by Comm, Joel
That might not sound like much—it means that 99 out of every 100 readers aren’t buying—but a product such as a GPS system costs about $135. On Amazon, that would mean an affiliate commission of around $5.40 for every sale. To earn $540 extra each month, you’d need make 100 sales during the month. At a modest 1 percent conversion rate, you’d need 10,000 unique users every month.
When you’re generating that kind of traffic—and a site that’s been up for a while should be generating that kind of traffic—that’s the sort of boost to your income you can expect from just one short post recommending an affiliate product.
You can characterize a post like this as an “experience” recommendation. It’s saying, “I’ve used this product. I like it. You’re going to like it, too.” You won’t always be able to say that about every product you’re going to offer because you won’t be able to try every product. (Although, if you are managing to earn $500 a month through occasional affiliate recommendations, then spending $130 or so on a gadget that you can recommend is still going to look like a good deal. You’ll be making a profit of $370 and have a cool new toy every month.)
An alternative approach is to recommend the producer.
This is something I do frequently. I know that if Yanik Silver or Shawn Collins or any of the other marketers I’ve met at conferences is bringing out a new product, it’s going to be good. I’ve seen them in action; I know their work; and I know they don’t produce garbage.
I can’t say what a product has done for me if I haven’t used it. But I can say what the product does; I can tell my readers what the person I’m recommending has done; and I can remind them that he hasn’t brought out a dud product yet.
Again, this comes down to trust.
For “experience” posts, sales are going to depend on the depth of the connection my audience feels with me. If readers identify with me, like me, and trust me, they’ll believe that they’ll like the same products that I do.
For these “publisher” posts, readers will buy the affiliate product because they trust my judgment. If I tell them that some publisher has good information and they should listen to that publisher, then many of them will trust me and do it.
Of course, that trust is only going to last as long as it proves to be true. If it turns out that the product is poor or that I’ve misplaced my own trust in the person I’m recommending, then I can expect my conversion rates to fall through the floor. As that trust dries up, so will my income.
So while these kinds of recommendations are very easy to make, they do carry an element of risk. Choose them carefully and don’t promote a product simply to help a friend. A real friend will understand why you have to refuse if you’re not 100 percent certain that the product is going to deliver.
A third kind of recommendation is the type that just lets people know about a product and tells them that it looks cool. These are probably the most popular, if only because they’re the easiest.
If you see a book, a gadget, or a tool that you’d like to buy, you can just write a post telling people about it. Include an affiliate link, and some people will click through and pay.
Clearly, the conversion rate will be lower than it would be for a product you can recommend personally or one that has been created by someone you can recommend personally, but you should be able to generate a few sales and give yourself a little extra income.
The number of sales you generate though will also be affected by its placement, our next topic of discussion.
PLACEMENT
I first saw the potential of the Internet when I realized that moving my ad units from one place on the page to another sent my click-through rate through the roof. Content was important, of course, but it was how I implemented the ads that had the most immediate effect on my returns.
It’s no surprise then that placement and implementation have a massive influence on affiliate earnings, too.
When you’re working with ad units, there’s a lot to experiment with. You can play with the color schemes and fonts. You can try different keywording to see which terms bring up the best ads. And you can use various sizes to see which are most likely to attract readers’ eyes. It’s a process, and although there are now guidelines that make the testing stage shorter than it used to be, you can still expect to spend a few weeks trying out different strategies to see what works best on your site.
For affiliate ads, it’s all much simpler.
Text links outperform banner ads, and links worked into the content outperform links placed in navigation sections.
The difference is remarkable, and it really does prove that simplicity is best. At least one blogger has found that simply swapping his affiliate banner ad for a text link increased his click-throughs by 60 percent. Others have seen even more dramatic improvements with equally simple changes.
Big merchants like Amazon offer a giant range of different types of ad units (including contextual versions that try to read the content on your page and offer products to match, such as AdSense). They’re all very neat and snazzy, but rarely do they perform as well as simply linking a key phrase in the content to the product and including your affiliate code (Figure 5.4).
Figure 5.4 Amazon offers a giant range of product links, banners, and widgets to attract users and win sales. But recommended text links will always do best.
The reasons are pretty clear.
First, a text link looks natural. A banner ad looks like an ad, so although it’s more likely to be seen than subtle hyperlinked text, it’s less likely to be clicked. Users feel that they can happily ignore ads, but when the publisher has taken the effort to link directly to a web site, that link appears to be recommended. It looks like additional information to the content on the page, something that’s worth reading.
