One of us is smarter than all of us

The wisdom of crowds comes not from the consensus decision of the group, but from the aggregation of the ideas/thoughts/decisions of each individual in the group.

At its simplest form, it means that if you take a bunch of people
and ask them (as individuals) to answer a question, the average of each
of those individual answers will likely be better than if the group works together to come up with a single answer. And he has a ton of real examples (but you’ll just have to read the book for them ; )

[Also] diversity increases the quality of the aggregated wisdom of the group.
If you have too many people who are alike, then no matter how smart
they all are, they may not come up with the same quality of answer than
if you have less smart folks who have a very different point of view. Diversity brings new information. And that new information is valuable.

In order for the crowd to have wisdom, the crowd has to be made up of individuals who argue! Or as he puts it in the book,

and independence are important because the best collective decisions
are the product of disagreement and contest, not consensus or
compromise. An intelligent group, especially when confronted with
cognition problems, does not ask its members to modify their positions
in order to let the group reach a decision everyone can be happy with.
Instead, it figures out how to use mechanisms–like market prices, or
intelligent voting systems–to aggregate and produce collective
judgements that represent now what any one person in the group thinks
but rather, in some sense, what they all think.”

And my favorite line that sums it up:

“Paradoxically, the best way for a group to be smart is for each person in it to think and act as independently as possible.”


How to win the search position game

Focus on what each click is worth, not on what position it should be in
general, if a purchase conversion is worth $10, and one out of 10
people purchases, you should pay about $1 per click. You should offer
that maximum price to Google (or another search engine) for the
specified keyword.

After you launch campaigns, continue to test them for conversion
metrics and adjust your top bid accordingly. Many marketers think that
if the clickthrough rate is higher, the keyword should be more
expensive. But you should determine the value of a keyword based on
conversion rate, not clickthrough rate, because you only pay by the

Heads or tails?
Head keywords are generic terms
that people search while browsing or doing product research, such as
“mp3 player.” Head keywords often benefit from being in first position,
because they capture a lot of “browsers” who just click on the first
link and may be exposed to your site for the first time. These people
may not buy now, but they’ll connect with your brand.

Tail keywords are often best in third or fourth position. These
keywords are specific and appeal to committed buyers, such as “black
ipod nano 8gb.” People searching for these keywords are usually more
ready to buy, so they’ll look at — and even click through
— several ads to find the best deal, even if that deal appears in
a link halfway down the page.

The upshot? Head terms get much more volume and are often more
expensive to boot, so to justify your investment you may need to
measure carefully which visitors return to your website.

Set a top position
This is a tool on Google you
can use to hold your keywords down in the rankings, even if you are
bidding enough to be #1. It’s always better to figure out first how
much your keywords are worth to your bottom line, and then find out
where that places you. But this tool can be useful if you find that
position #1 gets a lot of poor quality traffic that never converts.

Focus on the dirty dozen
Most marketers spend the majority of their budgets on a few top keywords, usually about a dozen, which are high volume and
have a strong conversion rate. Focus on fixing the position of these
keywords first, because correctly placing these top keywords will have
the biggest impact on total revenues. Let the others fall where they
will according to their conversion rates as described above.

Turn off Google Search and Content Networks
you don’t opt out of Google’s search partners, like AOL and,
your position numbers will reflect a blend of your positions across all
of those properties. To get an accurate picture of where your keywords
are positioned on Google itself, turn off the additional distributions.
You can always turn them back on after you finish your measurement.

Turn off Google Content Network. Ditto as above
figure out what your keywords’ true positions are, focus on Google
itself, not your position across all its content partners, such as New
York Times, MySpace and

Work weekends
Some keywords perform stronger on
the weekend, such as “gardening” or “beach wear,” for example. Set up
automatic bid increases for these terms to boost your position solely
on the weekends. (Google supports this at the campaign level; MSN
supports this at the Ad Group level; and Yahoo doesn’t support it right
now.) Remember: These boosts should be based on changes in conversion
rates, not click volume. Look for the pattern before you set the boosts.

