Clicks Are a Bad Metric (for influence or anything else)

Fast Company recently kicked off the Influence Project, an audacious effort to determine “Who are the most influential people online right now?” Unfortunately, they have made the same critical flaw in their measurement of influence that many first time and/or unsophisticated advertisers make — ie, focusing on clicks.

Under “How We Measure Influence” they clearly state their measurement philosophy:
The scale of your influence is based on…
1. The number of people who directly click on your unique URL link. This is the primary measure of your influence, pure and simple.
2. You will receive partial “credit” for subsequent clicks generated by those who sign-up as a result of your URL.

Essentially, this means, the primary form of measurement is clicks with a small component of the measure being people who sign up (or what we call “conversions”).

Here is why this is a problem. As of today, August 3rd, here is the #1 most influential person according to their measurement:

Yes, yours truly. However, if you go into rank view, you can see that I have only directly signed up 6 people. So, although I have sent a significant amount of click traffic I have driven very few conversions. In the online advertising world, this is equivalent of sending low value (or no value) traffic to an advertisers site and getting paid for it. This is a tactic we strongly advise advertisers to avoid, as a measurement of the post-click activity is far more important than the number of unique clicks.

Compare this to Mari Smith, an SEO consultant who has dominated the Influence Project rankings for most of the competition. She is currently ranked #2 in influence but has driven 346 people to sign up. Further downline from her initial signups (or conversions) she has driven thousands of additional signups as a result of her efforts.

In my opinion, there is no way you can argue that I am more influential than Mari as she is clearly driving more action from her referred visitors and more value to Fast Company’s efforts. In addition, some of the clicks I have sent I know are of zero value but there seems to be no measure of the value of an individual click.

In the ad world, businesses that pay on clicks quickly end up with 30-50% of their traffic being fraudulant. With no measure of click quality or value put on post-click performance, click traffic becomes the wild west. I would bet that Fast Company would see click volumes at 10-100x the current levels if they were paying even $.10 a click. According to some pundits, Fast Company should be paying per click given all the link love they are getting out of this controversial Influence Project.

I am considering testing the impact of sending high quality traffic to the Influence Project, rather than continuing my test of sending a stream of low to no value clicks, but to what end? Given that it is clear this project is focused nearly exclusively on clicks it is hard to imagine that any resulting rankings will be of any meaningful value. To argue that the “winner” is actually the most influential person on the web is simply a false argument.

The Power of Fail?

fail.jpg

FOX Business Live

Sand Hill Road Lessons - Techcrunch Guest Post

Today I wrote a guest post for TechCrunch called “Don’t “Pull A Patzer” And Other Lessons Learned On Our Trip Down Sand Hill Road.”

Here is the link: TechCrunch

The part of the article that seems to be generating the most amount of interest is the section titled the “Patzer Problem.” Essentially, this is the entrepreneur vs. VC problem of deciding when to sell a company, and highlights my surprise that VCs would actually say that they believe Mint.com’s sale was a mistake. I consider the company to be a huge success and this issue worth talking about.

In the comments, I note that…

I deeply share the concern “about any VC who says to you that any deal which puts tens of millions of dollars in a young founder’s pocket is something to avoid,” and that is why I wanted to shed light on the practice. I understand this phenomenon first hand, having been at Plaxo, a company that exited at $180M+ in equity value but had VCs saying it had failed behind closed doors.

My perspective is on this issue is that VCs are actually acting counter to their goals. They should tout these 3 year, $180M+ exits, and share the story with any budding entrepreneur who will listen — what better example is there of the incredible fruits of starting a successful company?

Furthermore, this exit in particular was at close to 40x revenue! It’s not like the company was a guaranteed success. Any rationale person would consider this a home run and I say congrats to Patzer and the entire Mint.com team.

POV - Point of Video

BrightRoll vs. Hulu — Why it Matters

My company, BrightRoll, recently announced that we are larger than Hulu in video viewership as measured by Quantcast. This is an important milestone for the company and for the online video advertising business.

