I can't really say without giving my details, and as you can see I like throwaways :) especially when I'm at work.
The difference was that the rest of our content was bland marketspeak whitewash garbage stuff. "Expand your client list 5x TODAY" type bs. Also useless PDF whitepapers on project we completed. Pro tip, nobody reads PDF unless it's the only content available, and on Google that's rarely true :).
The new articles were all focused on specific issues that our clients would have. Some favorites:
"TWC business modem speed up"
"AdWords tracking conversions"
"Choosing the right software development company"
"Pros and cons of outsourcing development"
"Online marketing for small businesses
"AdWords overview for CEOs and VP s"
"How to hire AdWords expert"
"Company outing locations in (cityx)"
"Most popular CMS for company websites"
"Small business tax audit guide"
Articles with a clear purpose and tailored to what our kind of client, and person, would be searching for. Admittedly this takes some really creative thinking but the results can be amazing.
Not all of our articles were related to what we did at all. They were built so that our target customers would search for them. Business owners, especially local. People looking for custom software even if it's not what they initially have in mind.
Our only channel for these articles was google search. In all, about 95% of our traffic went to new articles straight from Google. We trailered content mainly to our location, and this is probably a wise choice for any small business.
The content raised Google's opinion of how good our site was, even when the article wasn't bringing in sales. Most of the time they would be first page for targeted keywords, about 40% of the time the first result. Some of them are years old now and still first on Google, if that gives you an idea of how powerful well written pieces can be.
Writing the actual content: make sure the article is long and well written. The times/WSJ/new yorker should be a guide for what kind of writing google likes. Long form and extremely focused on a subject. Pictures help too.
Google looks at how many people view your article, for how long, and whether they go back to search afterwords if it wasn't what they wanted. They track future searches related to the article they just brought you to, to see if it has stimulated the person to read more into whatever your article was talking about. So for example the best CMS article. Did the user that read your article end up checking out the CMS you mentioned? They track how often users return to the same article, and probably whether they bookmark it. They even know how deep your knowledge is by comparing the distribution of uncommon words in your article to more trusted sources like journals and high quality books on the same subject. This is all admittedly speculation but google is extremely good at rating the quality of your content. The new AI processing google uses probably doesn't even know why an article is good, it just knows that it is.
The thing that helped me most writing the good articles was a deep background in the subject. For things I didn't know as much about, like managing the business, I would pick the brain of somebody that did know a lot for an hour or two then put it to writing.
If you don't have much knowledge about the things you should be writing about you need to find a decent writer that does.
If you don't want to deal with all that just pay google money for ads, and get really good at "the game". AdWords is extremely competitive and you need somebody with the drive to win at any cost, because the road from the start of your first campaign to something profitable will be expensive and hard.
Channels are overrated IMO :) you can reach around 90% of all web users on Google alone, and close to 100% on Google+fb. To start out at least, just pick a few very pervasive platforms and get really good at using them