Top

Increasing Impact & Engagement through Advertising Testing

Denyse Drummond-Dunn | Nov 23, 2016 53 views No Comments

Share on LinkedIn

If you prefer to listen to the PODCAST of this post, click HERE. Don’t forget you can subscribe on iTunes too!

One of the most popular evergreen topics on C3Centricity is advertising testing. Therefore, in light of the expanding channel options available to marketers today, I think it’s time I proposed an updated perspective.

There are countless posts which discuss how to A/B test a campaign on FaceBook or how to  pre- and post-test advertising. What seems to be lacking is an objective view of IF  you should be testing your advertising at all. So this is what I want to discuss here. I hope you will find it helpful in reviewing your own opinions concerning advertising testing. 

Should You Test Advertising?

If you work on the client side and ask your colleagues in an advertising agency, most of them would probably scream NO! That’s not very surprising. Countless teams have suffered at the hands of market research and the over-zealous testing of their creative – in a usually very uncreative way.

There have been many attempts at defining metrics to evaluate advertising. One of the biggest challenges from my perspective is whether or not you should test a campaign or each individual ad separately. But more on that in a moment. First I want to review the actual decision to test.



Should you test an ad built to increase awareness in the same way as one built for encouraging trial, purchase, repurchase, loyalty or advocacy? My answer would be a very Swiss “It depends”.

Firstly you have to be clear about why you are advertising in the first place; what your campaign is trying to achieve, its objectives.

It still amazes me how many companies develop a new campaign simply because that’s what they do each year. Hopefully, each new campaign has a link to the preceding one, but even that is not always obvious.

Therefore start by identifying with whom you want to communicate and why. And share this information with your ad agency; it’s surprising how many clients don’t! Remember to give as complete a description as possible of your target audience, including the who, what, where and why. (Our 4W™ Template is great for storing all the information in a one-pager)

When to Test

Many companies have a standard process of testing ads before they can be aired. This is usually referred to as copy-testing. An ad must appear in the “top quadrant” on both impact and persuasion before it can be used. While this is admirable for its thoroughness, it often results in multiple ads being developed, to ensure that at least one of them meets these criteria. The feeling is that more is better. If you test two, three or more ads, you can then hope for a “winner.” What’s wrong with that?

Well, in my opinion, quite a lot.

You’ve just wasted a lot of time, money and energy in developing multiple ads, when you know you’ll most probably only use one. Of course, your advertising agency won’t tell you to stop this practice, as they’re getting more work than you really need.

It’s time to think differently about ad testing and spend your valuable resources more wisely. (>>Tweet this<<) 

Once the ad agency has developed a number of campaign concepts or ideas that meet your carefully defined objectives, then that is a much more efficient time to test. OK, so the ads are likely to be in storyboard format or sketches, but most people will understand the message you are trying to convey – if it’s effective of course.

Don’t wait until you have gone further in the production and created animations, final prints or even complete films before testing. (>>Tweet this<<) That’s just a waste of resources.

If you wait until that later stage in the development process, you are also more likely to designate one “winner” when in fact they could all be good – or bad!

Working with concepts will help you identify the real winning ideas, which can then be developed into a final version or two for copy testing – if you must, but more about that in a moment.

The earlier you test, the more resources your ad agency can then concentrate on the most relevant concept(s), rather than diluting their efforts to give you the wide choice requested. No wonder ad agencies don’t like copy testing!

What to Test

Another reason for testing concepts rather than finished ads, is to ensure that they can be turned into a campaign.

I have witnessed many so-called “Big Ideas” that were superb as they stood, but which were impossible to visualise other than in the single version proposed.

If you show your early work to consumers, they might be inspired by the idea and suggest other related situations or portrayals. You will then have an indication of the campaignability of your idea.

For regional and global campaigns, there is often the added complication of the translation of the idea into multiple languages.

There are many concepts in English that don’t or only poorly translate into other languages.

English is a wonderful language that is particularly suited to advertising. It provides many opportunities for word plays, idioms, acronyms, slang, compound words and other wonders of its grammar.

In addition, English is known for its extensive vocabulary, which is especially useful in advertising copy-writing. Whereas in another language you might only have one or two words to express a particular meaning, English may have five or six, each with subtle differences.

If you’d like to see some great examples of advertising messages “lost in translation” check out this fun article from Business News Daily, or this one from The Balance. They’ll both have you laughing out loud!

How to Test

Depending upon their “standard” process, many companies will tend to use the same pre-testing method and evaluation, with no regard for the campaign’s objectives.

As previously mentioned, some clients I know must score in the “top quadrant” on the usual copy testing impact and persuasion metrics before their ad can be aired. However, there are valid reasons to accept lower scores on one or other of the metrics, depending upon the campaign’s objectives.