Here is the raw data:

Here is why this is important:

1. We Only Serve Video Ads. Hulu Serves Video Content And Ads. Yes, we are comparing apples and oranges — if you compared our reach to the reach of Hulu’s ads, there would be a much larger difference between our two networks. Agencies and advertisers often forget that the reach of a site is irrelevant, the only metric that matters to them is the reach of the site’s available ad inventory. As the gap between our network and Hulu continues to grow, it will become more and more clear that the most efficient way to reach targeted video audiences at scale are through video advertising networks.

2. TV Everywhere? More like Video Ads Everywhere. Although there has been a lot of press recently about TV Everywhere, the reality is that online video advertising will be far larger than online television. Why? Because online video advertising is being used by premium publishers to monetize all free content — including broadcast video, short form video, games, radio, social apps — and many of those publishers have much more reach outside of their video content area than within it. Plus, many premium publishers don’t have huge production costs outside of their broadcast content, so advertisers are flocking to more cost efficient placements.

3. By The End of 2010, The Majority of the Top Ten Video Properties Will Be Networks. As I recently predicted in MediaPost, this is beginning of what will be a long trend of networks and aggregators surpassing the largest video properties in total reach. By the end of 2010, the majority of the top 10 video properties (as measured by Quantcast, comScore or your preferred third party) will be video companies that don’t produce any meaningful amount of video content. This means the top 10 properties will be dominated by video ad networks (BrightRoll), video-sharing sites (YouTube), video syndicators (Grab Networks) and vertical video sites (Break.com). Yes, some of these players produce some content, but the vast majority of the views on their properties are generated from content they did not produce.

We look forward to continuing to lead the industry and driving innovations across our platform, pricing, targeting and research initiatives. If you share our passion about video advertising, please reach out or join our team.

Is this good for your brand? (Part Two)

In 2007, I posed the question “Is buying video advertising in social media a good idea?” as a way to highlight publisher behavior that was hurting the video advertising industry.

Today, I pose the question “Is buying video advertising from a video syndicator a good idea?”  The answer is, it depends.  But, again, bad actors are hurting the video ad business. Broadcast videos sites such as Hulu and CBS have done a phenomenol job of maintaining high quality users expereinces when their videos are in syndication.  Hulu distributes their player to MSNFancast and other sites, and maintains a universal user experience.  CBS, through the audience network, controls video content, placement and advertising, to insure consistency.

The problems for brands become more evident when we are talking about the content licensors such as RooNewsMarketJamboTVa slew of others or any of the video ad networks that offer video syndication.  In an attempt to compete with the broadcasters (or each other), these publishers syndicate their players far and wide, with little attention paid to user experience, site or site placement.  As a result, video advertising campaigns can end up on low quality sites or, worse, not even be seen by the consumer.

This problem is most evident when brands unknowingly buy fake pre-roll, either from a video syndicator or a video ad network. Fake pre-roll occurs when an advertiser buys pre-roll video inventory, but gets a placement where the video player autostarts a video ad in display inventory, usually with the sound off and often below the fold.

Examples of fake pre-roll are abundant (see screenshots below).  In 15 minutes one day I found examples of MarsState FarmNicoretteVan de KampsStarburst and Starkist running in low quality, fake preroll placements.  On day two HellmansJelloTideOil of Olay and Bounty.  Sites such as Luvcube.comSweetiessweeps.comHairpedia.com and Pajiba.com provide clear examples of fake pre-roll with video syndication from Jambo and video ads from many “leading” video ad networks.  This gaming of video advertiser’s budgets is hurting all of us for the following reasons:

  • These advertisements won’t perform for the advertiser, which hurts budgets moving online and the perceived efficacy of online video advertising
  • These placements are priced at banner rates (or lower), which drives down rates for the entire category and falsely gives agencies the belief of a market rate
  • Advertisers run the risk of being placed next to inappropriate content, which could result in an advertiser or agency pulling out of video altogether

The problem of fake preroll can either be solved by agencies (by paying more attention to the placements they are buying, comparing ad performance among players and demanding URL by URL reporting) or by the players in the industry (by not taking advantage of unknowing buyers, “gaming” the medium and returning to focus on selling a high quality product).Personally, I am much more a fan of self-regulation. 

vandekamps_autostart.jpg     starkist_autostart.jpg     luvcube_belowfold.jpg     target_autostart.jpg vlasic_belowfold.jpg  
 
 

How To Create A Vanity URL Shortener for Twitter

Since my name is Tod Sacerdoti and I twitter with short URLs such as http://tod.ly/aboat, I have received a bunch of requests about how I created my own vanity URL shortener.

