Alex Steer

Advertising effectiveness, analytics and strategy / about / archive

Word of mouth and relevant buying factors

343 words

Just finished reading McKinsey's briefing 'A new way to measure word-of-mouth marketing'. In brief, it recommends assessing (not really measuring) WOM impact by factoring in the volume of messages and various aspects which come together to form the impact of messages.

The aspects of impact are 'Network: Where are they talking?', 'Message content: What are they saying?', 'Message source: What is the trigger?', and 'Sender: Who is talking?' The second of these ('Message content') involves a judgement of whether or not what is said represents a 'Relevant key buying factor' for consumers.

Though this will appeal to marketers who are ultimately very sales-driven, this heuristic seems to undervalue damage to brands. A message that says 'Brand X is poisonous' obviously contains a 'relevant key buying factor', but does a message that says 'Brand X uses suppliers who use palm oil that is diminishing rainforests and leading to greater food insecurity in South America'? The latter may lead some very rainforest-minded consumers to avoid purchasing Brand X's products, but the total damage done may not translate quickly into sales. It may take years of incremental bad news for consumers to turn against Brand X and favour Brand Y. Alternatively it may lead to avoidance or restriction on purchasing in the whole category, if consumers come to perceive that the whole set of brands in the category are similarly undesirable. Or it may not affect sales much at all, but may damage the credibility of Brand X in a way that pushes up its required marketing budget by diminishing return on investment.

Identifying 'key buying factors' means knowing a lot about how consumers are feeling and thinking, and the impact of those thoughts and feelings on their purchasing decisions. The heuristic is only really useful, then, if you have a lot of data and insight to back it up. Otherwise there's a good chance your analysis will be wrong, and will turn a measurement tool into a rather formalised wild guess.

# Alex Steer (16/05/2010)