For instance, if the campaign’s objective is awareness, then a lower persuasion score may be acceptable. Likewise, if you are looking for your customer to take action, then a lower score on impact may be acceptable if the ad scores high on persuasion.

Some of the best – and most useful – campaign testing I have ever seen, was done qualitatively. But that alone won’t work unless you then allow the creatives, market research and insight groups to discuss the results together – ALONE!

In my experiences of this, it was exciting to share consumer opinions directly with the creatives. They too found it stimulating to share their ideas and get feedback based on real consumer input. Whoever said that creatives don’t like testing are wrong; they just dislike judgemental, sometimes disrespectful and bland numerical results with little if any depth of analysis.

In “Copy Testing: A Confident Path Toward Mediocrity,” Tom Bick, named one of the top digital marketers of 2014 by Ad Age, claimed that copy testing tends to penalise forward-thinking marketers. He argues that the average person will default to comparing a new ad with those they’ve already seen. In other words, they will err on the side of less creative campaigns. You can read more on this in the excellent Big Commerce article HERE.

A Unique Alternative to Copy Testing

One solution that I propose to my clients is a unique and powerful testing methodology developed by PhaseOne. Their scientifically based, proprietary technique, is based on over thirty years experience of academic work and real-world validation. Their knowledge base includes an extensive foundation based upon analytics in human behaviour, anthropology, culture traits, entertainment, education, communications and marketing.

This enables them to accurately explain how your target will react to your messages and even more importantly the reasons why, without actually speaking with consumers. This can be particularly useful for testing ad ideas for new product concepts.

In comparisons with standard copy tests, PhaseOne’s technique has been shown to give similar outcomes, but with a greater depth of understanding. More importantly, it provides a clear explanation of the reasons why consumers react to an ad as they do. This makes it far easier to improve the ad, whether by cutting out sections – which saves money – or improving the explanation of benefits. If you’d like to hear more about this unique methodology, especially if you’re having trouble speaking with your own target customers due to legal or confidentiality issues, I’d be happy to share some case studies. Just drop me a line HERE.

The Six Rules of Advertising Testing

In summary when it comes to testing your advertising:

  1. Know with whom you want to communicate.
  2. Know what your target audience wants to hear.
  3. Know why you are communicating, what the message is that you want to send.
  4. Know which concept(s) or ideas have the most resonance with both your target audience and objectives, and why.
  5. Know how the concept(s) will develop into a campaign across media.
  6. Know how you are going to communicate, the most relevant medium and channels for your target audience.

Do you abide by these six rules before pre-testing your own ads? If so, well done; if not, perhaps it’s time to review your own process.

The Future of Ad Testing

In conclusion, let me finish with a few words about the future of pre-testing. Although advertising testing supposedly started in the mid-1800’s, it wasn’t until the 1950’s that performance metrics became the holy grail of clients, ad agencies and media sellers alike.

From Day-After-Recall, to Persuasion, and from Brand linkage to Moment-by-Moment systems, it wasn’t until recently that the importance of emotional rather than rational responses to advertising gained support.

Today, emotional analysis has become widely available. Customers’ reactions to the ads are measured, usually on the six universal emotions (happiness, sadness, surprise, fear, disgust, anger) plus neutral. While it’s still early days in understanding the connection between emotional reactions and brand impact, things are definitely moving fast.

Digital and Traditional Media

Interestingly, when I was doing research for the original post on this topic a couple of years ago, almost all the articles I found were about the testing of online advertising, comparing PPC (Pay Per Click) and the positioning of paid, earned and owned media.

However, around two-thirds of budgets were still being spent on traditional media – at least in 2014 – and Statista showed that consumers still trusted it more than new media.

A more recent post on MarketingCharts still shows traditional media leading the ideal channel mix for marketers – but for how much longer?

Spend on digital is increasing more rapidly than was at first expected. In the US, the UK and China online is expected to surpass TV next year. Both MarketingCharts and eMarketer have made similar predictions for many markets in just the past couple of months.

I have covered primarily pre-testing here, yet I know many companies who are satisfied with running only post-tests. They admit that it is because they never have enough time to pre-test their ads. At least to me, this highlights a clear lack of concept testing in the first place. If you are one of these organisations, then this post has hopefully persuaded you that there is a better way.

Pre-testing is important, no vital, for clients, media and ad agencies alike, to do more of. At least doing early assessments will provide material for those development discussions – before it’s too late!

Do you agree? Do you have a different approach to advertising testing? If so, please share your ideas below.

C3Centricity used an image from Denyse’s book “Winning Customer Centricity: Putting Customers at the Heart of your Business – One Day at a Time

This post is an updated version of the one first published on C3Centricity in 2014.

Print Friendly, PDF & Email

Republished with author's permission from original post.


Recent Editor's Picks:


Categories: BlogDigital Marketing

53 views

No responses yet, why not leave yours?

Add Your Comment (All comments are reviewed by moderator, no spam permitted!)