Testing different adverts in your Google Adwords account is easy to implement and can be a real quick-win within your PPC strategy.

Testing can help you to drive better-converting traffic to the site, improving your Google Quality Score (QS) thus reducing your CPC’s and all the while making the auction tougher for your competitors.

Here are my top 10 tips for testing PPC ads... 

Get the basics right

Include the search query in the advert itself because:

  • it’s what the customer was looking for, so it will make the advert more relevant to them (which can help your Google QS as well, bonus!).
  • It appears in bold on the search page, which in itself can improve CTR as you have better stand out.

If you don’t do this already, test it.

Deciding what else to test….

What are your competitors using in their messaging?

  • Are they using prices and/or claims that are better than yours? If not, then test using your claims/prices!
  • If their offering is more competitive than yours, don’t bother trying to compete on this. Consider pushing a different USP or trying something different in order to stand out.

Do you have a strong brand?

  • If so, test using your brand name in the headline and within the main body of the ad.
  • If your brand uses TV and other marketing channels, then test aligning the PPC adverts with the messaging on other media so that it resonates with potential and existing customers.

Calls to action:

  • Try testing the call to action e.g. how does ‘buy now’ compare with ‘compare prices now’. 
  • It can be worth testing this throughout the year, e.g. retailers in the run-up to Christmas can test ‘free’ vs ‘guaranteed’ delivery; travel operators can test urgency in the rush for last-minute holidays e.g. ‘book online now’.

Look beyond CTR

CTR is undoubtedly a key metric when testing in Google, but focusing on improving click-through can leave you forgetting your ultimate goal, whether that be sales, enquiries etc.

So ensure you use conversion tracking (e.g. Google Adwords') to measure conversion rate (and cost) by advert.

Take into account other factors like the average position and day of week when reviewing results as these can often skew results.

Stay organised

It’s tempting to test a multitude of things across several different keywords at the same time as it’s so easy to implement, but it can be hard to stay on top of everything!

Generally it’s easier to keep one ‘control’ advert in place, so you have this to benchmark your results against i.e. champion vs challenger. So don’t forget what the control ad is, on what date you started the test!

Take your time before making any decisions to pause/remove adverts. Let adverts build up their QS and gather enough data before deciding on next steps.

Follow Google’s best practice guidelines

Typically, the guys ‘n’ gals at Google recommend having three to five adverts live in an ad group at any time.

This suits most advertisers as well, as running with fewer adverts doesn’t allow for enough testing, and running more than this at a time could become unmanageable and it could take a while to produce significant results.

Make the most of Google’s ad formats 

Google has introduced several new enhancements to adverts over the last couple of years including:

  • Extended headlines.
  • Site links. When you’re in the top three positions you can have up to six appear at a time, so this is another test in it’s own right. 
  • Click to call on mobile adverts, test sending people to your call centre instead of your website.
  • Product extensions. If you have a Google Merchant Centre account, this is a further opportunity to sell your products!

Also, make the most of display URLs. These don’t have to exactly match the destination URL and so provide you with precious extra characters to play with.

Make the most of Google’s free tools

If you are short of time, or for lower volume areas of your account, try using Google ‘auto optimise’ rather than ‘even rotate’ so that the best-performing advert gets served more over time. This can be optimised to either CTR or conversion, depending on which is more important to you.

Don’t forget the landing page

Make sure your landing page prominently displays the search query and is consistent with the advert as Google’s QS algorithm takes landing page relevancy into account. As with the advert itself, keep an eye on conversion rates from this landing page and tweak and test the page accordingly to improve conversions.

A tailored advert linking to a homepage or irrelevant page can lead to high bounce rates and therefore wastage in your PPC budget.

Data is the new oil... so use it!

Use whatever sales/ demographic insight you have (e.g. internal data/ analytics) to determine the type of keyword to testing on and tailor messages accordingly e.g.:

  • Generic terms like ‘camera’ might be used in the research phase, so messages using ‘compare’ could work best. 
  • Typically, branded keywords e.g. ‘canon camera’ are used closer to the point of sale so test adverts containing product prices and ‘buy now’.

‘Test, don’t guess’

My colleague @richielewis introduced this mantra into our team, and it’s true. Stop trying to second guess what might or might not work, just test it!

Heledd Jones

Published 15 November, 2011 by Heledd Jones

Heledd Jones is Marketing Manager at Admiral Loans and a contributor to Econsultancy. You can follow her on Twitter.

11 more posts from this author

You might be interested in

Comments (4)

Paul Walsh

Paul Walsh, Founder & CEO at Infinity Call Tracking

Good article Heledd, and your so right about looking beyond the CTR, especially if you have a phone number on the site you need to push all this data back into AdWords conversions to be able to see which ad copy is performing the best.

Working back from the Sale or Lead from the phone call and the PPC ad copy that contributed to the call happening, then pushing this data into Analytics and then linking to AdWords... Many steps but the only way to truly understand the conversion rate. This has been very beneficial for the clients we have done this a/b ad copy testing for who depend on phone calls for leads and sales.

almost 7 years ago


Sanjit Chudha

A great common-sense guide; thanks Heledd.

almost 7 years ago


Edwin Hayward

One important technique that is notably absent from your otherwise excellent roundup is to change the actual URL of the landing site to an appropriate exact-match generic domain name for your most important keyword group.

The tests we conducted for our PPC white paper (downloadable from our website) showed a 105% uplift in CTR by using an appropriate generic domain name vs a "brandable" domain name, with all other factors (headline, body copy, bids, keywords, landing page) kept constant.

For larger PPC campaigns, the ROI is likely to be measured in weeks or months, even after factoring in the costs to acquire a decent generic domain for your industry niche or main product/service, and of deploying a dedicated mini-site to act as "landing page".

almost 7 years ago



I would say that split testing continuously with your ads and landing pages is key. A considerable improvement can be made just with these changes.

Also, a small % change in conversion rates will make a much bigger difference then a few more clicks.

over 6 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.