You Are Here: Home > Articles > Testing & Tracking > Article

How To Create A Testing And Tracking System To Explode Your Website's Profits

by Jonathan Mizel

Probably the most important thing you will learn in this article is how to test and track your website's progress. In fact, that may be the most important thing you learn in your life. Direct marketing, both online and offline, is not about opinions, awards, focus groups, or art.

It's about making a profit.

And since I assume you want to improve your site's performance, you have to know where you currently stand in terms of stats, visitor value, hits, and sales to evaluate whether or not a particular technique or offer is increasing (or decreasing) your profits.

In fact, I submit to you that I'm not the greatest marketer in the world, probably no better than you. But what I do very well is test. Then, I can see what works and what doesn't and repeat my successes with frequency.

Note: Making more sales and improving your profitability means testing many different variations of your sales process against one another. It's not always easy, but when you properly develop your site's statistics, you are one step closer to putting your business on auto-pilot.

Over the next year, we'll be talking at length about banner ads, opt-in e-mail, link exchanges, search engines, e-zine advertising, and other traffic boosters. However, none of them will mean a thing if you don't track your sales to see which ones work for you.

The Internet community is fickle, and what works for one website won't work for another. Plus, the market is changing at an accelerated pace, and what made money last year may not even work today. Therefore, it's imperative you create a system to discover what works for your particular business and website.

The term I want you to become familiar with is known as empirical evidence (that which is indisputable). It and nothing else is the only thing that should be directing your marketing.

Don't ever accept anyone's opinion of what will work and what won't. I have taken great salesletters, ones that produced 3 or 4 sales per 100 visitors and 'fixed' them so they produced nothing. If I hadn't been tracking the results, I wouldn't have known why my bank balance was getting smaller.

Tracking will also help you determine weaknesses and flaws in your sales materials (and your process) and will show you exactly where they are so you can fix them or eliminate them.

Example: If your site isn't making any money, it's important to know why. Is it lack of traffic? Or are people getting to the site but not buying from your salesletter? Perhaps there are stumbling blocks preventing them from ordering, like a slow loading page or a defective order form? Maybe it's the headline? Or is it something else entirely?

Another benefit of tracking is people are quite predictable in their actions after you know what those actions are. You will advance your marketing sophistication many levels by simply testing different criteria like subject lines, offers, and prices, and incorporating what you learn into future promotions.

Every marketing technique or element you test, whether it is online or offline, must be tracked using a key, web page, or coding system.

If you market with e-mail, every subject line you run must be tested against other subject lines so you can see which ones generate the greatest number of hits (and which ones make the most amount of money). We drive the traffic into 2 (or more) separate web pages, each with it's own counter.

And always use a valid return address so you can see how many undeliverable addresses you have. If you have 50% of your mail come back, remember to recalculate your stats upward.

Tracking Guidelines

• Every web page you drive traffic to must have a counter or offer some sort of statistical reporting so you can see how many people clicked through to it (check with your provider to see what they offer).

• Separate web pages and autoresponders must be set up for separate campaigns, each with a counter so you know which campaigns are working (generating leads) and which ones aren't.

• Every online sales letter, whether delivered via the web or with an autoresponder, must have a code or operator number so you can determine which marketing pieces are profitable and what percentage of prospects turn into customers.

• If you are taking orders on your website, separate online order forms must be set up for separate campaigns so you can see which ones generate money.

There are no exceptions.

If you don't track your progress, you'll be like a ship trying to get into port before a storm...without a map. You'll never be able to increase your income unless it's by accident, and that's a scary way to run your business.

Here are a few things to keep in mind:

Remember, there's no such thing as an unsuccessful test. You will always learn something when you run a promotion and track your results. Even if your initial sales materials don't work, you will discover brick walls that exist in your offer, your sales process, or deficiencies in your mailing list (like undeliverable addresses) that will keep you from making money. At the very least, you know what not to do again.

Test only one element at a time. If you change your headline and your price in the same test, there's no way to know what affected the response. You may have guessed right with one element, and not another. Thus, you learn nothing at all.

Don't waste your tests. If you are sending unsolicited commercial e-mail (UCE), you may have only a few shots at getting your message out. Don't test ambiguous elements or things that make little difference. By far, there are four major elements that affect response, and they should be tested first and foremost, regardless of how you promote your site:

Your subject line and headline

Your price

Your teaser offer or banner ad headline

Your sales letter

Make sure everything works before you attempt to track results from a test. If your web page is down, or you put the incorrect link in a banner ad you are running, it doesn't matter whether or not you have a counter on the page. It's a virtual certainty you won't get any hits at all. Go through each element of your campaign before you roll it out.

