How Internet Marketers Can Use Split Testing to Greatly Improve Their Results

June 30, 2012 by  Filed under: Advertising 

Split testing is probably the single most important concept in all of marketing. At its core, this is a very, very simple concept. And perhaps that is the very reason more modern marketers don’t take advantage of it. It’s just too easy. It’s also not sexy. It has nothing to do with writing or producing great ads. It has nothing to do with creative genius. It is a purely analytical device.

So what is a split test? It is giving two or more advertisements an equal opportunity, and then tallying or measuring the results of each one to see which one performed the best.

You write two ads: ad A and ad B.

You present both ads in front of the same number of prospects, with each group of prospects being roughly equal, or the same type of prospect. You wouldn’t send one ad to a group of lawyers and then the other ad to a group of auto mechanics. This would represent two completely different groups, with different wants, ambitions, income levels and lifestyles. The results of the split test would be skewed simply because the two groups are so different. You would want to send both ads to either lawyers or mechanics, but not both.

Anyway, once both ads have been sent (or displayed in search results, or whatever), you then measure the performance of each one to see which one generated the most business, or the highest return on investment.

Return on investment: the amount of revenue generated less the amount spent (a.k.a., profit), divided by the amount spent.

The key here is the measuring process. You need some kind of way to measure the performance of your ads. This is one area where the Internet has any other form of traditional media beat hands down. What you have to have is something on your website that creates a conversion. A conversion is simply the completion of whatever action it is that you want your visitors to take. The most basic kind of conversion is selling something. This is also the easiest type of conversion to track and measure.

If you are using AdWords or another pay per click platform, you will be provided a script to be placed on your order confirmation page. This script will report back to the paid search platform whenever a conversion has occurred. Your reporting will then show you that not only has a conversion occurred, but where the conversion came from, such as what keyword triggered your ad in the first place, and what ad was displayed when the person clicked through to your website. With a little bit of simple scripting logic, you can configure the script so that it will actually report back the exact amount of the sale. That is what you want.

For service organizations or any website where a product is not actually being sold, tracking conversions can still be done, it just won’t capture a dollar amount. However, you can still assign an economic value to the conversion if you have some idea of what the average client contact is worth.

For example, let’s say you know that for every ten visitors who contact you through your website, one becomes a client (for a conversion rate of 10%), and that your average client value is $1,000. In that case, you would assign a conversion value of $100 to every conversion (contact form submission) on your website ($1,000 X 10% conversion rate). With this conversion value set, you now have a way to determine your best performing ads on a dollar-for-dollar basis.

Getting back to the topic of split-testing, you will also learn more by using two ads that are very similar except for one thing. In this way, you are testing that one thing, rather than the whole ad. If you use two completely different ads, you really won’t know exactly what it was about the winning ad that caused it to perform better. Maybe it was the headline, maybe it was the copy, maybe it was the graphics… who knows? The earlier you are in the split-testing process, the more different the ads can be. Initially, you probably don’t have any clue what will work (you might THINK you do, but you really don’t). So initially you can test two completely different ads. Once you determine a winner from those two, then you might want to test the winner against a completely different ad, or start testing slight changes to the winning ad, such as a different headline.

You don’t just have to test the ad itself. You can also test the offer. Maybe you have a product but are not sure what the best price is. You could do ads to different markets with different price points and see which one does the best. But here is something to keep in mind when determining price this way: it’s not necessarily about the number of units sold, it’s about your return on investment. Let’s look at an example:

Split test A: sells 500 units @ $50

Split test B: sells 250 units @ $100

Based on the math, both tests did the same, right? They each generated revenue of $25,000. However, if you factor a per unit cost into the equation, it makes a big difference. Let’s say each unit costs you $10. Now here is how the numbers look from an ROI perspective:

Split test A: sells 500 units @ $50, at a cost of $5,000 (500 X $10)

Profit: $20,000, ROI: 400%

Split test B: sells 250 units @ $100, at a cost of $2,500 (250 X $10)

Profit: $22,500, ROI: 900%

So now, as you can see, in this scenario the higher price generates a much higher return on investment, and a higher overall profit, even though it resulted in only half as many sales. The only way you can know this kind of thing – if one price results in a considerably higher profit margin over another, is through split-testing.

Paid search just may be the best tool ever invented for split-testing. It has the MAJOR advantage of speed. In traditional direct marketing using print, it can take weeks or months to create enough data to be able to make a decision. With paid search, more often than not, it’s simply a matter of spending enough money fast enough to generate data fast enough to be able to make quick decisions. You can literally find out in a single day what works and what doesn’t.

What you learn in your paid search campaigns can be used to improve the performance of those campaigns, as well as improving the performance of other media campaigns or your offer in general. As demonstrated above, you can use it to determine the best price for your product. Or you can use it to find the best advertising headline, which can then be used with other forms of advertising. Don’t just think from the mindset that Internet marketing is one thing and print (or radio, or TV, or whatever) is something completely separate. Use what you learn from your online marketing to improve your offline marketing.

Split-testing is a powerful tool that should be used by every business to improve the performance of their marketing. It takes the guesswork out of marketing and lets businesses make better decisions based on analytical data, rather than personal opinion.

Virtuoso Legal Marketing specializes in search engine optimization, paid search marketing and social media management for law firms on a market exclusive basis. Contact Virtuoso at 800-281-2226 or on the Web for a no obligation discussion of how they can help you use your website to generate leads for your firm at a fraction of the cost of traditional media.

Article Source:

Speak Your Mind

Tell us what you're thinking...
and oh, if you want a pic to show with your comment, go get a gravatar!

You must be logged in to post a comment.

Prev Post:
Next Post: