Monday, October 3, 2016

The Remarkable Resurrection Of Pre-Testing

By: Tom Ewing, Senior Director, BrainJuicer Labs

Ad pre-testing has had a rough ride in the digital era. A few years ago, the talk was all of testing as obsolete: why even bother painstakingly screening your communications, when online platforms were such a great opportunity to move fast, publish at will and learn as you went along? Agility was the watchword, and launch-and-learn was the strategy.

It’s certainly true that responding to this challenge helped give ad testing a necessary kick up the backside. Over the last couple of years, many research firms have finally understood the need for speed, and developed rapid-turnaround versions of ad testing solutions. The rise of self-service platforms like Zappistore has helped drive this.

But none of this matters if pre-testing itself is irrelevant. It's always had its enemies - no creative has ever said "Yes! Please test my work more!" - but until recently it never faced a potentially existential threat. Fortunately, the fundamental principle of testing content is as important now as it ever was. This isn't because the pace of change has lessened. In fact, it's because things change so rapidly that one year's revolution can become the next year's obsolete philosophy. And that's what happened to launch-and-learn.

The launch-and-learn strategy rested on two assumptions. Firstly, that measuring earned media (social sharing) was a reliable and consistent way of identifying the best content in the market and getting it to an audience. A few years ago, this was probably true. In the IPA's #SocialWorks initiative, which looks for validated case studies of social media ROI, one of the best examples came from Cadbury's Creme Egg brand. Creme Egg - a seasonal obsession of pumpkin spice proportions in the UK - optimised its Easter content by simply launching a lot of things, and seeing what its Facebook fans liked and shared. If the fans were into it, that’s where the brand put its money.

Unfortunately, shortly after that the organic reach for earned media on Facebook – governed by the proportion of fans who would see each piece of branded content – was drastically throttled by the platform. These the number of fans who will see content without paid promotion is tiny. At the same time, a growing understanding of how things spread and become successful meant an increased appreciation of how much of a role sheer luck played in social media success. The Creme Egg case study became the woolly mammoth of ROI - a magnificent specimen, but doomed to extinction by a changing climate.

The second plank of the launch-and-learn approach was transparency. If you can trust the metrics your ad networks and social platforms are giving you, they can be a big source of in-market insight. But this turns out to be a big “if”, with this week's Facebook video viewing figure controversy just the latest example. Viewability, ad fraud, and blocking of intrusive ads have become constant thorns in the industry’s side.

So launch-and-learn has lost viability and reliability. Throwing content at the wall and seeing what sticks is no longer the right strategy, any more than counting likes is. But it’s done its job. Faced with doom, pre-testing solutions evolved, and have become faster, more relevant, and a better fit for purpose than they were five years ago. Bulk pricing is becoming more common, as are turnarounds measured in hours and more use of innovative technologies like emotional measurement and in-context testing.

As the brief heyday of launch-and-learn fades, there's fresh interest in pre-testing as a digital tool from big brands and companies who know that the smart play is to make sure your content really will capture your audience's hearts and eyes, before you invest in it.

No comments: