The great media CPM debate
Are Facebook and YouTube ads better value for money than TV commercials? Professor Karen Nelson-Field, the executive director of the Centre For Amplified Intelligence, has an answer - and it's not good news for the tech giants.
Photo by Tim Mossholder on Unsplash
I presented some CPM work at AdWeek Sydney a couple of weeks ago and, let me tell you, I have been completely caught off guard by the reaction.
The work comes off the back of 18 months of large-scale research considering advertising effectiveness cross-platform and cross-device, incorporating 22 test and control groups, 2000 viewing hours and more than 50,000 test exposures.
So how is it that after all the controversial work we have presented - campaigning against the MRC’s viewability standard, highlighting how the poorer performance of Facebook and YouTube is compounded with time due to decay effects - that the humble CPM could cause such a stir?
But then I realised that proponents of online [advertising] are holding on to CPM as their last line of defence. Facebook and YouTube are perceived to be less expensive than TV and, as such, are theoretically considered good value for money.
This work shows they are not (caveat: this work was conducted in Australia, so at the moment it can’t be generalised, but the math can be applied cross country).
Let me explain…
To be thought of at the purchase occasion, advertising needs to cut through and be remembered. Without these two very important conditions, advertising can’t be effective. Our task over the past three years has been to examine how different platforms perform under these conditions and how this varies by device and country.
TV wins on impact
Findings from the 22 test and control groups have been resoundingly similar: TV impact is greater than its nearest rivals largely due to advertising viewability (pixels, coverage and time spent viewing). Take a look at the ThinkTV Benchmark Series comparing TV with Facebook and YouTube for more detail.
But what about CPM?
It doesn’t seem to matter how many times I prove this, one question keeps coming up. Online advertising is perceived to be cheaper, so in terms of ROI could it be better than, or at least cost comparative, to TV?
One of the great things about being an independent researcher is that we can answer some of our own research questions. And this was the perfect question to look at with the data we already had.
Our question: are the CPMs of poorer performing platforms low enough for the platform to be considered better value?
To answer this we first gathered average rate card costs (from four independent sources) for the test ads used in our Australian study. Then we considered the relative differences in CPMs between the platforms given their respective STAS (Short Term Advertising Strength) impacts.
Apart from STAS having a strong history as a gold standard measure of advertising impact, it inherently reflects the significant differences in viewability across platforms, as well as any other mediating characteristics that may have affected the choice to buy the advertised brand or not.
We calculated the uplift in baseline STAS divided by the CPM, providing a comparative measure of ROI for every $1 spent on each platform.
We found that every $1 spent on Facebook delivers 1.75 points of STAS uplift, compared with 1.3 for YouTube and 2.5 for TV. This means that, in Australia, of the three platforms, TV produces the best ROI for each advertising dollar spent.
Let me also answer your next question, about the fairness of CPMs. We wondered how much cheaper online ads must be to be cost comparative to TV. This question can be answered with basic algebra to discern the proportional difference in impact between platforms, indicating what the price difference should be.
Our findings show a Facebook impression needs to be one-third of the price of a TV impression (.34) and YouTube needs to be two-thirds of the price (.61) to generate a comparative ROI to TV.
Wait there’s more!
Cost comparative and value comparative aren’t necessarily the same thing. Even with cost parity you can still miss immediate sales opportunities from choosing a poorly performing platform. If CPMs for Facebook and YouTube were on par with those of TV for their relative return, the advertiser would still sacrifice significant sales per impression. In our Australian data the average is 50%.
And our recent Benchmark results show that advertising decay happens far more quickly on Facebook and YouTube. This means that ad impact is largely redundant after a week, at which time a new impression needs to be served to maintain any impact amongst the already exposed audience. As a result, the initial opportunity loss is compounded further with time.
Stop the short-term obsession
Advertisers need to start playing the long game and stop their obsession with upfront CPM when buying reach. Seeking low cost reach is like P. Diddy flicking money around.
While the digital landscape is shifting at a ridiculous pace, the question still remains: do you want your brand to grow in the long term and how much will it cost?