The Prescience of the Clue Train Manifesto & The Genesis of Web 2.0

Shannon Ryan, our fearless leader, is out in Redmond today attending the Microsoft Canadian Leadership Summit. He gave a pre-conference talk this morning on Enterprise 2.0: Social Networks Behind the Firewall. ( I’ve posted his presentation to Slideshare).

Be Right, Be First - and Then Hope People Remember

Shannon makes some really solid observations in this presentation – we’ll be capturing many of them in a white paper Social Networking in the Enterprise we’re releasing at the end of June. (Email me if you’d like a copy).But one observation really struck me. Those Cluetrain folks got it right. If you don’t remember them, four Silicon Valley writers published the The Cluetrain Manifesto:The End of Business As Usual in 1999. It contains observations that have proven remarkably astute.

While the business press seems to have given Tim Oreilly of Oreilly press the nod for coining “Web 2.0”, I’d suggest that the Cluetrain writers nailed the concept down years before.

Six of the 95 Theses

Here are six of the 95 theses they propounded that sound like an analyst from Forester or Gartner writing today:

1. “Markets are conversations.” (Is this what we mean by social media?)
2. “Markets consist of human beings, not demographic sectors.” (Hmm. Sounds like online word of mouth or social buzziness)
3. “We’re both inside companies and outside them. The boundaries that separate our conversations look like the Berlin Wall today, but they’re really just an annoyance. We know they’re coming down. We’re going to work from both sides to take them down.” (Tweet, Crackberry, Facebook status updates)
4. “Companies typically install intranets top-down to distribute HR policies and other corporate information that workers mostly ignore.” (Or can’t find – fix the bloody intranet search engine)
5. “The best intranets are built bottom-up by engaged individuals cooperating to construct something far more valuable: an intranet-worked corporate conversation.” (or, as the pundits would belatedly name it – Intranet 2.0)
6. “While this scares companies, they also depend heavily on intranets to generate and share critical knowledge. They need to resist the urge to “improve” or control these conversations.” (Corporate Wiki’s for knowledge retention; Facebook inside the enterprise)

Locke, Levine, Searls and Weinberger were early in the curve – and I admit that I glanced at the book when it came out, classified it as more .com lunacy and set it aside. But Shannon’s right when he says that their observation – that business is about humans and humans like to interact, share, talk and work with other humans – is at the heart of the Web 2.0 revolution.

“How much copy do we need to write for this page?”

It’s a frequent question. Like most important things in life, it doesn’t seem to have one answer.

Finally – Empirical Data

Last week Jakob Nielsen published a quick study (based on the data set from this – impressive – analysis of real-world browsing activities.) He reaches two conclusions that are important for marketers

• You can expect the average visitor to actually read about 1/5th of your copy
• Visitors spend about 4.4 for seconds on your page for every 100 additional words (Neilsen says “only 4.4 seconds.”)

This seems to be helpful. It’s well-founded, empirical proof that short copy is more likely to be read in its entirety. At the very least, it suggests that web sites should be structured with many pages, each dedicated to a single topic and containing the least amount of copy necessary to convey the intende message effectively.

But What About A|B Testing?

But how do we reconcile this discovery with reports from disciplined marketers that the conversion rates of pages containing long copy frequently outperform short copy pages? Consider this experiment on copy length conducted by Marketing Experiments.  It determined that:

• In a real world test, long copy outperformed short copy by nearly 400%.
• Where perceived risk or required commitment level are large, long copy may be critical.

At least in some cases, ensuring visitors read all of the content on a page isn’t necessary for delivering results. In fact, I suspect most marketers would argue that their goal is to convince readers, not convey ideas; to compel action, not inform. In this context, it may just be a numbers game: the more words on a page – the more benefits conveyed – the more likely one chunk of text will catch the reader’s attention and drive a download, free trial or purchase.

And Isn’t 4.4 Seconds Worth Thinking About?

Nielsen does the math and comes up with an equation that relates time on site to number of words on page. The length of time a visitors stays on a page seems to scale linearly with the number of words on the page. He interprets this to suggest that while users spend more time on pages with more copy, “they spend only 4.4 seconds for each additional 100 words.”

A more important interpretation might be – more words on the page drives greater visitor engagement. Adding 100 words seems to keep a visitor on your page for 4.4 seconds longer. Add 300 words and Nielsen’s data suggests your visitors will engage with your content for 13 seconds longer. Is this good? It depends on the objectives of your copy.

The Bottom Line: What this Means to Marketers

I don’t believe there’s a definitive answer to “how long should the page be?” The answer varies with the objective of the copy : is it written to inform or is it written to motivate an action?
This is what we suggest:

• If you are aiming for a specific outcome – a download, trial, purchase, etc. – then iteratively test copy length and approach so that you optimize results for your audience and business context.
• If you are trying to convey complex ideas – for example, if you have a long sales cycle and you are providing background information – then put the copy in a PDF. And write supporting copy with the goal of convincing visitors to download, print and read the document.
• If you are conveying simple ideas – the time a movie starts, the location of a restaurant, George Bush’s understanding of international affairs – then user the fewest words possible and leverage bullet lists and headers to improve scanability.

Finally, if you are not sure what to do, opt for well-written, lengthier copy. Even if your human visitors don’t read all of it, the search engines will. And this may increase the likelihood of ranking well for “long tail” search terms.

Get Interactive

Agree? Disagree? Think this post is too long? Comment or drop me an email.