Commentary
The Most Important Media-Mix Study, So Far
- by Joe Mandese @mp_joemandese, March 28, 2022
Bill Harvey is one of the smartest guys in the business. He is also a great writer. So when he sent me an email to an article he wrote last week, I definitely took notice and read it all the way
through.
I mean, how could you resist the headline?: …
7 comments about "The Most Important Media-Mix Study, So Far".
Read more comments >
Next story loading 
Joe Mandese is the Editor in Chief of MediaPost. You can reach Joe at joe@mediapost.com.
My co-worker friend for many years Beth Uyenco sometimes used to ask new media hires to estimate the prevalance of various groups within the U.S., such as urban/rural, college grads and so on. The result was great consistence in how these young people overestimated the proportion of all the groups that they themselves were members of. That's consistent with Harvey's observation of college kids inflating the role of social media.
On the larger question of media mixing, Harvey is spot on again. While balancing media choices is crucial, it's always the case that mixing produces benefits. There were spectacular store studies decades ago documentating the sales bump from adding radio to a retail client's ad schedule.
Hi Joe.
It's not often I see the Middletown Media Studies cited these days!
Your comments about unmeasured media are spot on. By way of illustration, I recall when we conducted the first of the studies in 2003, many people questioned why we chose to include use of the cell phone in what we measured - they didn't consider it a medium.
That all changed in less than a year of course, but it shows how you don't get to understand the whole picture unless you measure it.
Mike
Middletown and the follow up CRE studies, led by Mike and his Ball State group were incredibly useful in their time. I'm sure another would prove equally useful every ten years or so. Such collaborative efforts are sadly not likely anymore without finding from measurement suppliers.
Great post Joe.
You are spot on with your comments such as "I think the actual media universe is much greater than the one the ad industry typically looks at", and "they tend to measure the effects only of the media that they know exist and can actually measure", and "there is a vast dimension of media exposure the industry doesn't accurately measure, and may not know actually exists."
But they made me think of another pretty smart bloke, albeit not in the media sector. who has been attributed to have written on a blackboard "'Not everything that can be counted counts, and not everything that counts can be counted." Whether Eisntein did or didn't say that. it is still very sage advice. As a parallel, we only know a scintilla about the universe, but what we do know is mind-blowing and enough for us to put a man on the moon just over 50 years ago.
And then there is the delicious irony of the reference to the "walled garden" of the digital age. We may know a lot about what is within a walled garden (albeit self-generated) but it is still less than what we know about the much larger "media market" with its omissions, incompleteness and varying accuracy.
As Erwin Ephron espoused years ago regarding the value of increasing the mix of media, it's about target audience reach, preferably exposed i.e., Eyes or Ears-On. Frequency, as he opined in one of his famous newsletters, is "crab grass". CPM, which is driven by frequency, stands for "Completely Positively Mad", of course!
In addition, I would posit that ROI studies are campaign rather than media studies (ANA please take note regarding hyour Cross Media Measurment initiative) due to the overwhelming influnce of the creative to drive attention and therefore impact plus, the interest in the brand category and the equity of the brand being advertised going in. However, all things being equal, incremental target group Eyes/Ears-On reach will typically improve campaign outcomes based on the "right" message.
Please remember John Grono's warning in all this. Brilliant creative can overcome a bad media plan but not vice versa!!
While reach is obviously important it matters a lot whether we are considering the very beginning of a branding ad campaign---new product or new positioning for an old product---or whether we are in the second year of the campaign ---or the third year---by which time just about everyone who is going to respond to your message will have seen your ads many times and gotten your message..
Erwin was promoting the maximize one -week reach aspect of his "recency" theory. This called for a 70% reach week after week for the entire budget year. If executed---very difficult now---this would automatically have become a high frequency campaign by week two or three. It would have taken about 200-220 TV GRPs in 1995 to attain a one- week reach of 70%----- by the time you finished your second week you would have amasssed 400-440 GRPs with a reach of around 85-90%---remember , those were the good old days for TV. This meant that your average frequency in two weeks was about 5. Take the entire year---all 52 weeks--- and you had about 11,000 GRPs which meant That your yearly frequency was roughly 115. That's a lot of "crabgrass".
My point is that you can't avoid the "crabgrass" as it's part of the reach and frequency equation. Moreover, frequency is usually required to make a sale and later, to reinforce the consumer'spositive feelings aroused by earlier exposures. It's not something to be avoided, rather, it's something to try to orchestrate so you don't overdo it in any short period---like a day or a week---or bunch up your ads in the same episode of a program.
As for which medium comes first, which second, which third, etc. and how much to spend in each medium there is no way to develop a magic formula that gives you an answer which can be applied to all ad campaigns.
Thanks for putting into words what I think whenever I read about "the right media mix" including a larger investment in the type of media favored by whoever funded the research.