The Official Blog of Cogmap, the Org Chart Wiki


Archive for August, 2007


Online Video and Contextual Advertising

Monday, August 27th, 2007

Jeremy Liew has a nice post on whether contextual advertising will work in online video.  He notes the absence of contextual targeting for YouTube video ads and wonders whether direct response will work.  I think there are a couple of things he doesn’t mention that are worth noting:

  • I think everyone thinks “direct response advertising online” is a code word for “lower eCPM”.  This is probably true.  One huge difference between monetizing search results today, AdSense, and monetizing video with in-video ads is that in-video ads decrease your user experience.  Overlays are clearly an attempt to avoid this and be more like the “associated results” that you get with monetizing search results.  As in-video advertising impacts user experiences, pricing can only go so low.  Overlays probably are less constrained in this area.
  • Given the volume of videos on YouTube, unless they establish a model that serves the needs of direct response marketers, they will probably not be able fully monetize their inventory
  • Obviously you have to offer a CPC or CPA model to attract direct response advertisers.
  • Last and most importantly, Google is great at contextual targeting, but not for video.  Dissecting video is completely different from reading text.  Call me crazy, you will see companies like ScanScout get gobbled up for big money as advertisers seek more sophisticated targeting and companies like Google look for higher eRPMs on advertisements

Finally, I promised my buddy Jon Alsop that I would give him some super duper link love for his new blog (alas a blogspot user so no trackbacks).  Jeremy Gold has a blog as well.  None of it is remotely work related, but two super sales guys blogging, this is a good sign for the target market of Cogmap!

My perspective on online share of voice

Thursday, August 23rd, 2007

iMedia just ran some of my thoughts on share of voice as a byline:

Good stuff. Go read that, it’s all there.

AOL,, the online advertising market, & the New York Times

Monday, August 20th, 2007

Good article by the NYT on AOL. eMarketer says that online ad growth is 28% YoY but AOL and Yahoo grew much less. What I think has been discussed less is how that growth might be represented on the publisher side, putting pressure on Yahoo and AOL. Logically, to grow 30% a site must either grow page views or grow rates 30%. Growing page views is harder than it looks. We are not seeing big growth in new unique visitors online, so the growth must be in pages consumed per unique. Much of the page view growth online today, I would postulate, are in social networks. Sites like MySpace and Facebook have super high frequency and are where we are seeing significant growth in page views. Also, the law of big numbers kicks in for Yahoo and AOL with respect to reach. I think we are seeing that Facebook and MySpace have significantly less reach, so it is easier for them to add new uniques to their platform.

These high frequency UGC page views are occupying an increasing share of online page views and they are inexpensive. This puts pressure on the rate cards of large portals like Yahoo and AOL. I think collectively this means it is really, really hard for AOL and Yahoo to ramp page views at the same pace as the global Internet because they are under-represented in UGC content, whereas the broader Internet minus portals is over-represented in UGC.

PaidContent summarizes the NYT report and there is a comment from “Nick Charles” that goes like this: “ is a great company, but the idea that it would become more important for AOL than the the AOL brand seems misguided…”. He launches this basically ad hominem attack that implies that only AOL giving up could cause to surpass it, but all the evidence presented by the NYT is that is growing faster. All it has to do it keep growing like crazy and it has to happen, right?

Name-based APIs

Tuesday, August 14th, 2007

UPDATE: All of this has been deprecated:

New features for programmers trying to do things with Cogmap: Name-based map identification! It works like this:

  1.<Name of Company> will return either the map of that page or a page to create a new chart.
  2.<Name of Company> will return either the XOXO XML document of the organization structure of that company or a 404 error.
  3.<Name of Company> will return the unique identifier of the chart ID for that company or a 404 error.

Here is an example for Google:
This will allow programmers to integrate applications much more easily. If you want to use our chrome and make it easy for people to extend and update maps, just direct them to our pages with nameapi.php. If you are trying to consume Cogdata but want to capture it using names, then xoxonameapi.php will make it easy to access our data structure using only the names of companies.