Pony up for brand and “executive” keywords
you’re Coca-Cola, you just have to pay whatever it costs to have
“Coca-Cola” be in the top position — that’s crucial for your brand.
Plus you can use your company name in those brand-term ads, and other
advertisers cannot (call the support team at the search engine if you
see any violations of this). Likewise, if your CMO tells you the
company needs to be in top position for certain keywords, like “digital
camera” or “PC” to build your brand in those categories, then just pay
what it costs to be in the top spot (and pull the cost from the
branding budget!).

Chris Lien


Writing Style for Print vs. Web


Linear vs. non-linear. Author-driven vs. reader-driven. Storytelling
vs. ruthless pursuit of actionable content. Anecdotal examples vs.
comprehensive data. Sentences vs. fragments.

I’ve spent many columns explicating the differences between the Web and television, which can be summarized as lean-forward vs. lean-back:

  • On the Web, users are engaged and want to go places and get things done. The Web is an active medium.
  • While watching TV, viewers want to be entertained. They are in relaxation mode and vegging out; they don’t want to make choices. TV is a passive medium.

This doesn’t mean that you can’t have entertaining websites or
informative TV shows. But it does mean that the two media’s contrasting
styles require different approaches to entertainment and education.

The differences between print and the Web may not seem as strong,
but to achieve optimal results, each requires a distinct content style.

Example: Tall Travelers

I recently read an article in The New York Times about tall people’s travails on the road: “Coping With the Tall Traveler’s Curse.” The headline itself is actually an example of the differences between print and online content style:

  • In print, a phrase like “tall traveler’s curse” is a bit enticing
    and might draw readers in. Because the article featured a photo of a
    tall guy crunched in the back of a taxi, the article’s content was
    clear to anybody glancing at that page in the newspaper.
  • In contrast, putting the same headline online would fail several guidelines for writing for the Web:
    • The first 3 words have no information-carrying content. On the Web, you must start with words like “tall traveler” because users often scan down the left part of a list of items. They never see the last words in a link unless the first few words attract their attention.
    • The headline lacks keywords — such as
      “airline seat” and “hotel bed” — that are important for search
      engine optimization (SEO). No one will search “curse” when trying to
      find out which hotel chains offer extra-long beds or which airline
      seats are the least unpleasant for long-legged travelers.
    • The words “tall traveler’s curse” are insufficiently specific
      to tell users what the story is about. Because headlines are often
      presented as plain links removed from the article itself, the photo of
      the poor guy in the cab won’t be there to explain the story’s content.
      Online, the headline alone must provide enough information scent to let users predict what they’ll get if they follow the link.

Even though I’m not particularly tall myself, I read the entire article
in the printed newspaper. Why? Because it was well written and
contained several interesting anecdotes about tall business travelers,
ending with the story of a tall woman executive having to bend down to
use a hotel room makeup mirror.

I would never have read that same article on, because
the story lacks both immediacy and utility. Even though the article
surely attracted some pageviews online, it’s style is not optimal for
presenting information on the Web.

The Web rewards comprehensive coverage that’s more specific
than print content. On the Web, content for tall travelers should
feature ratings of airline seats and hotel beds for all the major
airlines and hotel chains, respectively. Even better would be to
differentiate coverage for tall men vs. tall women and for somewhat
tall vs. gigantically tall people.

This more detailed approach works online because the content is
searchable and you can sort and present it in personalized views for
each user. Say, for example, you’re 6-foot-8 (2.03 m) like the guy in
the article photo, and you’re flying United Airlines from San Francisco
to Chicago. A good site will tell you which departing plane has the
best seat configuration for you, and which seat you should book.

Narrative vs. Actionable Content

Print publications
— from newspaper articles to marketing brochures — contain
linear content that’s often consumed in a more relaxed setting and
manner than the solution-hunting behavior that characterizes most
high-value Web use.