Get it? Tod + bit.ly = tod.ly?  

It is actually quite straightforward.  To do so, follow these easy steps:

  1.  Register a (short) Vanity URL 
    • Make sure it is short - that’s the whole point!
    • Either a simple tod.com at GoDaddy or 
    • A foreign domain such as tod.ly from LibyanSpider
  2. Redirect your short Vanity URL to bit.ly (or similar url shortener)

After following the two steps above, you have to wait anywhere from 1-9 hours for your new DNS records to propogate.  Then, shorten a URL at Bit.ly and insert your new URL.  For example, I created http://bit.ly/aboat and changed it to http://tod.ly/about 

That’s it, your done!

Congrats, now you are truly vain.  :-) 

How To Spot “Fake Pre-Roll”

Yesterday I wrote a post on AdAge on “How to Spot Fake Pre-Roll,” as we have noticed a trend of online publishers and networks packaging low quality ad units as pre-roll. The article goes into detail about what to look out for and what questions you should ask to verify your pre-roll buys.

Here are the highlights:

Pre-Roll Quality Categories
1. Gold Standard: Traditional Pre-Roll, as defined above
2. High Quality: Video ads that play in the middle of long form content (such as an ad between segments of House on Hulu.com)
3. Medium Quality: Video ads that auto-start with sound in a publisher’s video section; no user initiation
4. Low Quality: Video ads that auto-start with sound on a publisher’s home page; no user initiation or attention
5. Questionable Quality: Video ads that auto-start without sound in display inventory (typically by an ad network without video technology)
6. Borderline Fraudulent: Video ads that auto-start without sound on a publisher’s video player which can be embedded by users anywhere

Key Queries to insure 100% Pre-Roll:
1. Will my video ads ever be played in the middle of content, as opposed to before content?
2. Will my video ads ever be served into an environment where video is not the main content on the page?
3. Will my video ads ever be auto-started, i.e. started without a user initiation?
4. Will my video ads ever be started with the sound off?
5. Will my video ads ever be served into display inventory?
6. Will my video ads be served into any syndicated content? (Syndication almost always means a reduction in quality.)
7. Can you provide me with a list of every URL my ads will appear on?

Video Ad Sausage: Pre-Roll Needs To Be Defined??

Call me crazy, but I thought the definitions of pre-roll and companion banners were well understood.

Unfortunately, many vendors in the marketplace have begun to offer a menu of what I call video ad sausage - a mix-mash of low quality ad units being packaged as pre-roll.

The simplest example of video ad sausage is running video and a companion banner jointly withing a 300×250 display unit and calling it pre-roll. In order to get to the bottom of this, let’s start with the proper defintion of Pre-roll:

    Pre-roll - A video ad that plays before the start of a video that a user has selected to play shown prior to the user’s selected content starting to play.

    Companion Banner - A banner ad that is displayed simultaneously with a video ad, often remaining on the web page after the video ad is complete.

There are a few important ingredients of pre-roll that media buyers must understand in order to avoid getting a plate of video ad sausage.

  • User Initiated - if the units loads automatically, without a user selection, it is not pre-roll
  • User Selected Content - if the user has no idea what video content is behind the ad, it is not pre-roll
  • Companion Banner - if the companion banner is served within the same unit as the video ad, it is not a companion banner

To be more explicit, running auto-start, sound off video ads in 300×250 display inventory is clearly a violation of the definition of pre-roll. Even worse, is running the companion banner also in the same 300×250, thereby not only reducing the value of the pre-roll but also double counting impressions. These tactics are a clear disservice to both agencies and their clients.

To learn more about true pre-roll, talk to your local BrightRoll contact.