superbowl
Popular Posts – Week Ending March 19, 2010.
HP Mini 311 Nvidia ION Netbook Hackintosh’ed
Facebook advertising metrics and benchmarks
What is Web 3.0? Characteristics of Web 3.0
social media benchmarks
Samsung 52 inch HDTV $9.99 at BestBuy – purchase receipt below (6:21a eastern time August 12, 2009)
1024-bit RSA encryption cracked by carefully starving CPU of electricity
How to manufacture a viral video sensation and make viral profits – Post 2 of 2
What can search volume tell you?
Evian baby viral video has much higher ROI than Etrade baby superbowl ad
The Evian baby viral ad (red spike) got almost as much search volume as eTrade’s Superbowl ad of 2009 (blue spike). But Evian paid millions less by skipping the expense of airing the video on traditional media; instead they just posted it to YouTube for free. But notice that in both cases the effect was ephemeral (not long lasting) — notice the narrowness of the spike. Interest in the viral video also subsided quickly. But at least Evian didn’t waste millions on producing and airing it — thus achieving a massively larger ROI than Etrade who paid to make the ads and then air it at great expense on the Superbowl for the last 3 years.
Etrade Baby Ad
Evian Baby Viral Video
Superbowl 44 Ads That Made It
Sadly only 2 made it on Google’s Hot Trends today (Day 1) after Superbowl 44. We may hit ZERO on Day 2.
Last year, by Day 3, the advertisers who paid for Superbowl ads dropped off the Hot Trends list.
See The Ephemerality of Superbowl Halo http://bit.ly/bUZJb6
Yep, like I said, by Day 2 (Feb 9) the 2 that were on dropped off. But Denny’s made the top 20 …
The Grand Unified Theory of Marketing(tm) – Digital String Theory
Just as physicists and mathematicians have been searching for the grand unified theory of the universe, I have been looking for a way to tie together the disparate disciplines of marketing and advertising, a way to correlate metrics from different industries that interrelate with marketing (e.g. market research, Nielsen, etc.), a way to put all past theories in context and perspective (Michael Porter’s Five Forces, Net Promoter, etc.), and a way to explain marketing successes and failures — all in one.
My method is the scientific method – which is simply put doing experiments and making observations that either support or refute hypotheses.
A grand unified theory will also need to be able to take into account phenomena such as social networks, etc. What are the organizing principles of such; what is the value? Why now?
Using digital tools — such as search volume trends — we can start to correlate marketing spend effectiveness across different forms of media and also different advertising and marketing techniques. The example below compares eTrade and @Drobo. What is most embarrassing is that eTrade, a well known brand from the first dot-com heyday, spent lots of money creating and airing TV ads which it hoped would go viral. They even paid for Superbowl ads for the last 2 years to promote the “eTrade talking babies” as you see from the 2 spikes in search volume during February of 2008 and 2009. However, when compared to Drobo (a startup company that developed a very easily upgradeable back up hard drive array), it is shocking to note that Drobo spent NOTHING on advertising and relied entirely on word of mouth and an awesome product. And their search volume is not only larger than eTrade but also sustainably larger despite zero advertising and media cost. The “totals” even suggest that the volume under the curve of Drobo is 8X (EIGHT TIMES) that of eTrade.
So if you consider that eTrade spent millions of dollars to create the TV ads and even more millions of dollars to air them on TV in order to drive interest, demand, and hopefully new customers, then Drobo can be considered to have gotten the equivalent of 8X more dollars in advertising and media – for FREE using techniques and channels other than TV advertising. So what does that say about the relative value of TV advertising compared to these other, newer techniques?
godaddy vs megan fox
What viral videos look like; what non-viral videos look like — by the stats
The first 2 are viral videos – notice the shape of the “total views curve” (quick rise and approaches the max asymptotically). The last 2 videos are not viral, and supported by paid advertising and promotion. It is a straight line that grows steadily over time. The 2 examples of non-viral videos were chosen simply to have similar view counts as the first and second examples.
Viral video examples – notice the asymptotic curve towards the max on the total views chart.
Frozen Grand Central ImprovEverywhere viral video – 18 million views – added on Jan 31, 2009. ”Other/viral” gave it its first big boost and embedded views gave it another big push.
No Pants Subway Ride ImprovEverywhere viral video – 9 million views – uploaded January 13, 2009; got onto YouTube homepage and got a major boost from it.
