Split Test your Ad Copy in 3 Easy Steps

All too often in meeting with potential clients, I encounter situations where they’ve been running the same ads month-after-month (or worse, year-after-year) with no attention paid to improving ad performance.  Failure to test ad copy deprives you of one of AdWords greatest benefits – the ability to scientifically measure response to varied messaging.  By testing your ad copy, you will find the winning ad combinations that deliver higher click-thru-rates and conversion, and at a lower cost-per-click.

It’s one thing to say “OK, I’m going to test” and then just start testing stuff haphazardly.  To get the maximum benefit, you will want to know why a particular ad does better (or worse).  And for that you need to follow a disciplined process. Here’s what I do:

Step 1 – Start with the Benefits
Begin by jotting down a list of your product or service’s benefits.  A good exercise is to answer the question, “why should somebody buy this from me instead of somebody else?”  Benefits typically fall into one of a handful of categories:  quality, service, selection, price, etc.  Identify yours.

Premiums, such as free shipping, coupon codes, discounts, etc. can also fall into this category.

Having listed your unique benefits, pick one and write an ad around it.  Test this ad against your current ad.  (Tip: make sure your campaign is set to “Rotate: Show ads more evenly” under your campaign settings). Watch the click-thru and conversion rates of these ads carefully.  (Another tip: make sure you’re evaluating a date range which corresponds to your test period).

If you think a winner is emerging, plug in your response data to this tool to determine if your test is statistically significant.

If the new ad fails to beat the old, cancel the new ad and move on to the next benefit on your list. Test that. If the new ads wins, cancel the old ad (which effectively makes your new ad the “control” ad) and continue testing the rest of your benefit list, always against the most recent winner.

Step 2 – Find a Winning Call-to-Action
Having found a winning benefit, write ads that combine this benefit with different calls- to-action.  Examples include: call today, search online, book now, request your free whatever, click here, etc.  Test and cycle through each potential call-to-action until you find your the best performer.

Step 3 – Find the Surprises
At this stage you’ve identified your winning benefit / call-to-action combo, and it is time find the other (and often unexpected) variables that can also make a big difference in ad performance:

• inserting dynamic keywords into your ads
• testing where things go in the ad. For example, play around with the benefit on line 1 and call-to-action on line 2, and then the other way around.
• include benefits and other adjectives in your title
• experiment with different capitalization, including your destination URL

One Final Note
Don’t make the mistake of judging an ad solely based on its click-thru rate. During this process you will write some very compelling ads that do a good job of driving incremental clicks.  However, if those additional visitors happen to be of lower quality, or if your site fails to deliver on the promise of your ad, the ad’s conversion rate may actually be lower.  Bottom line:  judge your ad winners based on conversion rate, not necessarily CTR.

And never forget, paid search is is wonderful because it allows you to test and measure everything.  Use this to your advantage and gain a leg up on your competition.