DM p20-21 Opinion Dec08.qxp 02/12/2008 13:09 Page 21
Opinion
assume that none of the factors seriously interact deliver 4.8% response.
with each other. Et voila!! In one fell swoop, we have learned a great
Assuming that a sample size of 40,000 is required deal about what hits the buzzer for prospects, rather
to achieve the required level of statistical reliability, than the ‘one step at a time’ testing regime of old,
we would have to run ten AB split tests amongst which proceeds at snail’s pace and takes forever.
800,000 prospects to gain these insights, whereas Interestingly, MVT appears to be a more sensitive and
The methodology
MVT only requires one test amongst a mere 480,000. accurate testing methodology due to its more holistic,
is a superior form
In these heady days of digital printing that can multi-dimensional approach, as opposed to the uni- of testing that
swap headlines, substitute copy panels and visuals at dimensional approach of conventional means.
delivers learnings
the drop of a hat, the production cost of working up Comparisons between the two suggest that some
as to the most
the twelve different test packs is not that great, and changes identified as significant by MVT are not
effective
when it comes to creating test emails or posting web- picked up by traditional testing.
pages, it’s even easier. However, as with all things new, proceed with cau-
execution.
In a nutshell, MVT works out the contribution that tion. MVT is a powerful technique, but it is not with-
each alternative headline, offer, format etc makes to out its complexities. Serious interactions between fac-
the response rates of all the different packs. In this tors can give an entirely erroneous view of the effect
way, the most potent headline, offer, format etc is of some of the test variables. Knowing how to identi-
identified in the chart on the left. fy interactions and adjust the orthogonal array
We see that changing the offer has the greatest accordingly is essential. And ensuring that reliable
impact on response; the free widget adds a whopping response rates have been elicited by the test, given
1.1% to response compared to the 10% discount. historical norms, is also important.
Other significant increases are format and proposi- Nevertheless, this is the first true innovation in
tion, while changing the covering letter, headline and testing disciplines that represents a real step forward
envelope have a relatively minor impact. that I have seen for years.
n
Hence, the optimal permutation is: longer covering
letter, plain envelope, family proposition, free widget, Ian Robinson is director of insighttmw, the
quality headline, beach visual, sea visual, informal data and planning arm of digital and DM
text, phone CTA and C5 format, which is projected to agency Tullo Marshall Warren.
www.dmarket.co.uk Database Marketing December 2008 21
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44