Ad testing help improve ROI by increasing conversion rate and reducing advertising costs. You can learn about what parts of your business offering appeals most to your prospective customers by honing ad text, especially your headlines.
Ad Testing Tips
- Be organised. Use Google Ads labels. Something as simple as a calendar can help you to keep a schedule. Document start and end times.
- Establish a testing threshold and stick to it. Wait until you have served enough impressions to be confident in your results.
- Limit how many elements you test within your ads. Make different variations similar enough that you learn from their differences.
Ad Rotation Settings
Ad Rotation controls what percentage of impressions each ad version receives. It is used to help test adverts (A/B testing) and ultimately control which adverts receive the most search impressions.
The end goal of ad testing is to work out what is the most effective adverts are- and make sure they receive the most impressions. Google can automatically show ads more often that have either the highest CTR or the highest conversion rate, or manual ad testing can be done.
The “% served” statistic showing how often each ad has been shown in relation to the rest of the active ads (within the same ad group.)
Automatic Ad Rotation (Testing run by Google)
This campaign setting automates ad testing by allowing Google to show the winning ad version more often. Google then automatically show ads more often that have either the highest CTR or the highest conversion rate.
- Optimise for Clicks (default setting): Allows Google to automatically show ad versions that get more click-throughs than other ads versions. This setting is useful for new accounts or campaigns to improve quality score and gather performance data. Optimising for clicks is showing the adverts with the highest CTR more often. This is useful was of improving CTR- or doing initial testing on new accounts if limited conversion data is available.
- Optimise for conversions: Automatically show advert versions more often that get the most conversions like leads or sales. This may generate fewer clicks & traffic- but more suited for direct response advertising.
The ratio of ad serving gets based on the degree the winning ad is better than the losing ad/s. When a clear winner is established- it is possible for the other advert/s to get less than 1% of the impressions. If no winner is established- the ads will serve on an equal basis.
Manual Ad Rotation
It is also possible to rotate the adverts equally- and not allow Google to show the best performing ads more often. This is for more advanced ad testing- where losing ads get paused manually. (The losing ad gets paused manually, and a replacement “experiment” ad version is created to beat the winning control ad).
- Rotate evenly: Gives equal preference to all adverts in the ad group, regardless of how each ad performs. After 90 days, the ad rotation in this ad group will automatically begin to optimise for either clicks or conversions.
- Rotate indefinitely: Shows ads in a manner similar to “Rotate evenly”, but does so for an indefinite amount of time, without ever allowing Google to automatically optimise.
Ad Copy Testing
Ad Copy Split testing has revolutionised advertising. Running more than one version of the advert, lets advertisers see what appeals most to their prospects and customers with concrete quantitative data.
Online PPC platforms such as Adwords, Bing, Facebook, Twitter, Linkedin offer A/B testing of adverts.
- Split testing means that advertising assumptions can be tested, proven, disproven and improved upon.
- Testing results show advertisers what customer think is the most attractive part of their business offering. Results can be measured using metrics such as click-through rate, conversion rate, cost per conversion, cost per impression, return on ad spend and more.
- KPI. Advertisers can test and learn what appeals to their customers while optimising efforts to meet the primary advertising objective.
- Every company has a business offer consisting of unique selling points, features, benefits, guarantees, promotions, calls to action, proof points, testimonials and qualifications.Even if your competitors are very similar, ad text results show what parts of your business offer and copy appeal to your prospects and customers.
- A/B test results show the best ad text versions while multivariate test results show the performance of ad elements like headlines, descriptions and even wording and phrasing as individual aggregates across ad groups or campaigns. Understanding the costs and conversions for each ad text and ad part is critical for making sure the most useful parts of your business offering and copy get emphasised.
- It is critical to look at the current ad copy performance- before creating new ads.
- Landing pages are important for both quality score and conversion rate. Google use automated systems and human evaluation to determine landing page relevance. Improvements such as increased time on site can lead to higher quality scores over time.
- Landing pages should be relevant to the advert copy by providing more detail about your business offering, features, benefits, guarantees, unique selling points, promotions, service, product, calls to action, locations and proof of services such as qualifications or reviews. A/b tests ads and landing pages consecutively to improve quality score and conversion rate.
This table is a way tool for capture your business offer as an elevator pitch- and testing what works the best.
Future of Ad Copy Testing
Imagine a website that could automatically optimise its design based on website user engagement by testing for the best variations of the advertisers business offer.
Using a combination of A/B and multi-variate testing to determine the best copy, the best colours, elements, weight prominence, sizes. An advertiser could merely enter details about their business offering into their web analytics system. The system could then test and refine a website until the engagement rate could not be increased
An advertiser could merely enter details about their business offering into their web analytics system. The system could then test and refine a website until the engagement rate could not be increased any more. This would be letting a customer majority engagement design the advertiser’s website