I was reviewing the keyword specific conversion rate data of a consulting client of mine. I have been working with this client for a few months now, helping her improve the sales conversion rate of her website and we have had very good results, taking average conversion rates at her site from below 1% to just over 4.3%.
Now, one of the keys to improving conversion rates is to continually test and measure every detail of your website, marketing and traffic. While reviewing the data from one of her Google AdWords campaigns I stumbled across a few hard to explain results that perfectly illustrate the very large effect small differences can have on conversion rates.
The keyword “piano lessons” had a conversion rate over the last 30 days of 5.09%. The keyword “piano lesson” had a conversion rate of only 1.64%. Both of these keywords had the exact same AdWords title and description, the same average position in the search results and the exact same landing page on my clients website. The only variable is one letter in the keyword… an “s”. Lesson vs. Lessons. That’s it! Yet the plural version of the keyword (piano lessons) outsold the singular version (piano lesson) by over 300%!
Another key phrase that had an even larger variation was “how to play the piano” vs. “how to play a piano”. Common sense would say that these two phrases would convert almost identically… Wrong.
Again, with identical titles, descriptions and landing pages, “How to play the piano” converted at 5.92%, while “How to play a piano” only converted at 1.42%. That is a whopping 417% difference between “a” and “the”!
Short-term variations and fluctuations in the conversion rates of individual keywords or landing pages are common. However, the data on both of these keyword pairs was measured over a full 30 days and several thousand clicks for each keyword. I don’t have any easy answers why adding an “s” to a key phrase or changing an “a” to “the” caused such a difference in conversion rates.
Sure, you can guess at it and draw a few conclusions such as “plural keywords convert better than singular keywords” or “using ‘the’ in a key phrase is more specific and targeted than using ‘a’”. However, these conclusions are just guesses. The only real conclusion you can draw from this case study is that you have to test EVERYTHING!
If one word or one letter can have that big of an impact in an AdWords campaign, then a similar change in your main headline or guarantee can have an equally significant impact.
The lesson here is to test, test, test. What should you test?
- Test headlines
- Test guarantees
- Test bonuses
- Test colors
- Test sub-headlines
- Test your copy
- Test different prices
- Test layouts
- Test autoresponder messages
- Test your USP (Unique Selling Proposition.)
- Test ads
- Test keywords
- Test policies
- Test images and graphic design
- Test deadlines
- Test navigation
- Test your checkout process
- Test font sizes
- Test payment options
- Test EVERYTHING!
Now, when you are testing a change, only test one item at a time. This part is critical. Testing only one change will help keep your results as accurate as possible. If you change 4 items at once and see a 1% boost in conversion rate, it is possible three items are helping and 1 is hurting, so your conversion rate boost would have been greater with just the three changes.
The bottom line is this…If you really want to take your website’s sales conversion rate to the next level, you must absolutely commit to a focused and long- term campaign of testing and optimizing every aspect of your site. You never know when a small change is going to give you huge results