The name field, because it is editable in the wiki, is a less reliable field than the unique identifier we assign each map. If you want to create static links to specific maps, it is better to use that maps chart ID. If you wanted to spider the site for all the chart IDs and map names, the best way to do that would probably be to consume and parse the alphabetical list:

For the numbernameapi.php, if a chart exists, it returns the number (CHART ID) of that chart, which could then be used to call the chart page (<CHART ID>) or the XML document (<CHART ID>).

Definitely shoot us emails if you build public applications using this. We love to share whatever SEO juice we have.

Analysis of Sweet Tea

Monday, August 13th, 2007

Once again, my wife has gotten me going on a topic that I must blog about, despite the fact that it is completely unrelated to everything else. Hopefully it will be half as popular as my basketball post, which was the last random rant I did and was linked to by ESPN. Yay.

Anyway, the topic is Sweet Tea! Huge props to Slate for doing a great article covering the phenomenon of sweet tea. No props for not having trackbacks. Lame.

Here are the facts of the matter as I see them:

  1. No restaurant in the northeast makes a good cup of iced tea. Too bitter or too sweet. They do not know how to make it happen.
  2. I remember my first day “up North”. I arrived at college, having never visited before applying/accepting/going, and I go out to eat. Appalled by a variety of terrible events: First, it’s not very good, which I kind of expected. Second, no free refills. No free refills defies the whole southern hospitality of sweet tea. I was glad to see that it got called out at the end.
  3. There is something I think was missed in the article and also missed in the 150+ comments on the article: True homemade tea needs to be SUN TEA. That is how I was raised. It heats and brews naturally for several hours in the hot sun and you are rewarded with the king of refreshment.
  4. I don’t feel the need to heat it hot enough to absorb all the sugar, because it must be sun tea! I am a believer in preparing simple syrup to accompany and sweeten.

Here is my recipe: Add one bag of “Lipton Cold Brew”, the large bags, and three bags of your favorite flavored tea, I like a nice peach ginger tea, to a one quart glass pitcher of water. Cover with Saran Wrap and put it out in the sun for a few hours. Remove tea bags, add some syrup (1 part water, 1 part sugar, boil and cool) and enjoy!

Good sun tea is one of the few undeniable pleasures in life.

Great example of using Cogmap: Oracle Products

Wednesday, August 8th, 2007

I just saw this great blog on a guy using Cogmap to keep track of the Oracle product line. He kind of implies that it would be great if he could lock the page to keep people from changing it, but I think he misses the beauty of a wiki. Now everyone can help him keep the page up to date! This is a good chart, check it out!

Online Ad War – NY Post gets props from bloggers

Wednesday, August 8th, 2007

The New York Post wrote a great article that rapidly turned into a big deal the other day.  Read/Write Web gave it great coverage (and I link to them because they support trackbacks!)

This article is right in my sweet spot, as I ponder how advertisers will aggregate sites like mine to make them buyable.

The logic behind this article is crystal clear: Every big online advertiser except Google misses their top-line numbers, so growth must be happening outside of the top sites.  Conceptually, I love this because the more growth diversifies away from sites that can monetize themselves, the more networks bring value to advertisers by aggregating.

Practically, this is potentially very conceivable because, if we say the market is growing 30% YoY, then to keep pace, Yahoo and AOL have to grow page views 30% Y0Y and maintain their sell-through or they have to increase rates 30% and do the same thing.

As the article says, rates aren’t sky-rocketing, so page views have to grow really fast.  This is hard.

Tons of things fall out of this logic:

  1. More people start to buy sites. bid on Clipmarks apparently.  Sites are adding little sites to get bulk.  Yahoo and AOL are buying aggressively.
  2. Everybody loves a social network.   UGC page views are growing super fast.

Ehh, I could do more, but what I really wanted to talk about was whether this article is even correct.  I went and looked at the comScore top 100 over the last 6 months and what I saw was more page views moving into the Top 10 compared to total Internet page views, and more page views moving into the top 50 compared to total Internet page views.  That would imply that potentially this assertion, while seemingly logical, is just wrong.  It could be that sell-through was down, or rates are down, or Internet growth slowed a bit and no one noticed.

Or it could be that I am looking at the data wrong.  The Top 10 includes MySpace and Facebook, which put up great growth numbers, certainly much better than their peers in the top 10.  This UGC content is kind of dismissed by NY Post, but it comprises much of the Internet today.

