situation ever happened, but I thought it would be an interesting experiment to ditch the banners for a while and rely solely on affiliate links that I sprinkle into my content (when the link is relevant to the conversation at hand, of course).

The Hypothesis

I proposed that removing affiliate banners would not negatively impact income.

Read on, Jared, below are the results.

First, My Trends

Before I can make any meaningful interpretation of the last three months of data, I need to set it in the context of several things:

  • Comparison relative to site traffic same period last year and previous quarter (presumably the more traffic, the more clicks)
  • Comparison to income same period last year (January – March 2014)
  • Comparison relative to income previous quarter (October – December 2014)

Site Traffic

Get ready – I’m pulling back the curtain and showing you my traffic data. This is like letting you sit on my couch.

Same Period Last Year

Interestingly, my traffic this first quarter of the year isn’t much higher than my traffic for the same period last year – it’s roughly 9% higher. That’s a little depressing to note, but then, I’ve also published very little content so far this year (40% less than this period last year). Note to self: Write more.

Traffic for Same Period Last Year
Blue line represents current data. Orange is historic.

Previous Quarter

Comparing the last three months of traffic with the previous three months, there’s a decent bump (almost 14%). That would excite me, except that the previous quarter was an overall flat quarter traffic-wise; Again, probably due to a decrease in writing.

Previous Quarter Traffic
Blue line is present quarter, orange line is previous.


In case you haven’t caught it yet, there’s a direct correlation between the amount of content you produce and the amount of traffic coming in.

Want to get more people to your site? Write more. Share on X

Affiliate Income

If sharing some traffic stats was like letting you sit on my couch, then sharing financial data is like letting you peek in my underwear drawer.

But I’m doing it anyway. That’s right – you won’t find Big Girl Panties in my underwear drawer, because I’m presently wearing them while writing this post.

Same Period Last Year

Now this is a fun one. Despite a modest increase in site traffic over the same period last year, I saw a 30% increase in affiliate income – $6,507 in Q1 2015 vs  $4,814 in Q1 2014.

Income ChartOne contributing factor is that a couple of my affiliate partners increased their payouts, but that certainly doesn’t account for the full increase. Somehow I’ve converted extra clicks that weren’t necessarily due to traffic. I really don’t have a good explanation for that, though I’m sure a deeper dig into site analytics could reveal something.

Relative to Previous Quarter

Good grief. If last quarter’s traffic and affiliate income were lower than usual, the pendulum has swung in the other direction for this quarter – $6,507 in Q1 2015 vs $3,864 in Q4 2014.

Previous Quarter IncomeKeep in mind this huge increase is AFTER I dropped affiliate banners. Also, the traffic increase for this quarter seems incredibly disproportionate to the income spike. But I’m not complaining.

The Verdict

If you’ve read this far in the post, then you already suspect the verdict: Removing affiliate banners from my site has not decreased income.

Removing affiliate banners from my site did not decrease income. Share on X

Now, arguably I could have made more if I was using banners in addition to links. The world will never know.

What I do know is that my site feels cleaner and less cluttered without the banners. Since this experiment showed no decrease in income as a consequence of ditching the banners, I’m going to leave them off.

One Last Thing

Stepping outside of this experiment, I did want to pontificate just a moment longer on the topic of affiliate marketing.

People read posts like this or listen to Pat Flynn’s podcast and think that affiliate marketing is some magical income source, where a fairy leaves deposits under the pillow every night. Now, I’m not talking about you here, because I know that you know better, but there are people out there that think that.

[The proverbial] you cannot buy a domain, throw up a website, stick a bunch of banners on like bumper stickers and expect the masses to come (and click your affiliate links). It doesn’t work that way, at least not for the majority of people.

It’s work. Writing content, earning the trust of an audience, building relationships — these are things that take time and effort. And you know what? I don’t do any of those things specifically for affiliate income. I do those things for my brand. Any affiliate income that occurs as a result is just gravy.

15 thoughts on “The Great Affiliate Experiment”

  1. Really interesting! Thanks for the candid view. And congrats on the successful experiment. I agree your site looks even more professional now.

    Clearly, your traffic has improved in quality more than quantity. At least that is true if you consider revenue as a definition of “quality” and who wouldn’t? Can you see that increase in quality if you dig into your Analytics? Are they staying on pages longer? Visiting more pages? Any stats that correlate to increase in revenue?

  2. Hey, Carrie

    Great post! Reveals answers to questions I have wondered about for moving forward with my blog.

    The conclusions you draw are spot on! Marketing 101 applies to blogs, just as it still does to print collateral. A visually messy, heavily banner-ed site MAY affect your particular target audience in an adverse way. So testing is a good thing, also a Marketing 101 principle.

    Your references to “affiliate marketing” take in one great way to monetize one’s blog property, no doubt.

    People should also know that, a second, and perhaps more “pure,” form of affiliate marketing is this: 1) create, launch and promote your own product /service in a particular niche, 2) build an email list from that launch promotion, 3) network with other product creators(now your peers) in your niche who also promote via email, 4) routinely promote these peer products/services and…. rinse and repeat.

  3. Congrats Carrie on building such a reputable site! I don’t doubt that tons of people click on your affiliate links because you are consistently putting out awesome content. 🙂

  4. Glad you made mention of how much work it is to make money by being an affiliate. So many of our clients hire us to do their site and have us put in banners or links to a ton of different products that they are affiliates for thinking that they are going to get this huge passive income each month.

    It doesn’t happen for them and they never quite understand why no matter how many times we tell them that they need to provide relevant content about whatever product they are an affiliate of in order to create viewer clicks, as well eyeballs to the site in the first place.

  5. I love posts with this kind of transparency, thank you Carrie!

    One question though, did you try to use link attribution in Google Analytics to see which of the links were actually being clicked?

    I suppose it doesn’t really matter since you make more money and don’t have your site clogged up with banners, you obviously made the right choice.

    1. Thanks, Kyle.

      I used AdRotate for awhile, which gave me click stats on banners. It was an interesting way to A/B test various banners, but as an overall percentage of clicks, banners were lower than text clicks (I don’t know the exact percentage).

    1. Amajjika, you bring up a really interesting point. I *think* only one of my affiliate programs pays out on annual renewals (most are a one and done type thing).

      1. Well I was more thinking about purchase cycle lead times. Often people like to research a matter, don’t have the budget but may have kept the product of “carrie dils” in their mind until they were ready to purchase. So perhaps the fruits of your content labour in the previous year (having conversations with your audience and building authority) are being realised now. I know that in web accessibility for instance I’ve submitted proposals 2 years prior to the work starting!
        Anyway, congrats again and it’s been fun going through your underwear drawer ;-P hehe

  6. Phenomenal post. Thanks so much for sharing these insights and pulling back the curtain so bravely. The challenge with comparing month to month is that traffic (both quality and quantity) can change so drastically with seasons and content you’re marketing. I agree that based on the data you presented though that the removal of the banner doesn’t appear to be loosing you any revenue on the affiliate side of things. One other way to tell would be to use an adserver (like Revive) and track clicks and use custom affiliate tracking tags (if the affiliate supports them) to do more itemized tracking but that’s a bit more work for sure.

Leave a Comment

Your email address will not be published. Required fields are marked *

Carrie Dils uses Accessibility Checker to monitor our website's accessibility.