dp thoughts.png

Ideas. Insights. Inspiration.

ADdicted: That Coinbase Ad

If you arrived at this page by scanning a QR code that appeared in my newsletter or on my LinkedIn feed, I'd like to thank you in advance for helping me with a simple experiment.


The experiment: what would you do if your curiosity was piqued by a mysterious QR code?


You didn't know where the QR code would take you, or if it would be relevant to you at all.


But you trusted me enough to know that it wasn't going to be malicious.


And you were curious.


So you scanned it and landed here.


That pretty much explains what Coinbase did during Super Bowl LVI last weekend...


On Super Bowl Sunday, the cryptocurrency company bought 60 seconds of airtime (at a cost of $14 million USD), and showed the 117 million people watching the football game this ad:

Now, this was clearly not your typical Super Bowl ad.


There were no major celebrities... no people at all, in fact. No mascots. No silly humour, epic montages, or big musical scores. No narrative storytelling of any kind.


Just a colour-changing QR code for 56 seconds, followed by four seconds of a bright blue screen that read, "COINBASE. PAID FOR BY COINBASE (C) 2022 | COINBASE, INC."


But before that blue screen appeared, millions of people watching that not-at-all-typical Super Bowl ad collectively thought, "I wonder where that QR code will take you?"


And 20 million of those people were curious enough to pick up their phones and scan that QR code.*


Now here's some quick marketing math: if you show something to 117 million and get 20 million of them to act, you've achieved a 17% response rate. That's extremely impressive!


Except simply scanning the QR code isn't the final response Coinbase wanted to achieve.


You see, the 20 million people who scanned that QR code were taken to this page...

(Note: the offer wasn't expired when the commercial went live)


The indisputable win for Coinbase would have been had all 20 million people clicked that bright blue button and opened an account.


And it's HIGHLY unlikely that happened.


In the marketing community, there's been a lot of debate this week as to whether that Coinbase ad was the best Super Bowl ad ever... or the absolute worst.


You're free to have whatever opinion you like. But I like to deal in facts whenever I can, so here are seven important facts as we collectively understand them:


1. Coinbase wasn't the first company to use a moving, mysterious QR code in a television spot: Burger King did this in early 2020. Scanning Burger King's QR code in real-time didn't get you $5 in bitcoin, but it did earn you something arguably better: a free Whopper!


Coinbase deserves full credit for being the first company bold enough to do this during a Super Bowl... but they don't get points for originality.


2. The cost of this commercial would have been about $17 million USD. How does that break down? The airtime reportedly cost $14 million. That doesn't include the production for the spot (which is likely very low because, well, look at the ad)... so let's add another $500,000 for that. Also, it doesn't include the cost of the $5 bitcoin incentive that was offered to those who opened accounts; Milkroad estimated the ad earned Coinbase 500,000 new customers, so that would be an additional $2.5 million expense. So if these numbers are accurate, we arrive at a $17 million cost for the ad. That's not a lot of money for a company with a market capitalization of almost $45 billion (as of today)... but it's not exactly chump change, either.


3. Not all 117 million people watching the Super Bowl would be Coinbase's target customer. The reason you buy Super Bowl ads (at extraordinary costs) is to achieve massive reach, not narrow targeting... so the ad was wasted on a lot of people. That's not an opinion, it's math.


This isn't unique to Coinbase, though: not everybody who sees a Budweiser, Rocket Mortgage, or Pepsi ad during the Super Bowl ad is going to be the target customer for those products either. But this has to be said for all those "performance marketers" yelling about how all Super Bowl ads should be this measurable: not everybody watching (and not even everybody clicking) is equally valuable to Coinbase.


4. The landing page did not tell visitors who Coinbase was and what they did: it assumed everybody already knew. But as #3 above has already suggested, that's likely a bad assumption: many of the people scanning that QR code wouldn't know Coinbase... and may not even be familiar with "bitcoin". (And... what's WAGMI?!?)


(This fact is especially curious because Coinbase's regular homepage does a nice job of succinctly describing the company's reason for being: they're "the easiest place to buy and sell cryptocurrency." You just wouldn't know that if you only saw the landing page.)


To be fair to Coinbase, I'm going to assume they knew this would happen. But it underscores the fact that the 117 million people who saw that ad and the 20 million people who scanned that QR code aren't the right metrics to use if we want to measure success.


5. There was indisputably a percentage of people who scanned that QR code JUST out of curiosity, arrived at the landing page, and... did absolutely nothing. There are definitely people on this Earth who know exactly how many people signed up for accounts after seeing that ad (or who clicked the less valuable white button simply to enter the contest), but all of them work at Coinbase and I doubt any of them will share that information publicly.


6. When 20 million people tried to access the Coinbase landing page within 60 seconds, the website crashed. That traffic was more than six times higher than any previous load that Coinbase had ever experienced, and it overwhelmed the site. And no, friends, that's not a positive story that underscores how popular this stunt proved to be: it shows they weren't prepared enough to anticipate overwhelming success.


Put another way, how many of those 117 million people who WERE the Coinbase target customer couldn't access the site to sign-up for an account because non-qualified customers were visiting the page out of sheer curiosity?


7. A lot of people downloaded the Coinbase app after watching this ad. After the Super Bowl spot aired, Coinbase shot up the list of "Most Downloaded Apps in the App Store", going from #186 to #2. This means the ad absolutely, unquestionably had a positive impact on Coinbase, at least in that regard.


And those are the facts.


Now here's an opinion: the Coinbase ad did not tickle my creative fancy...


... but that doesn't matter at all.


As I wrote earlier this week, "what matters is if the commercial achieved its desired result with the brand's intended audience, as outlined in the brief."


It's reasonable to assume the desired result for Coinbase was "getting as many new customers as we can."


But only the people at Coinbase would be able to provide us with those numbers, which means they're the only ones who can really determine whether or not this ad "worked".


And I doubt those numbers will be made public any time soon... if ever.


In the best-case scenario, Coinbase earned not only broader name recognition and widespread attention for their marketing stunt (just as Reddit did with their 5-second-ad during last year's Super Bowl), but also more than enough new customers to achieve a positive return on their marketing investment.


In the worst-case scenario, they didn't... but marketers will assume they did, believe the lesson is to abandon creative storytelling entirely, and attempt to repeat this same stunt with another client during next year's Big Game.


Let's hope the best-case scenario applies.


 

* It's important to note that while scanning a random QR code would generally be ill-advised (in fact, PCMag proclaimed, "Coinbase's Mystery QR Code Super Bowl Ad Is a Security Nightmare") this wasn't a random QR code: it was a Super Bowl ad. And it would be reasonable to think that Super Bowl broadcaster NBC wouldn't want to be liable for allowing something malicious to enter our homes via their broadcast, and thus would have pre-vetted that code before allowing the commercial to air. Without that basic level of trust, I can pretty much guarantee the click rate would not have been nearly as high as it was.




If you liked this post, don't miss the next one: get dpThoughts delivered to your inbox up to three times each week. 

(Or add me to your RSS feed and get every post in your reader as soon as it's published.)

Disclosure: As an Amazon Affiliate and a member of select other referral programs, I may earn a commission if you click on links found within my blog posts and subsequently make a purchase. The commissions earned are negligible, and while they help fund this website, they do not influence my opinions in any way.