7 Ways To Evaluate Blog Post Quality — Tuning your BS detector

(Reading time: 5 – 8 minutes)

Would you recognize a bogus blog post if it bit you on the butt?

While some blog posts are obviously bogus, for others, the bogosity is not quite so obvious.

I was recently doing a bit of due diligence on SEO for Website In A Weekend readers, and it occurred to me that most of what I was reading about SEO was not very good. As in, essentially useless, which is to say, bogus.

It turns out I have a sideline career as an occasional reviewer for a couple of academic journals, so my BS detector is fairly well tuned.

Here are some of the indications I find when content has little or no value:

1. Rehashed content. The same old, same old. Over and over again. On the one hand, it’s reaffirming to read and reread (ad infinitum) “what we know is true.” On the other hand, that’s a not a good way to solve problems and make progress. Check for actionable content, if you can’t find any, it’s not worth saving or bookmarking.
2. Inapplicable content. You click through to some page, and the title of the article bears little or no relationship to the text of the article. Bogus!
3. Written anonymously. Private Label Rights material on SEO. You can buy the same article everyone purchased. Post it under your name (or make up a fake name), just like everyone else does. Bogus!
4. Irrelevant content. Does the subject of the article really apply to your business or interest. For example, do you really need to care that much about SEO? You might not. It depends on your goals.
5. Zero human interest. Really strong articles will have a human voice. You will hear it. Bogus articles read like they were written by machine. In fact, sometimes articles are auto-generated, purely to game search engine results! The yield on these cannot be very high.

   Human interest serves a couple more purposes: 1. helps you track down content thieves, and 2. helps you connect with readers. Your little stories (ok, my stories) might be boring and pedantic, but a machine isn’t clever enough to achieve boringness.

   This article was actually motivated when I was doing some research on some topic (I forget now), and kept running into machine generated-pages. I refuse to link to any of these pages, even as example, but in the future, I’ll update this article and comment on a screenshot.

6. Stuffed with keywords. These articles can be hilarious, keywords and phrases bolded, emphasized and repeated ad nauseum all through the article. If Google doesn’t have a keyword ratio, they should. Something #keywords/#words total. Anything over a certain value would be considered stuffed. Send me an article if you see one of these, I’ll compute the ratio.

   These aren’t that hard to find either. Just do a search on a hot topic. You will be lucky if half the top page of search results is useful information.

7. Weak comments: no comments, poor comments, bogus comments, and no author response in comments. It’s amusing to see articles with 15-20 one line comments along the lines of “Nice post!” or “Very useful information, thank you.” Perhaps some of these are real. Almost surely not all of them are.

   Smashing Magazine may be an exception here. Smashing doesn’t allow URL linking, and runs a pretty stripped down commenting system: name and email only. The last couple of articles I read had quite a few comments, but they were pretty weak comments overall. Worth watching though.

The obvious (to me) current example of all of the above is the SEO thing: a big space filled with little content. Plus, “black hat” SEO can game up almost any article like the above conditions into a very high ranking on search engines. You’ve surely seen all the above.

Here’s what you can do to add quality signal and reduce the noise:

1. Specifically link to quality pages. Don’t worry about losing your “juice.” If the linked page is high quality it will benefit your readers.
2. Do not link to low quality pages at all.
3. For the long term, be careful about accepting links in from garbage pages. I generally refuse trackbacks to page having no information. That is, I do, when I catch it. Some slip through.
4. Link to people, real people, not “websites.” For example, I have linked out to Sean, Holly, Jan and Extreme John. And Larry too. These are real people, not just websites.

   I’ve linked out many others here, and will continue to do so. You’ll meet more of these folks in the future. In the meantime, introduce some of your audience!

I’m really sure that many of these bogus websites and blogs -for now – get more traffic than Website In A Weekend. I’m also really sure that none of them will ever develop a real, live audience of human beings.

In the long term, it’s the people that make the difference.

Can you add to this list of bogus detectors? Please do, I’m sure I missed a few things.

[Update November 6, 2009]

From The Freelance Writing Jobs Network, 10 Tips for Telling if an Article Contains Reliable Information:

Bad content floods the web. It’s so bad that schools are giving out guidelines for sites to avoid when collecting information for reports. Many writers also use the web to research information, but how can we know if it’s someone else’s unreliable content rewritten ten times, or if it’s a realistic investigation or expose?

---

Would you like more? Send me a letter...

Tagged as: Black Hat SEO, Bogus comments, Comments, Research