In print, you can spice up linear narrative with anecdotes and individual examples that support a storytelling
approach to exposition. On the Web, such content often feels like
filler; it slows down users and stands in the way of their getting to
the point.

For example, in print, discussing the tall-friendly rooms in
the Palms Casino Resort in Las Vegas feels somewhat interesting. That’s
not the case online when a user is looking for tall-friendly rooms in
Chicago (or wherever he or she is going next week).

Web content must be brief
and get to the point quickly, because users are likely to be on a
specific mission. In many cases, they’ve pulled up the page through
search. Web users want actionable content; they don’t
want to fritter away their time on (otherwise enjoyable) stories that
are tangential to their current goals.

Instead of a predefined narrative, websites must support the user’s personal story
by condensing and combining vast stores of information into something
that specifically meets the user’s immediate needs. Thus, instead of an
author-driven narrative, Web content becomes a user-driven narrative.

Print’s narrative exposition calls for well-crafted, complete sentences. Online, less so. Fragments often let you pull information-carrying keywords to the front, while also reducing froufrou word count. Because Web users read only 18% of added verbiage, cutting words is well worth the accusing squiggles that MS Word will throw at your sentence fragments.

E-Learning: An Oxymoron?

I continue to believe in the linear, author-driven narrative
for educational purposes. I just don’t believe the Web is optimal for
delivering this experience. Instead, let’s praise old narrative forms
like books and sitting around a flickering campfire — or its
modern day counterpart, the PowerPoint projector — which have
been around for 500 and 32,000 years, respectively.

I continue to write books, and I continue to develop training seminars, because I believe these media are best for deep learning of new concepts.

We should accept that the Web is too fast-paced for big-picture learning. No problem; we have other media, and each has its strengths. At the same time, the Web is perfect for narrow, just-in-time learning
of information nuggets — so long as the learner already has the
conceptual framework in place to make sense of the facts.

For example, I dated “learning around the campfire” to 32,000 years ago
to coincide with the emergence of high culture and the Cro-Magnons. Not
that the Neanderthals didn’t have campfires — they simply didn’t
have the cultural depth of modern humans, so I don’t think their
storytelling was equal to my seminars. So, did I actually remember that
Cro-Magnon culture started 32,000 years ago with the Lascaux cave
paintings? No, I looked that little fact up online.

Writing for Selfish Readers

In linear media — such as print and TV — people expect you to construct their experience for them. Readers are willing to follow the author’s lead.

In non-linear hypertext, the rules reverse. Users want to construct their own experience
by piecing together content from multiple sources, emphasizing their
desires in the current moment. People arrive at a website with a goal
in mind, and they are ruthless in pursuing their own interest and in rejecting whatever the site is trying to push. Banner blindness is only the most extreme manifestation of this selfishness.

Particularly on commercial sites — whether they’re B2C e-commerce or specialized B-to-B
sites — users cherry-pick the information and concentrate
narrowly on what they want. If you’re smart, you’ll write accordingly:
make your content actionable and focused on user needs.


Don’t design ‘what if’ navigation

The left column navigation should point forward; drive you
towards your destination. The BBC website is a really great
design. When you get to the homepage, you get two major options:
News and Sport.

Clicking on Sport gets you to a left column navigation beginning
with: Football, Cricket, Rugby Union, etc. (Notice it is not
alphabetical; it begins with the most popular, Football.)

Clicking on Football removes all the other Sport options and
shows you just Football ones. What if someone is interested in
Cricket? If I was interested in Cricket I’d have clicked on

You keep clicking until you select, say, Manchester United. In
the left navigation you now see links such as Squad Selector,
Results, Fixtures, etc. The worst navigation systems would
continue to offer you Cricket in the left navigation. Of course,
that would result in a huge list of links in the left navigation
that would cause great confusion.

Gerry McGovern