When blogging, it’s tempting to use your web analytics software to examine what people are searching for, and then build your new blog posts around the search terms people are using. After all, if people are searching for that, it’s a popular topic, so you should write about it, right?
You already wrote about it, and they’re already finding you.
You’re looking at statistics based on the people who are already coming to your web site. For example, if I went by my own web site metrics, then I’d be writing about Perfmon and SQL Server setup checklists every day, because those are why people are coming here now. Those search metrics are almost worthless, because they only report on the content that I’ve already written.
If I’m trying to write blog posts based on what topics people need help with, I need to look beyond what I’ve already written. I need to look beyond my own web site and look at a bigger picture, like:
- Search engine metrics – there are sites that gauge what search terms are popular at Google. Frankly, I think this approach is idiotic, because taken to the extreme, you end up blogging about Britney Spears. Don’t get me wrong – I love reading What Would Tyler Durden Do (very Not Safe For Work) but I couldn’t write that stuff.
- Q&A web sites – read through sites like StackOverflow or forums and get a sense for what questions get asked frequently. For example, people keep asking why their query runs slow, so I wrote a tutorial on how to get help with a slow query. The nice thing about this approach is that I can then use my blog post as my answer or comment in StackOverflow – when someone asks for help with a slow query, I can very quickly say, “Go here to my blog entry, then come back and post more info about your question.” Same thing with my Perfmon tutorial – I can quickly say, “Follow these directions, and send me the output so I can figure out why your server is slow.”
- Your own web searches – whenever I’m trying to solve a problem, the first thing I do is search the web. It’s pretty rare that I have a problem no one else has had before. If I don’t find a good answer in the first page of search results, I make a note of that. When I’m done solving the problem, I blog about what I learned. After all, I can’t be the only one having the problem!
No matter what method you choose, write about things you know. If you choose topics based on popularity, and not your own expertise, your blog topics will start writing checks that your skills can’t cash.
The web site search metrics are still good for something: refining your existing blog posts. I look at the search terms people are using when they come to my Perfmon blog post, and I look to see if I haven’t addressed any of the terms in their search. I try to reverse engineer their real question using their search terms to find out what they really want to know, and then I’ll expand my blog post to include that answer. Over time, your post will become longer and longer (like my Twitter RT FAQ monstrosity) and it’ll snowball to help more and more people.