NON-viral video examples – notice the straight line of the total views chart.
Perfect example of NON-viral video that had help with paid media – in this case, GoDaddy supported these videos with costly Superbowl ads — which led to nice bumps-up in total views.
In the case of Smirnoff’s Tea Partay, it was not supported by paid media so it took longer to grow and the shape of the curve is a nice blend between the straight line of a non-viral video and the asymptotic line of a viral video.
Finally, blatant ads don’t go viral – Sony’s grand central product demo stunt. And even if they are discussed in dozens of blogs it is not enough to get past the first tipping point.
How Megan Fox got the equivalent of $50 million of Superbowl ad spending – for FREE! — the marketing of Megan Fox
godaddy superbowl ad spending led to sharp spikes in search volume every February for the last 5 years straight. Other advertisers who spent on Superbowl ads have similar lift in search volume from the TV advertising.
Source: Google Insights for Search
If you believe that lift in search volume indicates interest and intent and if you consider that each 30-second ad cost $3 million in 2009 (WSJ: NBC Super Bowl Ads to Cost $3 Million) and assuming GoDaddy’s ad did not air more than once, they spent $3 million to get their ad in front of a TON of people and to get people’s attention. Those people who saw the ad and were interested enough to take action went online and searched for more information by typing godaddy into search (see lift in search volume during February of each year) .
If we assume that it took $3 million to generate a certain lift in search we can use multiples to calculate the media dollar equivalent of any lift in search — for example, if godaddy spent $3 million to get X lift in search, then a 2X lift in search would have required $6 million of media (in a very very simplified back of the envelope estimate; it usually would cost more than 2x to get that lift) — i.e. it would have cost at least $6 million in superbowl ad media dollars to achieve a 2X lift in search volume.
So, if we now compare search volume on megan fox side by side with godaddy search volume, we will see that in Feb 2009 Megan Fox was indexing at 21 while godaddy was indexing at 12 (this is normalized to a scale of 0 – 100). So search volume on megan fox indicates she was getting the equivalent value to $6 million of super bowl media ad spend – FOR FREE — roughly 2X the search volume of godaddy in the same time period.
At the peak of her search volume in June 2009 (corresponding to the release of Transformers 2: The Revenge of the Fallen), she was indexing at 100 and godaddy at 7. This is 8x the index of godaddy of 12 during the Feb 2009 time period when they were airing their superbowl ads. This implies that she was getting the search volume that would have required the equivalent to a $24 million super bowl ad spend to achieve — again for FREE!
If you want to research futher, use the following link to bring up Google Insights for Search to see relative search volume
http://www.google.com/insights/search/#q=godaddy%2Cmegan%20fox&cmpt=q
In February 2008, Megan Fox indexed at 8 and GoDaddy at 8. In 2008, Superbowl ad spots cost only $2.7 million — so she had the equivalent search volume as a paid advertising spending $2.7 million on a Superbowl ad.
In 2007, Godaddy indexed at 6 during Feb 2007 Superbowl. Megan Fox indexed at 43 during the July release of the first Transformers movie — this is an 8X multiple on Superbowl ads that cost $2.6 million — or $21 million
So the perfect “product placement” of Megan Fox in the two Transformers movies garnered her nearly $50 million worth of advertising based on search volume equivalency. This does not even take into account her sustained and increasing search volume, compared to most advertisers’ search volumes which drop right back down to pre-ad levels once the ad is finished airing.
Lift in search volume indicates that customers …
… saw the ad, recalled it, and thought it relevant enough at the time to make the effort to take action — search for more information about it (beyond the tidbit of info contained in the 30 second ad, print ad, radio spot, or banner ad).
But traditional ads are still very very costly and inefficient due to the extremely large media cost.
For example, at the extreme cost of a Superbowl ad, the following advertisers were able to drive fleeting (short-lived) lift in search volume: godaddy, etrade, sobe lifewater, dennys.
related: http://go-digital.net/blog/2009/03/lift-in-search-due-to-paid-tv-advertising/
lift in search due to paid TV advertising
List of 2009 Superbowl spots on AdAge.com
http://adage.com/superbowl09/article?article_id=134136
Lift in search is a great indicator of interest. Modern consumers may be inspired by TV ads, but they usually go online to do more research for themselves, to inform their own purchase decision. The following examples show the lift in search after Superbowl commercials or for launch of products like Subway Footlongs. The use of unique, made-up words makes it easier to detect lift in search (see related post: made up words are great for tracking buzz and search volume ). There is now a correlation between offline paid advertising and online behaviors of modern consumers that can be tracked and ultimately related to sales.
