Copy Matters: 30 Minutes for a 10% Lift

Pad of Paper & PenOne of the more common excuses I hear for marketers not conducting experiments on their websites is a “lack of resources.” This generally refers to not having designers and developers at their disposal to create new versions of site pages, new graphics, new layouts, etc.

However, that is not an excuse from testing. That is perhaps an excuse from “radical redesign” testing. But there are still ways to run tests and get conversion rate lifts without tapping those resources. If you have access to a copywriter, or can write a bit of copy yourself, there’s plenty of testing (and learning) you can do, and I’ll give you an example in this post.

As I go through a test that I oversaw, what was tested, and the analysis of the results, keep in mind that the client probably invested around 30 minutes in this experiment…total! They used Optimizely, which is an extremely easy to use tool particularly if you’re only testing copy.

Also keep in mind that copy tests are literally the oldest optimization technique in the book. If you come from Direct Marketing, you’re nodding your head right now 🙂

The Setup

The client wanted to run a simple test that didn’t require any design or development resources. They also wanted to learn about what kinds of messaging worked better on their homepage, which is also their primary landing page. How they position their service to prospects is crucial to their bottom line. The test was run as a simple, A/B/C/D test where only the body copy above the fold was changed. No code changes to the site were required because the Optimizely tag was already in place site-wide.

The Versions

To protect the client’s emerging competitive advantage, I won’t show the copy or even attempt to paraphrase. Instead, I’ll lay out what the distinguishing characteristics were in each version, so you can see the differences for yourself. The website’s conversion goal is an application for a financial service, by the way.


  1. The service is flexible.
  2. Long-term, the service will save you 50% over competitors.

Variation 1:

  1. The service is fast and flexible.
  2. The short-term benefit is X.
  3. You can exit the service early with no penalty.

Variation 2:

  1. The service is fast and flexible.
  2. The short-term benefit is X.
  3. It’s free to apply and only takes 15 minutes.

Variation 3:

  1. The service is fast and flexible.
  2. The short-term benefit is X.

The Results (prepare to be fascinated)

results table

The Analysis

So what can we make of this? Here are a few high-level interpretations, all of which could use further verification via follow-up testing:

  1. “Fast and flexible” appears to perform better than “flexible” by itself.
  2. Emphasizing short-term benefits over long-term savings is a clear winning approach.
  3. The “early exit with no penalty” message introduced anxiety and was a conversion-killer.
  4. The simpler message of Variation 4 significantly outperforms the slightly more complex addition of “free to apply and takes only 15 minutes.” The free application and/or the 15 minute mention introduced anxiety or lessened the perceived value of the offer.

Holy, moly! That’s a lot of insight for 30 minutes of copywriting and test setup, don’t you think?!? And a 10% conversion rate lift is nice, too 🙂

What other interpretations can you make from the data from this experiment? Does anyone still not see the value of testing, whether or not you have the “right resources” available?

Note to legitimate commenters: I’m backed up with 1,600 comments, 99% of which are SPAM, so please be extremely patient with me, and if you know me personally, let me know that you’ve commented so I can approve you faster.

Leave a Reply

Your email address will not be published. Required fields are marked *