Traffic. Traffic. Traffic.
The first thing that we hear about when we’re establishing a business these days is getting traffic online. And yes, traffic is very important. After your website starts getting the traffic that you want, what happens next?
TNMC members now have all the traffic they can ever want or need using our TNMC app, however the way you now treat your new prospective customers and customers is a big determination to how much income you will be able to make!
The fact is you can have tons of traffic coming to your website and still not make a sale, get one opt-in or create any leads. When traffic comes to your website, you want those visitors to do something. That “something,” that action is called your most wanted response. Once you really know what you want them to do then you have a goal. Every time they do the desired action, you win. Every time they don’t, you lose. This is critical if you are using PPC (pay per click) to drive traffic to your website. Every click costs you money and if that click does not result in the traffic taking the action you want then you are losing money and wasting time.
From here on out, the act of getting that visitor to perform that desired action will be referred to as “conversion.”
Again, conversion is another name for the visitor coming to your website and taking the action that you want them to take. Nothing more, nothing less. Conversion is a yes or no metric. Either they do what you want them to do or not.
They give you their name and email address – Yes or No
They purchase your product – Yes or No
The fill out a form and submit it – Yes or No
You see, with conversion either they take that action that you want or they don’t. Your job as the website owner is to make the road to taking the desired action smooth and easy. A lot of factors play into this:
Readability – How easy is it to read and understand your webpage?
Color – Are you using the right colors?
Call to Action – Are your instructions clear in regards to what you want the person to do?
Source of the Traffic – Where did the traffic that is seeing your page come from?
All these are easy to change and therefore easy to test.
Tracking and Testing
For some reason, people often have a hard time with testing their websites and landing pages. Testing really is silently asking your visitors a few simple questions.
Question 1: Do you like this? Or do you like that?
Question 2: Do you like it better if I say it this way? Or do you like it better if I say it that way?
Question 3: Do you like the way this looks? Or do you like it better if I present my page to you this way?
That’s it. Testing is really about asking questions to your visitor and letting them make a decision in the moment. This is what I call a forced survey. Why a “forced survey?”
When you do a standard survey, people know that they are being asked a question. When folks know they are being asked a question for the sake of a survey, many times they change their answers to accommodate the person asking the question. In the case of a “forced survey” you just present the situation to the visitor and let them respond. Either they like it, or they don’t. Either they move forward toward your most wanted response and take the desired action, or they don’t. It is that simple. Yes or no. Either they like what they see or what you are saying, or they don’t.
A forced survey like this is commonly called split testing. That is when you show some of your visitors one version of your webpage or landing page and you show some of your visitors another version of your landing page or website.
Split testing can be as simple as changing the:
Color of the Background
Use of Video
Use of Audio
Anything that is on the page that the visitor sees is something that can be split tested.
So why do you want to split test your pages? Good question…
Just like when you were in school, testing makes you better. Testing allows you to get to know what your visitors like and don’t like. Split testing allows you to better know:
How to present your product or service
What pictures they like
What words they like to hear (or read) when being presented your product or service
What visuals they like to see as you present your product or service
What type of benefits they like to hear
What features they what to know about in relation to your product or service
Knowing these answers will let you fine-tune your site, and you can sell more product or services to the people who are already coming to your website.
Testing can get ultra complicated if you let it. Really, though, it is just getting to know your visitors better and finding out how to best communicate your message to them.
There are plenty of tools that allow you to split test webpages.
One that is free and simple to use is Google Web Optimizer – http://www.google.com/intl/en/websiteoptimizer/tutorials.html.
This video series shows you how to set up a test in about 5 minutes.
How do you know when the test is complete? This is handled by something called Statistical Confidence and Validity. This long word set – Statistical Confidence and Validity – means I did this test enough times to know that it was not a change occurrence and I am pretty sure that what happened during my test will continue into the future as long as I don’t change the traffic source. Statistical Confidence and Validity relies on the number of times the page was seen and the desired action was taken in relation to how many times the visitors saw the other page and took the desired action when they saw that page. It is as simple as that. There is some interesting math that goes on in the background. But we don’t have to concern ourselves with that here.
Another way to check the validity of your results is to try this tool at the website
You’ll see that the tool asks for the “Number of Clicks.” This is where you enter the number of sales or leads that you have gotten from your landing page. Then, enter the number of people who have seen your landing page.
Do this for version one and for version two.
The math will be done in the background and the program will tell you how confident you can be of your future results for the test that you are running.
Go ahead and play with the tool:
Enter 100 as the number of impressions (how many times the page was seen) for both Ad1 and Ad2.
Enter 4 for Ad1 number of clicks. Enter 7 for Ad2.
You should see a message that reads: Confidence in the long term outcome of your ads can not yet be established. You should let your test run some more to gather more data.
Now enter 12 for Ad2. You should see a message that reads: Ad 2 has a higher CTR than Ad 1. You can be 95% confident that this result is real, and not due to randomness.
Notice that Statistical Confidence and Validity is based on the difference in the number of actions taken by the visitors. The larger the spread in the number of actions, the more confidence you can put in the winning ad continuing to win.
So how do you get big differences in response to an ad? The simple is answer is to test big variations for version 1 to version 2.
If you are testing headlines, don’t change one word in the headline. Change out the entire headline and the message on the headline.
Headline 1: How To Train Your Dog To Play Fetch.
Headline 2: Seven Easy Steps To Training Your Dog To Play Catch And Return Games
Big variations in testing should give you some pretty conclusive results.
Split testing might seem like a daunting task; however, it is a skill that is well worth the effort to learn. Follow the steps above and you’ll find that split testing can be easily performed and will yield valuable information for converting your traffic to sales.
Same idea said two completely different ways. One of these will work better in the marketplace.
If don’t know which one. I have a suspicion but until I take them to the marketplace I don’t know.
This brings up a very important point. You can guess and speculate all day long. But until you put the ad, webpage or landing page into the market and let real people see it and respond to the advertisement, you are only guessing and speculating. The true test occurs in the market with real people voting with clicks ands and dollars.
Forget about outcomes and what you think should work and let the market tell you what they like and are willing to respond to.
One very important point to remember: The key to split testing is to only test one thing at a time.
If you test more than one thing at a time you will not be able to keep track of what made the different in your split test.
Don’t make this harder than it needs to be. Test (try) something. See if it works and does better than what you had better. If so, keep it. If not try something else. That is all split testing is. Try this then try that.
Finally. You are probably asking yourself so what do can I test? Here’s a list to get you
Greeting : Dear Friend, Dog Owner, etc
Size of headline
Color Of Headline
Submit Button Color
Submit Button Text
Call to action
Color of the background
The Header Banner
Position of Images on Page
This was written for me by David Bullock (known as one of the smartest marketers alive) and used with permission.