Brand advocates promote the necessity of defining a clear and evocative image, easily allowing consumers to identify and buy into the brand and corresponding products. This is achieved using “top of the funnel marketing” – packaging (a website and clear identity), findability (Google and mainstream newspapers, magazines, and blogs), and importantly premium advertising such as sponsoring sports events, rich media banners, billboards, radio, and TV.

Direct response advocates quantitative methodologies, as attributing ad-spend with KPIs such as new customers, their corresponding value, and the viral effect. The primary ethos is that advertising monies should be spent according to what achieves the best response per $1 – so the more high quality and targeted customers that can be acquired, the greater the overall revenue increase will be.


The inherent difficulty of working out whether one approach is better than another, the ideal mix or efficacy is the quintessential problem of attribution. In the case of brand spend, it is possible to measure to a certain extent customer perception (before and after) – whether using social media buzz, surveys, or up-lift in sales. However, even the most ardent analytics companies will accept measuring sentiment is as much an art as it is a science.

Whilst it is easier to attribute and thus measure direct response spend it is still far from perfect. For example a $100k campaign driving 100,000 clicks, 50,000 downloads and 10,000 sales looks simple enough. How many of those 10,000 sales would have happened anyway?  Additionally, how many sales have occurred that were not tracked because the sale occurred outside of the time-tracking window, or due to a technical error?

How to know what to do?

It appears based on the conflicting evidence that it is impossible to know which part of your spend is the most effective, bringing to mind the old adage that “50% of your advertising budget is wasted”. I recommend running a series of tests to try and ascertain with as much accuracy which steps to take and how to allocate budgets.

For example, consumer goods products – such as P&G and Unilever are more suited to high profile brand and experiential spend. Practical or utility products such as computer games are more accustomed to direct response spend.

Taking a practical approach

It is possible to reconcile Math Men and Mad Men. So taking mobile media for example, the one constant is that if brands wish to advertise they have to run banners – the question of where, at what price and with whom is ultimately up to the decision of the marketer and the nature of the product and their individual leanings.

However, it can clearly be accepted that if banners are going to be utilised, one should look to make the banners as effective as possible- whether one is adopting a direct response or brand methodology.

An interesting experiment that was run recently on Facebook showed utilising blank banners (as opposed to text and colours) were clicked 1.5x more than regular ads. While this is not necessarily the best approach for high-profile brands, it does show that the “brand effect”, non-scientific creativity, clearly creates a huge different in the results – which can thus be measured using direct response methodology (quantitative techniques).

For best performance agencies and brands should utilise both their mathematicians to measure the tests and their creatives to come up with cool and quirky ideas to gain attention & direct the most suitable people to engage with the ads.