#19 of 25 Things we know about what we don't know about Marketing Effectiveness
Television is the most effective marketing communications medium. That's what the TV networks claim. That's what the ad. agencies claim. Proven TV case histories abound. It's what the econometric models "prove". So it must be true. Or is it?
It may not be true! We've never even given other media a chance or even a look in.
It's something that's bugged me for years. For sure, television is a very powerful and persuasive medium. But why is it that the volume of evidence is so swamping & favouring the domination of television as the brand leader effective medium?
It took me years to work it out and I have only recently proved the empirical reason. It's not that television IS the most effective medium. What happens is that whenever we plan TV campaigns we do three things that introduce a massive & unique bias in favour of TV.
- First, we deploy far more absolute amounts of cash behind TV than we do for any other medium.
- Second, we then compound the matter by planning that money in far more intensive slugs of cash per advertising week.
- Third, in multi-media campaigns we ALWAYS start with TV
Econometric models always start with the first and largest effect. So if we plan TV that way it's inevitable that TV will be proven to work best. If you index the average TV campaign weight at 100, the average newspaper/magazine campaign is planned at 35 - a third of the weight. If the intensity of the average TV campaign is set at 100 units per week, the average radio campaign is planned at 25 - a quarter of the weight per week. I have as yet insufficient internet campaign evidence to be certain, but the initial analysis suggests that if the average TV campaign eats up $100 per week, the average internet campaign cost only $15 per week.
With this massive built in bonus in favour of TV it's no surprise that TV "works best". No other medium stands a chance of proving its effectiveness.
The usual response to considering alternatives to TV is "but it works, so why drop it for some experiment in marketing effectiveness?" That's an understandable reaction. But if we donate to the TV networks such a head start in extra money, it comes at no surprise that TV effects dominate.
The evidence suggests that if marketers start planning non-TV with the same weight. intensity and scale as has become standard TV practice you will be amazed how cost effective alternative marketing communication channels can be.
Hi John,
There is a danger that we confuse effective with efficient. Effective in absolute terms doesn’t mean that funds were well allocated which is a mistake many people seem to be making. Econometric models can be highly misleading if they only look for average effects as opposed to marginal ones for individual channels. By analysing the marginal returns from activities, we can then state with greater confidence as to whether a channel really requires a greater or lesser proportion of the overall funding for a campaign.
The statement you make "Econometric models always start with the first and largest effect. So if we plan TV that way it's inevitable that TV will be proven to work best." doesn't have to follow if modelling is done carefully and without bias. How often either of these two goals is achieved is another matter entirely and very much depends on how and why modeling is done. Is it being done by a specialist independent analyst and is it a post-hoc back patting exercise or will the analysis feed into future planning.
Obviously there's another factor at play here - traditional media agencies depend on having large TV buying operations and that causes bias in their thinking. How often they use econometrics models to prime their thinking is open to question. Not enough might be a reflection of my experience.
This is a fantastic series of posts.
All the best,
John
Reply to this