Marketing Insights and News | Website Magazine: Digital Marketing Tips and Blog

Conversion Tuning Crash Course

Written by Peter Devereaux | Aug 7, 2006 5:00:00 AM

Your website could be better. Many companies spend large amounts to drive traffic to their website, only to see this traffic land on pages that convert poorly. Improving the efficiency of these pages should be just as important as driving traffic to your site.

The Importance Of Conversion Tuning
Why should you care about conversion tuning? In most cases you must pay market rates to get significant volumes of quality traffic to your site. If your landing pages are not optimized to take advantage of your investment, you might as well be throwing your money away.

The economics of your program depends on your cost-per-acquisition (CPA) for a new client or prospect. The CPA depends on two factors: your cost-per-click (CPC) for getting traffic to your site, and the conversion rate (CR). Your CR is the percentage of unique visitors who take a desired, trackable action on your website - clicking through to another page, filling out an online form, downloading content, or making a purchase.



By increasing your conversion rate, your cost per acquisition drops without changing what you pay for the traffic.

What Can You Tune?
The focus should be on the "mission critical" parts of your site that lead to your desired trackable action. This can include any of the pages from your initial landing page to the last step in your trackable action process.

Once you have identified the pages involved in the test, you can decompose them into specific elements to test. For each of these elements (variables), you should test your current version and at least one alternative. Common variables include headlines, call-to-action, the promotion, sales copy, and the page layout.

Selecting a Tuning Method
The tuning method that you choose depends on three main factors:

Size of your test - The size of your test depends on the number of possible website versions or distinct "recipes." For example, if you are testing three headlines, four offers, and six button colors, then there are 72 possible recipes (3x4x6=72) in your test. As you increase the total number of variables and the number of alternatives for each one, the possible number of recipes grows very quickly. Some tuning methods can only handle a few total recipes, while others can routinely find the best answer out of millions.

Traffic levels required - Conversion tuning relies on statistics. You can never be absolutely certain of the answer, but you can get a very high degree of confidence. The amount of data required depends on the confidence level that you want to have in your answer. Do you need to be right three out of four times, nine out of 10, or 99 out of 100? Once you have picked your confidence level, you must wait to collect enough data to reach it. Making decisions based on too little data can end up costing you money.

Variable interactions - What are variable interactions? Simply put, it is when the setting for one variable in your test positively or negatively influences the setting of another variable.

If you believe that there are no interactions, then you must also believe that there is a best headline regardless of the accompanying offer. Clearly this is not the case. Each variable depends on the context in which it is seen. In online marketing, we want to create positive interactions. We want the offer to reinforce the headline, sales copy, the picture, and the call-to-action.

Some tuning methods assume that there are no interactions between variables. While you may get some positive results by ignoring interactions, you will not be getting the best results.

How To Get Started
You have three options to improve your conversion rate:

1) Do it yourself - If you are testing a few simple alternatives, you can rent the software to run your own tests in-house. You will probably need support from your marketing, usability, Web design, and I.T. staff.

2) Pay for an engagement - You can hire a company to run an engagement for you. The fees can be substantial, often ranging from $10,000 to well over $100,000 per year for multiple tuning tests. There is no guarantee of results.

3) Pay only for results - SiteTuners has pioneered performance based website tuning. It gets paid based on the profit improvement generated as a result of the tuning engagement. With its proprietary TuningEngineTM technology, SiteTuners specializes in large-scale tests.

Tim Ash is the President of SiteTuners.com and can be reached at Tim@SiteTuners.com. This article is a shortened version of a SiteTuners.com whitepaper that is available for download on the company's website.