The text around the link also leads users to click. It puts the product in context and raises curiosity. Having read about the product, the user is likely to want to click through and see the product itself.
Text links also make for cleaner Web pages. Instead of a page filled with flashing graphics that serve mostly to drive users away, the user will be able to see a page designed primarily to serve content. The ad will be embedded inside that content.
That doesn’t mean that affiliate graphic ads never pay. In fact, while you can expect your text links to dramatically outperform affiliate graphic ads, the best strategy is often to combine different methods to increase the chances of winning click-throughs and sales. In the sample post I provided earlier for example, I would expect most of the affiliate sales to come from the text link containing my affiliate code. However, I would also use an affiliate image of the product to illustrate the post.
I’m going to want a graphic element there anyway to attract eyes, so using an affiliate ad from Amazon or even linking an image with my affiliate code provides one more way for users to learn more about the product—and gives me another chance to generate a KaChing.
It’s also possible to place affiliate ad units on the page. Just don’t expect them to generate more clicks than a conventional CPC ad unit while earning money only for those that convert into sales. In general, that’s more likely to happen on sites such as AppCraver, ones that discuss products.
The position on the page isn’t as important as it might seem, either. When you’re using Kontera, you’ll want to put some effort into ensuring that the ads aren’t sent down to the bottom of the page where no one will see them. But Kontera gives you no direct control over which terms are linked. When you’re inserting your own affiliate links, you get to choose where to put the links and more important, you get to write the text that surrounds those links.
Affiliate advertising gives you complete control.
Clearly, placing the affiliate link on the last word on the page is always going to be a bad idea, but it’s unlikely to matter whether it appears in the first, second, or fifth paragraph—provided the preceding text has generated enough interest and enough enthusiasm to persuade reader
s to click through.
SEND AFFILIATE E-MAILS
Spam has caused a huge amount of damage. According to Ferris Research, the global cost in lost productivity caused by unsolicited, unwanted e-mails was an incredible $130 billion in 2009. All that time spent deleting e-mails and flicking through folders looking for lost messages adds up.
It’s no surprise then that those offers of herbal pills and fake watches have had an effect on legitimate direct mail marketing. Comparing figures produced each year by the Direct Marketing Association (DMA) reveals that between 2006 and 2009, as the volume of spam shooting around the Internet continued to set new records, the return on investment for e-mail promotions fell by about 25 percent. That sounds like a marketing strategy in trouble... until you look at the figures a little more closely.
In 2006, according to the DMA, every dollar spent creating and sending an e-mail with a promotional offer generated $57.25 in return. By 2009, that ROI had fallen to $43.62. But that’s still twice the amount earned from search-based advertising. (By way of comparison, the ROI of traditional, mailed catalogs was estimated at just $7.32.)
The reason spammers continue to send out billions of e-mails every year is that even their irritating, scattergun approach delivers profits. E-mail advertising works, and it works better than any other form of marketing on the Web.
Spam, though, has frightened people off. According to the DMA, marketers spent $600 million on e-mail advertising in 2009. That’s just a fraction of the $11.2 billion spent on search advertising, even though it delivered only half the KaChing for the buck. That means that marketers who know how to practice e-mail marketing, who understand how to do it in a way that delivers information about quality products to people who have chosen to receive that information, can make the most of a massive opportunity.
Proper e-mail marketing isn’t difficult to do. Even the law provides a bit of help.
The CAN-SPAM Act of 2003 threatens fines of $11,000 for each unsolicited e-mail delivered to each recipient. Considering the number of recipients in each spam campaign, that should be a hefty threat. Officials have occasionally handed out the odd giant fine, collecting piles of cash and confiscating Porsches from unlucky spam artists. But most spammers feel—for good reason—that they’re never going to be caught, so they keep sending out the messages. The act might not be very helpful at keeping out spammers, but it does provide useful guidelines that everyone marketing through e-mail should understand and follow. You can think of them as rules for good e-mail business, and they cover three areas:1. Unsubscribe compliance. Every marketing e-mail you send has to have a way for recipients to pull themselves off your list, and those requests have to be honored within 10 days. Usually that just means an e-mail link at the bottom of your message marked “unsubscribe.”
2. Content compliance. “From” lines and subject lines have to be relevant, and you have to include your physical address, not just your e-mail address, in the message.
3. Sending behavior compliance. This requirement is designed to clamp down on the most deceptive behavior. It prohibits sending e-mail through open relays to harvested e-mail addresses and using a false header. Those are the kinds of things that only a real spammer would do and they don’t affect legitimate e-mail marketers.