My next analysis of this argument would be to look at certain content areas and compare their growth rates to overall Internet page view rates.  The flaw with that is that growth might be coming from the “long tail”.

This really illustrates my axe with this article.  Long on fluff, short on real analysis.  Ms. Sanders says “established Internet players” are suffering.  Maybe I demonstrate my skew by my natural inclusion of MySpace in that bucket.  Who are these “established” players?  How can I identify them to find out if this is a systemic issue or a blip on some sites?  Here is an article where the New York Times refers to MySpace as an “established Internet company”.  I know linking to contrarian Times stuff has to leave the Post burning!

Brand arbitrage

Tuesday, August 7th, 2007

I told Steve Smith the other day that many of the Behavioral Insider columns are so far out on the bleeding edge that they don’t have value to the day to day practitioners in the industry. The Behavioral Insider team at MediaPost has responded with some great stuff recently. I thought their interview with Burst Media was great.

While behavioral targeting offers great value for direct response marketers by allowing them to pinpoint consumers most likely to convert, many direct response marketers have found their ROI is still more than acceptable by buying large volumes of inexpensive run-of-network impressions.

Brand marketers are different. They are so concerned with the quality of the inventory they run on (and brand associations created), in many instances they find themselves paying exorbitant prices for small pieces of inventory. Behavioral marketing allows brand marketers to arbitrage the people they are seeking to less expensive inventory. And not all less expensive inventory is “risky”. Mail inventory is a great example of inexpensive inventory that is high quality but has low demand.

It’s great to see this message getting distributed in the market.

My analysis of “Who’s Now”

Thursday, August 2nd, 2007

Who’s Now“, ESPN’s event to figure out who is the most “now”, has been pretty thoroughly ripped in the media.  Do I think it is dumb?  Yeah, probably.  Figuring out what “now” means is pretty dumb, and I assume it means something like “Who is the most popular person at this instant”, which is definitely not a contest I think about a lot.

Having said all of that, I want to break down the contest for you and predict my winner based on rigorous anti-science:

  1. First, let me start by saying that the beginning was also the end.  LeBron and Tiger had the highest votes of anyone in the first round, crushing their competitors, and now they are in the finals.  In fact, Tom Brady was the only person who #1 seed to not make it to the semi-finals and was also the only person who had the most votes in his bracket to make it through.  You expect #1 to cream #8, but the degree of creaming told the tale of the tape.  Peyton, LeBron and Tiger slaughter their #8 and cruise to the finals.  Tom barely beats David Ortiz and gets tripped up by Shaq in the semi-finals.
  2. Only two upsets would really have mattered.  If you look at places where people came close to winning without winning, only two of them makes a big impact.  Tony Parker upset Federer, but if Federer had won, Shaq would have gotten him next round.  Jeter beat Reggie Bush, but if Bush had pulled it out, LeBron would have crushed him.  Here are some ideas though: If TO had beaten A-Rod, then he would have faced off with Kobe.  Bad boys that bring it on game day.  The closest defeat was LT over Beckham in the first round.  If Beckham wins, he faces Nash and the winner then faces Tiger.  If LT had lost, its a different contest but Tiger still wins out.  The next closest defeat was Shaq over Phelps.  If Phelps wins, then it is Phelps vs Parker to see which underdog gets his clock cleaned by Brady.  Shaq actually defeated Brady, so that is a big swing.
  3. Parker’s upset and Shaq defeating Tom Brady are the only events in the contest where a lower seed advanced.  Federer being seeded #2 was clearly a bad decision by the committee that failed to recognize how “now” Shaq is (#3 seed).

I am sure that is more than you ever wanted to know about “Who’s Now”.

Red Herring is done, stick a fork in it

Thursday, August 2nd, 2007

Red Herring is struggling financially, so they announce that instead of printing the magazine, they are going to go online.  As we have discussed before, editorial-based ideas get destroyed online.  This is financially not any more viable and Web 1.0.  Expect them to not get any more issues out digitally.

I used to love Red Herring.

Also, of course Valleywag and Paid Content don’t support trackbacks.  Dismay!