Use Online Testing For Higher Profits
The best way to maximize your marketing results is by testing various elements of your marketing to find out what works best. BizGrowth founder Brian Offenberger gives you an overview of how testing can help you boost profits while lowering costs.
To be able to effectively analyze how to get the best results and profits from your website, it is crucial that you understand what online testing is and why it should be a major part of your online marketing efforts.
You may have heard the joke about the business owner who said, “Half of my marketing is working real well……I just don’t know what half.” Testing will answer that precisely, so that you know exactly what is working and what is not, often in just a few hours. As a business owner, it’s important to understand that you are not your customer, so any attempts to figure out what they will respond to versus what they won’t are just guesswork. Testing removes all the guesswork and shows you in real numbers, in terms of traffic and dollars, what will get your prospects and customers to act.
So what is testing? Testing is changing of an element or elements of your marketing message, tactics, channels, design, pricing, promotional offers, etc. to determine what works best. Testing allows you to ‘do things wrong quickly’, and get rid of what doesn’t work before you waste a lot of your marketing dollars.
Why should you want to do this?
- Testing allows you to get the maximum return for the minimum investment, the ‘holy grail’ of all marketing.
- You can cut your losses quickly and get rid of ineffective marketing
- Learn what your prospects and customers are willing to pay for
- Gather marketing intelligence that can be applied to other sales channels
As with many things, there is good news and bad news with this. The bad news is that it takes a lot of trial and error to determine what works best; the good news is that when you determine what works best, what marketers call ‘the control’, you can use it successfully for years and years.
What Are The Things I Can Test?
There are many aspects of your marketing you can test online. You can test:
- Price points
- Promotional offers
- Website design
- Headlines and copy
- Email messages and subject lines
In the beginning of your testing, you want to test one variation at a time. For instance, you would have one landing page with one price point, and another one with a different price point and measure which has more conversions. Many people who test price points test with the regular price and a lower one, but experienced marketers know that testing the upper limit is more profitable! They keep testing higher pricing levels until orders start to drop off dramatically to determine the highest price with the corresponding higher margin and profits they can sell for.
Common Types Of Testing
The most common type of test is the A/B or split test, where one group gets a message, and the other gets a different message and you see which one gets better results. It’s important to test properly to get a relevant result! You can compare one price point against another, one headline against another, different ad copy, even the background color of the website; just make sure the comparisons are ‘apples to apples’.
Another method of testing is called ‘multivariate’ and as the name implies, in this test you are testing a combination of things at the same time.
Then there is ‘Taguchi’ testing, which is like a multivariate test on steroids.
You do not need to do all of these tests, for most people getting started the A/B split test is the easiest to perform and there is no guess work or interpretation that is involved. One thing works better, gets more sales, and you go with that one and discard the other. With the other types of tests, there is some subjectivity and interpretations as to which of the changes produced the better result.
How To Get Started With Testing
- Determine what part of your marketing you want to test
- Put the test online
- Begin testing with search engines, either organically or through paid search
- Use the FREE testing tools provided by the search engines
You want to use the search engines first because of a number of reasons:
- You get a consistent, and often predictable flow of traffic
- You can control how much money you spend on the test
- You get more and better marketing and customer data from this source than any other option
- You have the most control over the testing process.
- Once things work well with search engines, they will most likely work with all other methods.
Following the roll-out plan (below) in the order listed. The costs and risks, and the lack of measurability increase with each channel. This is the most important reason to start with search engines first. It’s cheap and measurable!
The Marketing Channels – Get It Right and Spread The Word
- 1. Search engine test
2. Search engine optimization of website (conversion optimization)
3. Other search engine advertising
4. Email promotions and newsletters
5. Blog tagging, podcasting, and other online channels
6. Affiliate marketing
7. Press releases
8. Direct mail
9. Print advertising
10. Other media (radio, TV)
I can’t stress this enough; start with search engines because it is least expensive and you get unambiguous results fast. Once you have success with the search engines, then expand your marketing efforts with these other channels using the same things that led to that success. Many marketing plans have been failures because people have started with the most expensive methods and have gone broke waiting for meaningful results.
Google is by far the most used search engine on the planet, and they provide serious analytic tool for FREE if you set up a Google account. The three most important tools they offer are:
Numerous books and countless articles have been written on each of these Google tools. The more you learn, the more you will earn!
A Word About Statistical Validity
When I went to college, I had to take several courses on statistics and validity scores. Very complicated formulas that I won’t bore you with, but you do have to have some understanding about statistical validity or you can make the wrong conclusions and sabotage your efforts.
The most important rule is that you have to have enough data to make any valid determination of what your customers are ‘telling you’. For example you cannot make any real determination of results from 50 website visits. A test with only 50 people landing on your test sites is not nearly enough to draw any conclusions. You probably need several hundred before you can start to see the real patterns emerge. Some companies won’t make any conclusions from a test until they have thousands of pieces of data to interpret. Depending upon your traffic patterns, you could achieve this in a day or a month or more. The more data you have, the more relevant the results.
Google AdWords and Google Optimizer will help you to determine validity, but be careful, these are only basic tools.
If you need help setting up a testing program for your business, tell us how BizGrowth can help you.
Tags: a/b testing, AdWords, Google analytics, Google optimizer, Marketing Testing, multivariate testing, online testing, testing
Bookmarks [ del.icio.us | Digg it | Blogmarks | Diigo | FeedMarker | Google | Netscape | Yahoo MyWeb ]