These requirements are the minimum, and there are no restrictions against e-mailing existing customers or people who have inquired about your products. CAN-SPAM calls these “relationship” messages, and they’re perfectly legitimate.
In practice, legitimate e-mail marketers go further than CAN-SPAM’s rules. They use “double opt-in” lists. Subscribers have to click a button declaring that they agree to receive marketing messages from you. They then receive a confirmation e-mail that they have to click again before their e-mail address is added to the list.
That makes it impossible for someone to add someone else’s e-mail address to a list maliciously, and it also makes it impossible for anyone to say that they shouldn’t be on the list. If that sounds like you’re putting a pile of hurdles in front of people you’d like to be marketing to, understand that it’s a practice that benefits marketers as much as recipients. It ensures that the e-mail addresses on your list are functional and accurate, and it reduces the bounce rate when you send out your messages. You don’t need to be a programming genius to use this system, either. Mass mailing services often provide free scripts to their customers. SendBlaster (www.sendblaster.com), for example, supplies a simple PHP script that you can upload to your server and explains in five steps how to use it (Figure 5.5). It’s very simple, and most decent mass-mailing systems supply a toolbox for their clients to use. They want to make your e-mail marketing easy so that you’ll become a subscriber. It’s a good deal.
Of course, you still have to build your list, but part of your Internet business strategy should consist of picking up e-mail addresses whenever possible. You’re going to be handing out lots of free samples, reports, and white papers to bring people to your web site and spread your name around the Web. The price for those should always be an e-mail address that allows you to send them information about products they might want to buy in the future.
Figure 5.5 SendBlaster helps you build a double opt-in system in seconds.
Thousands of people are going to flow through your web site every month. You want to make sure that they’re leaving with a hook that enables you to pull them back at a later date.
When you’re signing them up, you’ll also have an opportunity to push other products in their direction. If someone has shown enough interest in the information you’re providing to give you their e-mail address, then some of them will be interested in other free information, too. People who sign up for the free copy of my AdSense Secrets e-book, for example, have to leave their e-mail address. They’re then taken to a “thank you” page that advises them to check their spam filters for their confirmation message. The bottom half of the page is an advertisement for a free trial of my Top One Report newsletter.
Enough people sign up for that report to more than cover the cost of the free books that I’m handing out.
In effect, I’m using an information product to grow my e-mail list, and at the same time, I’m signing up paying subscribers to a publication. All it’s costing me is free copies of my e-book.
It really doesn’t take a great deal of effort to produce a free information product, place it online, and capture e-mail addresses of people who want to enjoy it.
As you build your list, you’ll want to make sure that it’s managed well. Or rather, you’ll want to make sure that they’re managed well, because the more products you make available, the more lists you’ll create. Those lists are likely to be for products about similar subjects—someone who downloads a report on low-cost home repairs, for example, may also download an e-book on home carpentry—so you can send the same marketing information about similar products to different lists.
When you have that kind of overlap, it’s a good idea to make sure that you’re not sending the message to the same recipient more than once. It’s easy to do, and it’s the kind of thing that quickly pares down your lists, as people make use of that unsubscribe link.
So what should you be sending to the people on your list?
Affiliate marketing by e-mail is different from affiliate marketing online. Your visitors see your Web page only after choosing to visit your site. An e-mail is presented to someone who has not come to you; you’ve come to them with a product you think they’re going to like.
It’s the difference between store assistants who ask shoppers how they can help and people who hand out flyers on the street. Shoppers are willing to listen; passersby—even those looking in your shop window—will give you only a second of their time. You’ll have to work fast to keep them interested and lead them into the store.
That means e-mail messages that contain affiliate links tend to be more “sales-y” than Web content. The risk, of course, is that a subscriber who receives a
sales message will glance at it and make a beeline for the unsubscribe link.
That’s fine. If subscribers don’t want to be bothered by sales messages, if they don’t want to know about the products that I’ve chosen to offer them, then there’s no reason for me to continue sending them e-mails. They aren’t going to buy, so I don’t want to annoy them. I’d rather send messages to 100 people who are interested in buying than to 1,000 who have no interest at all.
Every time I send an e-mail to my subscriber list then, I expect to receive a number of unsubscribe requests. Those numbers tend to be lower than you might expect—that’s the benefit of double opt-ins and offers of good products—and I keep track of the numbers. If an offer results in a higher than average number of unsubscribe requests, I’ll make a note not to offer that kind of product again. Not only are my subscribers not interested in buying it, they don’t want to see it, either. The response I receive tells me something about my subscribers and the information they’re looking for.