What is harder to do is track lift in search from smaller TV media buys or from terms which are generic — e.g. American Express OPEN, Proctor & Gamble’s TAG (men’s deoorant), etc. And furthermore, people may or may not remember the brand name itself and may type in a more general search query — e.g. “talking baby” instead of” e-Trade” or “dancing lizards” instead of “SoBe LifeWater.” And most people usually forget to type in special URLs specified in the ads. So the opportunity is to 1) use made-up words which can be used to detect lift in search and 2) search-optimize around other more generic terms that people may search for if they remembered the ad, but did not remember the brand name itself.
key learnings include:
1. only the superbowl TV ads generates enough awareness to drive lift in search volume detectable above the noise or normal levels
2. made up words are useful in correlating paid advertising and subsequent online actions (e.g. search) because most users forget or are too lazy to type special URLs
3. is is always better to have real analytics from the site to see when paid campaigns hit; site analytics will also reveal more information about users including demographic information, what they are looking for, and even whether they “convert” to a sale or a desired action — like print off a coupon, etc.
Notice the January spikes for several of the examples below — these are their Superbowl ads in action. But also notice how sharp the spikes are — most of them go back to prior levels within 1 – 3 days (see related post: the ephemerality of the Superbowl halo )
Source: Google Insights for Search
Denny’s is still getting “social amplification”
denny’s spike in search volume after Superbowl ads, lasted only a few days, then back to normal
compared to very steady stream for years
About Me
Tags
advertising
amp
android
apple
brand
com
company
content
day
email
Engadget
example
Facebook
gawker
google
information
iPhone
lot
market
marketing
mdash
microsoft
nbsp
number
online
page
Permalink
post
Product
search
search volume
site
something
Source
time
tmpPost
today
twitter
use
user
video
volume
way
Web
year
Popular Posts
- HP Mini 311 Nvidia ION Netbook Hackintosh'ed
- Facebook advertising metrics and benchmarks
- When NOT to use Groupon (as an advertiser)
- How-To View Gmail for iPad on Your Regular Computer - Chrome and Safari
- social media benchmarks
- What is Web 3.0? Characteristics of Web 3.0
- Facebook's Security Check Asks Users to Identify Photos of Friends' Dogs, Gummi Bears
- Vapor4 May Be the First Bumper Worthy of the iPhone 4
- Two Social Success Stories - Groupon and FourSquare
Recent Posts
- 1535
- ‘we are prioritizing our Android platform’
- 1531
- 1529
- 1527
- HP Labs teams up with Hynix to manufacture memristors, plans assault on flash memory in 2013
- Amazon planning subscription video service to challenge Netflix and Hulu?
- It’s Time To Make Standardized Ratings For Gadgets
- Arcade Fire and Google Pushing HTML5 Together
- New ARM architecture (likely Eagle) better suited for OS virtualization
Recent Articles by Dr. Augustine Fou
- Augustine Fou | ClickZ
- ClickZ Welcomes Augustine Fou | ClickZ
- The ROI for Social Media Is Zero | ClickZ
- A New Definition of 'Digital' | ClickZ
- Social Commerce: In Friends We Trust | ClickZ
- 10 Commandments of Modern Marketing | ClickZ
- Digital is the DNA of All Advertising | ClickZ
- Experiential Marketing | ClickZ
- Social Intensity: A New Measure for Campaign Success? | ClickZ
- Beyond Targeting in the Age of the Modern Consumer | ClickZ
Pages
Archives
- September 2010 (7)
- August 2010 (101)
- July 2010 (61)
- June 2010 (28)
- May 2010 (28)
- April 2010 (26)
- March 2010 (33)
- February 2010 (21)
- January 2010 (12)
- December 2009 (4)
- November 2009 (2)
- October 2009 (14)
- September 2009 (6)
- August 2009 (19)
- July 2009 (34)
- June 2009 (11)
- May 2009 (4)
- April 2009 (6)
- March 2009 (13)
- February 2009 (32)
- January 2009 (25)
- December 2008 (1)
- October 2008 (1)
- November 2007 (1)
