Central Control, Inc. retweeted
The only real ROAS is experiment-based, incremental ROAS.
Accept no quasi-experimental imitations.
Central Control, Inc. retweeted
Thanks to @MarcJGoldberg for including me in his Trust Web Interview Series! I wax on about advertising experiments (natch), the @ResearchWonks, and visions of the days when we're back to airports and concert halls!
trustwebtimes.com/trust-web-…
Central Control, Inc. retweeted
Announcing "Measured Response" our new #podcast on #advertising, #marketing & #media #measurement, #research, #analytics & #datascience
Ep 1: @rickbruner @SKTornquist w @aripap @BeeswaxIO re "dead pool", NYC startups & more + @MarcJGoldberg @method_mi
youtube.com/watch?v=UJBXkQKs…
Central Control, Inc. retweeted
Happy to be an expert source in @eMarketer's new report "Ad Measurement and Revenue Attribution 2021." I wish it went deeper on experiments (of course), but these two charts show just what a burning priority accurate measurement is for advertisers.
emarketer.com/content/ad-mea…
Central Control, Inc. retweeted
Good New Year's resolution from @ThinkwithGoogle
One question: why no "ghost ads" for paid search?
Also, @avinash, I recommend you have a look at @the_ARF's "RCT-21" study. I'll shoot you the methodology doc, if you're interested.
thearf.org/submit/arf-rct21/
This Post is from an account that no longer exists. Learn more
Central Control, Inc. retweeted
"Of primary concern to advertisers is the potential cannibalization of free organic clicks by paid search clicks... Cost per click does not provide a
complete picture... A more useful metric is the cost per incremental click."
Google researchers, 2011
static.googleusercontent.com…
Central Control, Inc. retweeted
Facebook Private Computation Framework for RCT: so intuitive!
"Cryptographic methods help two parties, Alice & Bob, compute a function on each of their secret inputs and receive outputs...a garbled circuit-based 2pc program."
github.com/facebookresearch/…
And nice privacy example.
Central Control, Inc. retweeted
From "Great lessons in 'Correlation is not Causation'," I came across this one over the weekend. No, Lefties do not die younger than Righties, at least not as described by this classic case study in bad research.
snopes.com/fact-check/left-h…
Central Control, Inc. retweeted
My expanded comments on the @Freakonomics two-part episode, "Does Advertising Really Work?"
Reminds me of: "How many psychologists does it take to screw in a lightbulb?" A: "One. But it will take a long time, and the lightbulb must want to change."
linkedin.com/pulse/does-adve…
Central Control, Inc. retweeted
I liked the reference to the "Principal-agent Problem," underscoring that everyone in the industry other than the brand (the CFO really) has a vested interest in supporting the flawed ROI measurement of the status quo.
en.wikipedia.org/wiki/Princi…
Central Control, Inc. retweeted
I just finished listening to the recent two-part @Freakonomics podcast, "Does Advertising Actually Work?" In answer to that provocative question, let me say I would have given my left nut to run an ad for @_CentralControl on those episodes.
freakonomics.com/podcast/adv…
Central Control, Inc. retweeted
"While business experimentation is — rightfully — framed as a gold standard by scholars and leading practitioners, [...] Only 12.6% of the 6,777 companies we looked at had conducted a recent RCT"
hbr.org/2020/10/marketers-un… by @julian_runge
Central Control, Inc. retweeted
It almost argues for validation of self-reported lift by the big media companies, using some independent experiment mechanism...
thearf.org/submit/arf-rct21/
Oof! Good reminder that experimentation platforms need constant experimental validation and need to be as robust as possible to code changes elsewhere in the org.
Big picture though: experimentation platforms are super important! adexchanger.com/platforms/fa… @AdExchanger @OSchiffey
Central Control, Inc. retweeted
Replying to @billharveymedia
@BillHarveyMedia and @RickBruner explain how companies are changing marketing models to stay afloat amid waves of #Covid, how @CocaCola became first to shore, the medical mindset marketers may need to adopt and provide insight on @The_ARF’s RCT21 Program.
mediavillage.com/article/Coc…
Central Control, Inc. retweeted
From the new @the_ARF member newsletter:
"And our new RCT-21 partnership initiative to advance Cross-Platform ROI analysis through application of Randomized Control Trial measurement is game changing."
@_CentralControl @billharveymedia @605Data
thearf.org/submit/arf-rct21/
Central Control, Inc. retweeted
Are you a Wonk?
If you work in media/market research, you should be.
(If you don't, you shouldn't be.)
ResearchWonks.com
Central Control, Inc. retweeted
A quarter century after P&G gave us click-based ROI, a $100BN+ US digital ad industry is based too much on magical thinking.
Here's my modest proposal: P&G could give the industry a make-good.
Marc Pritchard talks plenty about the bad ROI metrics. I've got just the project.
Central Control, Inc. retweeted
Trivia question for Advertising Twitter. Prize is a copy of Central Control's 100+ page "Best Practices Guide to Advertising Incrementality": centralcontrol.com/best-prac…
Question is:
Who invented cost-per-click pricing, the beginning of the ruination of digital advertising?
Central Control, Inc. retweeted
Good ad-industry read.
Notable: Coca-Cola is “reassessing our overall marketing return on investment on everything from ad viewership across traditional media to improving effectiveness in digital,” said James Quincey, CEO.
Bring it. #Incrementality
nytimes.com/2020/07/28/busin…
Central Control, Inc. retweeted
Our latest research initiative RCT21 in conjunction with 605, Central Control and Bill Harvey Consulting, was recently covered in Forbes ow.ly/xH1u50APAqw #the_ARF #research #advertising #billharvey #605 #Forbes #CentralControl #advertisingresearch #crossplatform #ROI











