You are here: Home > A/B Split Testing
Archive | A/B Split Testing RSS feed for this section

Internet Marketing Lesson: A Sheep, A Duck, And A Rooster

by on June 16, 2010 in A/B Split Testing, Copywriting, Email Marketing, Internet Marketing

A little story for you, which may not seem related to Internet marketing right away. I promise to bring it all home for you in the end.

If I were to say “Montgolfier”, most people would answer that that name belonged to the two brothers who flew the first ever hot air balloon…  and they would be close but not quite correct.

See, Montgolfier did invent the balloon and were the first to build one. However, they weren’t the first passengers on board their invention.

Since up to that point, human beings had never had a chance to become airborne, the effect of the air flight on a living organism was unknown. The Montgolfier were offered criminals as their guinea pigs. They declined. Instead, they picked a sheep, a duck, and a rooster.

Obviously, that story has a happy end. Nowadays, thousands of people get up in the air in airplanes, helicopters, gliders and, of course, hot air balloons. They don’t think twice about it and obviously don’t remember the three little creatures that paved the way to human flight.

What’s the moral of the story, you may ask? And what does it have to do with Internet marketing?

Well, imagine you are creating a new product. You’re excited. You’re so sure it’s going to be a smashing success. The website it up, the shopping cart is ready to take orders. You go ahead and spend a small fortune on the traffic and, to top it off, line up a bunch of JV partners.

The big day arrives. The traffic comes in fast and furious and…

… nothing! Your squeeze page didn’t squeeze and your sales copy didn’t sell. At least not to the extent that would be considered “reasonable”.

The money spent on advertising is gone forever. And so are the partners who vouch to never participate in your lame launches again.

That’s one possible scenario.

Now imagine doing things a bit differently.

You create a product and put together two different versions of the same squeeze page. You set up an A/B Split test and send the traffic to these squeeze pages through a link rotator, which you then promote to a relatively small in-house list (the online equivalent of the Montgolfier’s sheep, duck, and rooster.)

You generate a bunch of clicks, just enough to tell which squeeze page works better (and if it works at all). You pick the better version (“control”) and send it out to all the lists you can get your hands on.

Your affiliate partners come in and start promoting your product too. This time, the conversions are high. Everybody is making money on the upsells and everyone’s happy.

What was the difference?

It’s easy to see: Instead of trying to predict how your squeeze page would weather out in the marketplace, you did a limited test before opening the floodgates to major traffic. A small extra step in planning and testing your campaign can mean the difference between making it and not making it in Internet marketing.

In a few days, one of my JV partners will be sending out my offer to his super-big list. So guess what I’m working on now? Exactly: I’m setting up an A/B Split test for my pages that I will promote to one of my own list. And when the big traffic hits my squeeze page, I will put my best foot forward because I will know which one is, in fact, the best.

Technorati Tags: , , , ,

Comments { 6 }

The Easiest, Fastest, Most Overlooked Way To Double Your Pay As an Internet Marketer

by on June 14, 2010 in A/B Split Testing, Copywriting, Internet Marketing

This short article may read like a math lesson. However, its purpose is to simply show you the revenue mechanics of your Internet marketing business and help you get that “pay raise” you’ve been working so hard on for so long.

Every successful Internet marketing business has a way of collecting people’s contact details and adding them to a database. That is usually accomplished with a squeeze page.

I think it is safe to say that having a good squeeze page is critical to “making it” in Internet marketing.

Yet a lot of people don’t test their squeeze pages. They don’t tweak them to squeeze more juice out of the traffic they are getting. They don’t look at their conversion rates.

If you happen to be in that number, I hope to give you enough evidence to make testing your squeeze page a priority.

So let’s begin.

Let’s imagine that your squeeze page is converting at a very respectable 35%. You decide to conduct an experiment and test several headlines. After a series of A/B Split tests you end up with a headline that improves the conversion rate for your squeeze page by 2%.

Next, you change the graphics and the location of your opt in box that results in another improvement of 5%.

You don’t stop there and continue testing. Adding the privacy statement results in another 3% bump.

