Automation of advertising is an issue to which I’ll return, repeatedly. We often are involved with clients who have large catalogues of products, one of the areas in which automation works best. Everybody has a different structure to the catalogue, different needs and different rates of update. Complexity is added with the presence or absence of product descriptions and more or less rational naming of pages on the web site. Of course, this is all compounded when working in multiple languages.
The current issue exercising me is creative – the advert copy. Just how much templating can you use and when do you flag that you need an (expensive) human copywriter to rescue the machinery? There’s plenty of people who write that templated copy doesn’t work. Our experience says that templated copy can achieve better than 50% CTR.
Are there obvious factors that make templated copy more or less likely to work?
I think there are, but this ties back to another area I’ll probably revisit – strategy and the role of paid and organic search. Search marketing works best when users know what they want, but they just don’t know where to find it. That’s also the case with templated adverts – if users know what they want, advertisers can offer a template that puts the right names and benefits exactly where users expect to see them.
If you have a niche product without significant market awareness of what you are selling, then templated adverts probably won’t work. If you are selling well known products then a template probably can be made to work.
The decision about templates comes down to understanding the consumer and their likely intent. And intent is another major topic I’ll be revisiting.
