Ad Copy Testing for the Google Adwords Beginner

Ad Copy Testing for the Google Adwords Beginner

/

Just starting out in Google Adwords? Then you might be wondering about the best way to begin ad copy testing for your campaigns – new or old. Every campaign demands its own attention and optimization efforts, however, all have a similar foundation for ad copy testing. These ten ad copy testing steps below will cover what you need to get on your way in your first round of testing.

Where to Begin

Knowing where to begin can be half the battle in larger campaigns. I generally start most ad copy testing in the campaigns that are spending the most of my, or the client’s, budget.

1. Find the big spender(s). Take a look at what campaign is spending the most money first, and then make a list from there. Inside each high-spending campaign, you’ll find an ad group (or two, or three) that is spending more than others. Note these ad groups – that’s where you will begin testing. You want to optimize where the money is going, right?

2. Determine your testing goals. Do you want more conversions at a lower cost-per-action? Do you want more clicks and subsequent views on your site? Once you know what you want, you can tailor the tests for the goal and will know exactly what to look for in the results. Within each high-spending ad group are keywords that spend the most. Determine the keywords with the right traffic, a good click-through-rate and conversion rate, and begin thinking of ads for those well-performing keywords.

3. Set up the testing parameters. Make sure that the campaign settings within the dashboard are set to “rotate evenly”. Recently, Google made a change to this setting that would automatically deliver the best performing ad after 30 days. However, after a rather public backlash by those who argued that a 30 day window was not enough time to determine a winner, they changed the setting to 90 days with an optional “opt-out”.

NOTE: Make sure that the Campaign settings-Advanced Settings-Ad Delivery: Ad Rotation is set to “Rotate Evenly”.

 Adwords-rotate-evenly

What to Test

It might be tempting to test a completely new ad all the time – who wouldn’t want instant validation of their creative abilities? But your best course of action lies in patience with an A/B testing schedule that will leave no room for doubt as to what element tipped an ad towards success or failure. Test one variable at a time, measure your results, then move on to the next one using the winning feature of your last tested ad.

Variables to test on different areas of your ad depending on your product or service can include: discounts, trials, free shipping, full access with a subscription, etc.

4. Start with headlines. Headlines are the first text a person will read when they view your ad in the search results. You want something that will stand out from your competition as well as represent exactly what you are offering. Don’t offer “Model Train Sale” if you are not currently running a sale.

5. Move on to the description. Make your offer/product/website as tempting as possible, without leading the user on. You want a click that turns into a conversion or an interested visit to your site, not a click that bounces back 97% of the time, thereby wasting your money (and decreasing your Quality Score – but that’s another blog post.)

NOTE: It is best to test Line 1 of the Description, then Line 2. Try not to test both at once so you can clearly identify where your results came from.

6. Make a strong Call-to-Action (CTA). Be specific here – this is the final “closer” before a user decides to click your ad over your competitor’s ad. And remember, a CTA is just that – it is asking the user to DO something at the end of your ad “pitch”, so don’t waste precious ad space with confusing or weak CTAs.

Example:

Bad CTA: “Check Us Out.” Good CTA: “Book Your Stay Now!”

7. Test your display URL.While Google doesn’t allow you to have a display domain URL that doesn’t match your actual destination domain URL, you can test other variables like the keyword in the URL subfolder and whether to capitalize first letters of words or not.

Example:

Control Display URLKnittingSupplies.com Test Display URLKnittingSupplies.com/Bamboo-Needles

How to Measure

There are a number of metrics you can use to determine the success of your ad copy testing. Keep your own goals in mind when determining a winner. Click-through-rate is one of the first metrics you can look to for judging the success if an ad copy test.

8. Measure click-through-rate (CTR).  If your main goal was to get more views to your site, then an increased CTR on an ad = winner.

9. Measure Conversion Rate (CVR). Although many may consider this to be an additional or optional metric, it is a good number to keep in mind. If you wanted more people to complete an action on your site, then an increased CVR = winner. Additionally, look at the Cost-Per-Action (CPA) and whether it is within the winning range that you need to be in for a positive ROI.

And finally, one round of testing alone is never enough. Therefore:

10. Rinse and repeat. Continually test new ads and stay on top of your competition.

Ad copy testing is specific to each industry, each campaign, and the ROI of your company or your client. Make sure that all your tests comply not only with Google’s Ad Copy Rules, but also with the right brand or product messaging of your company or client. Stay tuned for more advanced copy testing techniques!

Have a question or comment? I’d love to hear it!

 

 

 

Sarah is a former LunaMetrician and contributor to our blog.

  • http://www.clickclickmedia.com.au adwords management

    A good article and an important part of any adwords campaign, thanks.

  • http://www.lunametrics.com Sarah Peduzzi

    Thanks for reading!

  • http://www.swanmigration.com.au venkat

    Hi Sarah,

    great article. simple, lucid and upto the point.
    I wish to enter into adwords a few weeks from now. Very useful for me.

    Venkat
    Perth, Australia

  • http://semexpert.co.za Colin Brazendale

    Hi Sarah,

    Good article. I agree with your points above.

    How do you tell a client why not to add his telephone in add copy?

    The whole point is to get the browser to click your advert. Nobody will just phone you without seeing what you are offering.

    Regards

    Colin

  • http://bcwebpros.ca Dave

    Excellent article! It’s been a while since I’ve had to run a PPC campaign, have mainly been doing on page/off page optimization. It’s a whole different beast that’s for sure. Great refresher for any one new or old to Google Adwords, thanks Sarah.

  • http://adproof.com Jonathan Nelson

    These are some excellent tips, Sarah. I particularly liked the tip on testing your call-to-action copy.

    Listen, we just launched a new app called AdProof that makes it super easy to test and optimize your PPC ad copy. It’s pay-per-click user testing for Google AdWords and Bing Ads!

    Head on over, try it out and let me know what you think.

    http://adproof.com

    Jonathan Nelson
    Founder of AdProof

  • http://na Carrie

    I’m curious can you add more than one new ad to an adgroup to test at one time?

    I usually add 2-3 new ads in adgroups with 2-3 ads currently running.

    I recently was told that in order to complete a proper A/B test there needs to only be one ad running and one new ad tested against it.

    Can Google’s system figure out what the best performing ad will be if there are more than 2 ads? What is the minimum volume of impressions/time needed to test this?

    • http://www.lunametrics.com Sarah Peduzzi

      Hi Carrie,

      I usually pause all ads that are running, copy the one that will be my “control” and then enable that one. Then I create a new ad to test against the new “control”. You can set campaign settings to distribute ads evenly and Google will do it’s job to send traffic in equal parts to whatever ads are running. My rule of thumb for testing is at least 100 clicks, but I generally work with niche clients so that’s pretty high for them. For higher volume clients, I would bump the click amount to about 500. This rule of thumb can change constantly though! Just make sure you have a statistically significant amount of data in front of you when making any decision.

      Hope this helps!

  • http://- Cody

    I have a question alike to Carrie’s (above). What is the best way to get dead accurate results to know which of your two ads is the winner. I have heard of many processes like creating 3 copies of the control so that it isn’t a time run bias in favor of the older control etc. What is your take?

    Thank you for your time!

    • http://www.lunametrics.com Sarah Peduzzi

      Hi Cody,

      I answered Carrie’s question, so let me know if that helps!

      Thanks for stopping by!

Contact Us.

LunaMetrics

24 S. 18th Street, Suite 100,
Pittsburgh, PA 15203

Follow Us

1.877.220.LUNA

1.412.381.5500

getinfo@lunametrics.com

Questions?
We'll get back to you
in ONE business day.