Market Research Forum
Pre Copy Testing - Printable Version

+- Market Research Forum (http://www.dobney.com/researchforum)
+-- Forum: Market Research Questions and Answers (http://www.dobney.com/researchforum/forum-3.html)
+--- Forum: General Research Questions (http://www.dobney.com/researchforum/forum-5.html)
+--- Thread: Pre Copy Testing (/thread-1159.html)



Pre Copy Testing - Ammar Affan - Thu 9 Jan 2020

Need to inquire ,if i need to test an ad in two phases pre and post, what should be the fieldwork methodology.

Should i recruit respondents and have my research conducted in Central location (CLT) or should i opt for random door to door methodology.


RE: Pre Copy Testing - Saul Dobney - Fri 10 Jan 2020

It depends on where you are, who you're targeting and your budget. Full random door-to-door (with a properly designed sample) is the highest quality of fieldwork - used for major government projects - but it's potentially very expensive the more strict you make the ESPEM design (equal probability of selection method). A cheaper face-to-face option may be able to hop on the back of a face-to-face omnibus study - a 'pooled' research survey of the type run by the large research agencies.

However, door-to-door is only feasible for consumer projects in relatively densely populated countries. It's unlikely you'd want to do door-to-door in the US for instance. This is when Central Location - sometime called hall testing - can be an alternative. Recruitment is less strict - it used to be mall-intercepts in the US for face-to-face studies, or a street recruit in Europe, or a RDD telephone call (with a suitably large incentive). You can also use central location for recruiting from small businesses as geography is less constrained.

For pre- and post- you'd be looking at quantitative research to measure the effect of the ads on behaviour and attitudes - so good sample sizes are advised - also adding to the cost. Face-to-face gets even more expensive if you're screening for particular target groups from a wider population as you have to make contact with a lot of people on the ground to get the target sample.

For a lot of these reasons, most people now would just jump straight to an online survey. There are still reasons for not using online - particular target groups who don't engage so much with surveys online, countries where face-to-face is still the only real reliable random sample, theoretical considerations, or a subject matter that doesn't work online, or the need to show or test physical product.

If you know how much the ad spend will be, what percentage of that spend can you afford to pay for testing to validate, and better still improve, the ad performance?