Excited, you add a testimonial. The one from a male client doesn’t seem to help that much. However, the female version results in another 3% improvement.

These numbers may seem small and insignificant. But let’s add them up.

2% + 5% + 3% + 3% = 13%

(Actually, adding the percentages here is not mathematically correct. I will spare you the calculations but if you were to do the math right you’d end up with 13.62% improvement. Think “compound interest”.)

A 13% improvement may not seem like much. However, let’s look at what’s going on. You page is now converting at (35% + 13%) = 48%. So basically if in the past your squeeze was  converting approximately one out of three clicks into a subscriber, now it converts one out of two.

This additional 13% is in reality a (13 / 35) = 37.14% improvement to your conversions.

Let’s see how this improvement may help your bank account.

With the new squeeze page, every dollar you invest in the advertising of your business goes a whole 37 cents further. You also get 37% more leads from every joint venture. Now the campaigns that were dogs are starting to make you money, while the rest have gone from mildly profitable to wildly lucrative.

Bear with me for a little bit. We are not quite done yet.

Every business operates at a certain profit margin. Traditional “brick-and-mortar” companies may have a profit margin of 3-15%. An Internet business can, in fact, enjoy a much higher one.

Let’s say your profit margin was at 30% before all the testing and the tweaking you’ve done. Now you have just expanded the number of leads, and hence the amount of sales, by 37%.

And the best part, it didn’t cost you anything at all, other that some time, to do that! It is all pure profit.

There is a moral to this story. Here it is:

Every time you don’t find the time to properly test your squeeze page, remember that by taking the shortcut you are accepting a cut in your pay check. And while there is always something to pin your lackluster sales to — bad economy, changing search engine algorithms, too many product launches going on at the same time, or your customers “not spending” — deep down inside, you know that it only takes a quick glance in the mirror to know the real reason.

Technorati Tags: , , , , , , ,

Comments { 3 }

A 100-Year Old Trick That Makes You More Money Online

by on June 13, 2010 in A/B Split Testing, Copywriting, Internet Marketing

Even though Internet marketing has only been around for a very short time (at least on the historical scale), the principles that online money-making is based on have been formulated, tested, and perfected over 100 years ago by the direct response marketing industry.

One of the simplest, most powerful, and yet probably one of the most overlooked “tricks” that comes from that time is A/B Split testing. (And it is overlooked, of course, mainly by the rookies. Every successful Internet marketer I know uses it with an almost religious rigor.)

Let’s take a look at what A/B Split testing is all about.

Back in the day, an advertiser would run the same ad in two cities at the same time. The ad in New York would use Headline A and the ad in Boston would use Headline B. The sales numbers would come in and the results would be tracked.

A week later, they would run the same ad again. However this time, the headlines would be reversed: Headline A for Boston and Headline B for New York.

When this experiment is over, the advertiser would know with scientific precision, whiccdh headline generated higher sales numbers. The winning ad would be declared a “control” and  would be subsequently used to attack the national market and build another business empire.

As Internet marketers, we are blessed to have a great variety of tools available to us to help us measure pretty much anything and everything our heart desires. Computers make it easy and inexpensive to do.

Yet so many online business owners still neglect to test much of anything, either due to ignorance, arrogance, or just the lack of proper time management.

Let’s see what can — and should –  be tested in your online business.

First is your “ad”, something that gets you a click. That can be a PPC ad, a banner, or the email copy that your JV partners use to send you referrals.

The next thing would usually be your squeeze page. (Shame on you if you don’t have one.) Just a few tweaks here and there can significantly improve your conversions and result in an almost instant pay raise for you.

As we continue, we need to test the sales page, the offer, all the upsells, and downsells, and, of course, the email follow up sequence.

Here are some of the less obvious ideas to test: How about adding a physical product option (CDs, DVDs, product binders shipped to your customer’s door) and adding direct mail (postcard, greeting cards, and “3d” mail items that grab attention) to your follow up? Many online marketers never think about moving their business partially into the offline, so if you manage to create a campaign that works, you have a potential of create an enormous competitive advantage.

Technorati Tags: , , , ,

Comments { 1 }