Here's the simple step-by-step process we developed to track all of the different ingredients of our various campaigns.

Establish a key system for all salesletters, whether autoresponder documents or web pages. Make up an operator code (for telephone orders), department number (for mail-in orders) or separate web page (for online web orders). Then, when sales come in keep track of the codes and you'll know which marketing piece generated which dollar.

Set up separate web pages for each testing criteria and put a counter on every one. We'll sometimes create 4 or 5 different web pages for a single promotion. When evaluating a promotion, it's only a matter of counting the hits to see which element pulled the best.  Most of the web hosting services have counters you can use for free, but if yours doesn't, or they only give you one, here are 3 premium analytics services we recommend and have used in the past: Adobe Digital Marketing Suite, HubSpot and SuperStats.

All you do is sign up and they give you (or your webmaster) the HTML code to insert at the bottom of your page. You can even keep your hit count private if you want. Check your stats using your customer login to the services.

If we are testing banner ads or link exchanges, we also set up separate web pages (with counters) for each promotion. We'll sometimes create dozens of pages, all identical in content but with a different address. We give each one of those separate URLs to people who want to trade a link or cross promote our site. Then we can see how successful their efforts are.

Remember, while tracking is the easiest method to improve profitability, it's the most overlooked and ignored process when trying to determine why something does, or does not work.

Determining Your Visitor Value

After you have tested a few headlines, offers, and other attributes, you'll probably end up with a page or site that is somewhat profitable. This now becomes your control piece.

A control piece is a salesletter or web page that has a predictable outcome. In other words, you have a reasonable expectation that when a people visit the site, a specific percentage will buy.

For example, the website that sells this newsletter generates one sale per 100 visitors. I'm not saying this is good or bad, it's simply a fact.

And it's also a fact that if we sell our newsletter for $100, each person who visits is worth a buck.

Here is the math: 100 visitors equals one $100 sale. $100 divided by 100 visitors is one dollar per visitor.

Now you need to do it for yourself. After you develop your control piece, you want to find out how many people purchase and express that figure as a percentage. Multiply that figure by the amount of profit you make on your product or service and that's your own "value per visitor".

Why is this information so important? Simple:

The days of PROFITABLE free traffic on the Internet are over.

Search engines are starting to manipulate their own indexes based on who is giving them advertising dollars, plus they are developing their own content that competes with your listings.

Unsolicited commercial e-mail, while highly profitable if done correctly, is getting more and more difficult. Many businesses don't want the hassle or expense of keeping multiple accounts open and others want to avoid being labeled as "spammers" (especially businesses in small or closely knit industries).

That means we are all going to have to start paying for traffic, and in order to do so at a profit, we need to know what our visitor worth is. Here's an example:

Let's say I decide to run some banner ads with Link Exchange Express (LEE) and I develop a banner ad that gets 2% of the viewers who see it to click through to my page (a separate page set up specifically for the LEE Promotion).

I'll have a cost associated with each visitor because the banner ad isn't free. To determine that cost, I'll divide the price of the ad into the number of hits I receive. Then, I'll compare my cost per visitor with my value per visitor.

If I'm paying 20¢ for visitors that are worth a dollar, I'm getting a 500% return on my advertising dollar.

Note: With that information at hand, if I decide I want to make $10,000 a month on the newsletter promotion, knowing my visitors are worth a dollar but cost only 20¢, all I need to do is spend $2,000 a month (or 1/5 what I want to make) on similar banner ads.

Of course, if my visitors are costing me more than a dollar, that means I'm losing money, and I need to reduce what I pay per visitor or develop a more powerful control piece.

You should always be working to improve 2 areas of your website marketing:

• You should constantly be trying to develop new offers to beat your control piece. To increase it conversion rate and it's sales appeal.

• You should constantly be trying to reduce the costs associated with generating traffic.

When you follow the above guidelines, you'll be in the top 1% of all marketers in the country. Those who know their numbers, improve their stats, and generate profits with their businesses and their websites!

Sometimes testing can be hard on your ego. After all, you may have felt sure a particular headline, guarantee, price, or other element of your testing was right on when in fact, it bombed. Don't worry! As I said earlier, the best marketers are actually the best testers and the only scorecard is what's left in your bank account after the expenses have been paid. Never quit.

Jonathan Mizel is a well-known and respected Internet Marketing expert, and is often found advising well known companies such as Microsoft, Intel, and American Express.

Share this article: