Write Articles, Not Blog Postings

by Jakob Nielsen on July 9, 2007

Summary: To demonstrate world-class expertise, avoid quickly written, shallow postings. Instead, invest your time in thorough, value-added content that attracts paying customers.

I recently served as a "consultant's consultant," advising a world leader in his field on what to do about his website. In particular, this expert asked me whether he should start a weblog. I said no.

You probably already know my own Internet strategy, so it might not surprise you that I recommended that he should instead invest his time in writing thorough articles that he published on a regular schedule. Given limited time, this means not spending the effort to post numerous short comments on ongoing blogosphere discussions.

Weblogs have their role in business, particularly as project blogs, as exemplified on several award-winning intranets. Blogs are also fine for websites that sell cheap products. On these sites, visitors can often be easily converted and the main challenge is to raise awareness. For example, a site that sells pistachio nuts should post as much content about pistachios as possible in the hope of attracting quick hits by people searching for that information. Some percentage of these visitors will buy the nuts while visiting the site.

Avoid Commodity Status

For many B2B sites with long sales cycles, quick hits to commodity-level content are insufficient. Instead, these sites need to build up long-term customer relationships based on respect.

Take my own business, for example. When I talk with people at my usability conferences, they often say that they've wanted to attend for ages, and only recently secured their boss's approval to come. To address this issue, we added a "convince your boss" section to our conference sites, explaining the benefits of spending money on usability training. Still, realistically, I expect to wait 3-5 years before meeting new readers of my site in person.

Blog postings will always be commodity content: there's a limit to the value you can provide with a short comment on somebody else's work. Such postings are good for generating controversy and short-term traffic, and they're definitely easy to write. But they don't build sustainable value. Think of how disappointing it feels when you're searching for something and get directed to short postings in the middle of a debate that occurred years before, and is thus irrelevant.

Obviously, I am referring to the user experience and to the style of the content in this analysis; not to the technology used to serve up this content. Thus, what I call "articles" might be implemented on a blogging hosting service. What matters is that the user experience is that of immersion in comprehensive treatment of a topic, as opposed to a blog-style linear sequence of short, frequent postings commenting on the hot topic of the day. It doesn't matter what software is used to host the content, the distinctions are:

  • in-depth vs. superficial
  • original/primary vs. derivative/secondary
  • driven by the author's expertise vs. being reflectively driven by other sites or outside events

Demonstrate Leadership

For the sake of argument, let's say that you're the world leader in your field. We'll quantify that as being the #1 expert among the 1,000 people with websites in your field. In other words, you are in the 99.9th percentile.

(Although you might think you have many more than 1,000 competitors, the Web thrives on specialized content, so it's better to conceptualize yourself as leading a smaller subdiscipline, unless you're so good that you're #1 out of millions of people.)

We can measure expertise as some combination of intelligence, education, experience, correct methodology, professionalism (say, avoiding profanities and politics), and willingness to be frank. The exact metric doesn't matter here; let's just assume there's a way to quantify how good people are within their field. The metric probably follows a normal distribution, meaning that the 1,000 people have the following levels of expertise:

Normal distribution of one thousand data points
Histogram of expertise scores for 1,000 authors. Each dot is one person.
Stupid people are on the left; clever ones are on the right.

Assuming that you're this good, you have to show it to gain customers. And blogs aren't the way, as we'll see once we plot the distribution of postings as opposed to writers.

Variability of Blog Posting Quality

Assume that the 1,000 people each write 10 blog postings. The resulting 10,000 postings will follow a much broader distribution, because the quality of postings is extremely variable.

Let's assume that a given writer's posting quality is normally distributed, with a mean representing that person's level of expertise and a standard deviation 3 times as large as the SD for expertise among people. I don't know what the actual number is, so this is just a rough estimate. But it's reasonable to assume that posting quality is more variable than expertise for several reasons:

  • Sometimes people toss off a posting in a minute. Other times they spend hours.
  • Sometimes a writer happens to know a lot about the topic at hand, possibly because they've just spent several months working on that exact problem. Other times people know nothing — which doesn't keep them from voicing their opinions :-)
  • Sometimes people are lucky and get a blinding insight. Other times they post more out of duty than anything else.

The following chart shows the distribution of the quality of 10,000 postings in one Monte Carlo simulation I ran:


Distribution of ten thousand data points, each representing one blog posting, ranked from low to high quality.
Histogram of 10,000 blog postings' quality.
Each dot is one posting; the highest-ranked expert's postings are shown in red.
Bozo ramblings are on the left; insightful stuff is on the right.

Of course, if I'd run many more simulations, the histogram would be smooth, but the overall shape would be the same.

In the above histogram, each of the tiny dots represents a blog posting. The larger red dots indicate the ten postings by our leading expert (who was ranked #1 out of the 1,000 bloggers we're considering). Although our expert tends to write good postings, a few of the many lower-ranked people will sometimes write even better postings.

Even if you're the world's top expert, your worst posting will be below average, which will negatively impact on your brand equity. If you do start a blog despite my advice, at least screen your postings: wait an hour or two, then reread your comments and avoid uploading any that are average or poor. (Even average content undermines your brand. Don't contribute to information pollution by posting material that isn't above the average of other people's writings. Plus, of course, follow guidelines for blog usability.)

In my simulation, our expert's best posting happens to be #25 from the top. The expert's second-best posting was ranked #300 from the top. It might seem fine to be the author of postings #9,700 and #9,975 out of a group of 10,000 blog postings. But in fact, it's nowhere near good enough.

The beauty of the blogosphere is that it's a self-organizing system. Whenever something good appears, other blogs link to it and it gets promoted in the system and gains higher visibility. Thus, the 24 postings that are better than our expert's very best attempt will gain higher prominence, even though they're written by people with lower overall expertise.

Prospective new customers don't even have time to read 24 postings, so they'll never make it down the list of rank-ordered blog postings to reach our expert's best.

Beating the Internet

It's almost impossible to fight the Internet: you're up against millions of people who are willing to work for free. But you have to do so, because if you work within the prevailing Web paradigm you're letting the search engines take 98% of your content's value. That's okay if you're not in the content business. Our pistachio site doesn't mind that it's not making money off its recipe for delicious pistachio ice cream. Just as long as it sells nuts.

If you're an expert who wants to live from adding to the world's knowledge, you must go beyond the mainstream Web model of single page visits driven by search traffic. It's easy enough to build a website that freeloaders will use, but that shouldn't be your approach. You must change the game and create content that's so valuable that business users are willing to pay for it.

You should also focus on material that lower-ranked content contributors can't easily create in their spare time.

Both of these needs are met when you produce in-depth content.

In-Depth Content Is Value-Add Content

It might take you only an hour to write a blog posting on some current controversy, but a thousand other people can do that as well (in fact, they'll sometimes do it better, as shown above). And customers don't want to pay for such a tiny increment of knowledge. Sure, sometimes a single paragraph holds the idea that can increase a site's conversion rate so much that a reader should have paid a million dollars to read it. But they don't know that in advance, so they won't pay.

In contrast, in-depth content that takes much longer to create is beyond the abilities of the lesser experts. A thousand monkeys writing for 1,000 hours doesn't add up to Shakespeare. They'll actually create a thousand low-to-medium-quality postings that aren't integrated and that don't give readers a comprehensive understanding of the topic — even if those readers suffer through all 1,000 blogs.

Thorough content's added value can rise above the threshold where customers become willing to be separated from their money. This is the true measure of a sustainable business.

You have to identify opportunities with a non-linear utility function: where paying customers assign more than 10 times higher value to something that costs 10 times as much to produce. The old open-source manifesto " The Cathedral & the Bazaar" holds much truth: when you're the duke, you can't trade in coffee beans, because the bazaar dealers will always undercut your price. You should build a cathedral, because a thousand tents can't compete with the Notre Dame.

The following chart shows another example from my own company: trends for key statistics across three editions of our report on e-mail newsletter usability:

Trends for page count, price, and sales throughout three editions of a report. All three values went up for each edition.
Statistics through three editions of the e-mail newsletter usability report.
All numbers are indexed to make the first edition the baseline with index 100.

As the chart shows, the fatter the report became, the more it has sold. Of course, page count (the blue line) is only a rough indication of the amount of insight, which is what customers are really paying for. The new edition has a large number of eyetracking heatmaps, showing how users read various newsletters, and these many illustrations eat up pages ferociously. Still, there's no doubt that each report edition contains significantly more information than previous editions.

The report's price has increased less than its page count: as we keep doing this research, we become more efficient. You could argue that customers are getting more for their money, and that's why they're buying more. But this argument works only if customers in fact assign extra value to more comprehensive reports. So either way, I conclude that in-depth content sells.

Why are paying customers (the people who matter) attracted by detailed information? Because systematic and comprehensive coverage is more actionable. It also protects them against the risk of losses caused when something important is overlooked.

In my report example, consider an Internet marketing manager who's in charge of the company's email newsletter. The report's price is trivial compared to the millions of dollars many companies would gain from increased subscription rates, increased open rates, increased clickthrough rates, and enhanced customer loyalty from content that's both better appreciated and read more often. To improve these key performance metrics for her newsletter, the manager could spend a week surfing the Web and reading a thousand short pieces about newsletter design. The result? A scattered set of imprecise advice that neglects many important issues. Instead, that manager could spend a day gaining much deeper insights from reading a single, well-structured report with all-inclusive coverage of the topic (and based on empirical data instead of each blogger's personal opinion). Saving 4 days is worth a lot in business, which is another reason to target business customers with value-added information.

In-depth content provides more value in less time than numerous superficial postings. That's why business customers have empirically been willing to pay, and that's why you should emphasize fewer, better pieces as your content strategy.

Expertise vs. Content Usability

This has been a very long article, stuffed with charts, mathematical modeling, and theoretical concepts — like standard deviations and utility functions — that I know most readers find difficult. Recommending in-depth content flies in the face of all guidelines for Web writing, which call for fewer words and scannable information.

The content usability guidelines are correct: they are indeed the way to make a site easier for most people. Thus, you should follow the guidelines — rather than emulate this article — for normal business websites and intranets. (When I say "business sites," I include government sites and non-profits, as well as e-commerce and corporate marketing sites.)

For most sites, the content is not the point. Instead, you want to answer customers' questions as rapidly as possible so that they'll advance in the sales cycle and start buying (or donate, or sign up for your newsletter, or whatever else you want them to do).

Elite, expertise-driven sites are the exception to the rule. For these sites, you don't care about 90% of users, because they want a lower level of quality than you provide and they'll never pay for your services. People looking for the quick hit and free advice are not your customers. Let them eat cake; let them read Wikipedia.

Still, even if you run an expertise-driven site, you should comply with the bulk of content usability guidelines: be as brief as you can; use bulleted lists and highlighted keywords; chunk the material; and use descriptive headings, subheads, and hyperlinks. The small percentage of users who are qualified prospects still read in an F-pattern, so a headline's first words are more important than its last words, just as they are for normal sites.

See Also: Writing for Social Networks: Usability of Corporate Content Distributed Through Facebook, Twitter & LinkedIn.

Full Report

Full eyetracking report on how users read on the web is available for download.

Share this article: Twitter | LinkedIn | Google+ | Email