How to Up Your SEO Income by a Factor of 10: Testing

The following is a guest post by Ari Ozick, a member of our SEO Community who frequently shares great insights. :)

The above graph is based on an intensive questionnaire of over 2500 world class SEOs, including freelancers, in-house corporate types and SEO entrepreneurs. In the questionnaire, I asked one simple question – what is the most profitable activity in your business. The results are in, and as the graph clearly shows, Twittering away the day is by far the most profitable activity for most SEOs. Apparently there is more money to be made on twitter then there is in link building and out ranking the competition.

Obviously (I hope), I’m lying. I didn’t conduct any survey. I just made up the graph with Smart Draw. In fact, if I had to guess, I imagine the most profitable activities for companies would be conversion optimization, link building, and public relations.

Here’s the thing, though: I don’t have to guess about any of these answers for my business. I know. If you’re guessing or following other people’s advice on pretty charts (even if it’s backed up by third party expert opinions), then you’re leaving a lot of money on the table. Let me show you how, with a little change in thinking and how you approach business, you can make a lot more money out of what you do, everyday.

Most SEO Blogs have an Agenda

Aaron recently mentioned that there are over 5,000 SEO blogs out there today. A lot of people read the more popular SEO blogs as if they were the Gospel itself. What most of them don’t realize, or don’t want to realize, is that almost all of these blogs have one of a few agendas:

  • To promote the business running the blog so they can get more clients
  • To promote the blogger as an SEO expert so he can get more clients.
  • To get lots of links, so that they can rank for [SEO] and, you got it, get more clients and sell advertising

Granted, there are a few exceptions. The bottom line is, however, that most of these blogs are fundamentally aimed at increasing their readership, their clients, and their reputation. That means that the information they offer is less aimed at being fundamentally useful, and more about furthering their goals. It very rarely is about providing concrete, useful tips that will lead to a direct increase in your ROI.

Don’t Believe Anything Anybody says

Often, someone will come to me and tell me that they’ve stopped writing content, or that they don’t want any reciprocal links, or that they don’t want to be listed on site X. Then I ask them why. Invariably, the answer I get is “I read it on a blog” or “It was on a forum”, or “I saw it on Twitter”. You need to make a business decision based on data, not on what something written somewhere on some forgotten piece of the internet.

Different sites will also have different focuses that provide higher ROI. E-Commerce stores may put more focus on optimizing for conversions from existing traffic, while sites like blogs and forums that sell advertising on a CPM or flat rate model will optimize for higher traffic – link baiting and public relations. Niche Adsense publishers and others operating on the longer tail of search will look to create volume quality content and build links in a more focused manner, sending link juice to the pages that need it most.

Data is King

The only way to make a decision is by looking at data gathered and seeing if the data provides you with enough information to make an informed decision. I think Diorex said it best:

I will share with you the same answer that my employees get.

“Do you have any data?” The answer is usually “No”, or they would not have asked the question.

To which I respond “Well why don’t you run a test and get some data.” Once a test has been run, they no longer ask my opinion because they now have an answer (good or bad, testing will give an answer)

I have said it before and will probably say it again, buying data in the form of testing is the best investment you can make in your business. It is not cheap, which is what scares most would be internet marketers away.

Getting the Data

So Data is what helps make informed decisions. How do you get the data? If you’re doing anything PPC/CPM based, you just need to start running a test campaign and use that data to scale, or alternatively shut down the project before you lose too much money.

SEO is a bit different – no two web sites have the same link profile, and minor differences can lead to very different results in terms of how certain types of links and on page changes will effect changes in search results. That being said, it’s best to have a secondary group of sites so you can measure the effectiveness of different link building methods, without endangering your profitable, money making sites.

Ideally you also need to have a good idea of the link graph in the verticals you work in, and an idea of what competitors are trying to accomplish. To that end, I highly recommend Majestic SEO and SEM Rush (I’m a happy customer, nothing more). There’s nothing like having fairly accurate data without being at the complete mercy of a search engine. It’s a liberating feeling.

What Data Has Taught Me

Data has taught me that what works for one site doesn’t necessarily work for another site. Strong sites with aged links have consistently performed better when they receive low quality links, while newer sites have languished until they received some better links.

In one test, we sent low quality links to an aged authority site in a competitive niche. These are links that are probably not your top priority on your link building list, and certainly not given the time of day on most SEO blogs, yet we saw a definite increase in rankings on competitive terms. In the vertical we had a newer, less linked to site – there was absolutely no movement in either direction for that site. Our testing on authority sites has shown us that you can send almost any type of link and get some benefit, either in rankings boost on a specific keyword or a larger net for long tail keywords. Yet if you tried to rank a new site using the same tactics that clearly work on an old, trusted crusty site, there’s a very good chance the new site would at the very least be filtered, and at the top end of the spectrum be penalized. Of course, defining what is an authority site is another issue – I suggest you go out and test what exactly is an authority site, and reach your own conclusions.

The Bottom Line

You need to be actively running tests and making efforts to build your business and your sites. The only data that you should trust is your own. While it’s good to have an idea of what’s going on in the larger SEO community, what really matters is your rankings. Everything else is, and should remain, secondary.

Ari Ozick is CEO of Wired Rhino and occasionally blogs at He would love any constructive feedback or questions you have, either in the comments or direct via email: first name @

Published: October 9, 2009 by Aaron Wall in seo tips


October 9, 2009 - 12:15pm

I followed the link to his/your blog, and must say I found the navigation/usability of your blog too confusing to stay there long enough. I tried clicking the categories (SEO, blogging, and another one) at the right side of the bottom of the page, but that only ended up showing me one head line (I expected to have all posts in that category listed).

Then again Im not too knowledgeable about general PC stuff, and that might have happened because I have about 30 browser windows opened right now? (dont think so but u never know)

I also think the headline of this post is a little bit overdone & not too descriptive of the acutal post.

Generally I agree with your ideas. I have friends who have been betting on sports games (legal in Germany) for years...and...simply accept the idea that their familiarity with those sports helps them "earn money on the side". Naturally they laughed at me when I started doing the same and tracked my results using the minimum bets =). Anyway, I agree that testing is really important.

Im also considering checking out the two tools mentioned...Ive been a bit afraid that yahoo might take their site explorer offline before I can grab the data I would need from it (Im in college and wont really have time to do that before the semester ends).

Is any of those 2 sites good for linkage in that it shows you as many links as yahoo's site explorer currently does? Not sure if I understood it correctly (went to their sites to check ito ut briefly), but it seems that they offer linkage data in a similar way that the yahoo site explorer does, right?


October 9, 2009 - 12:58pm

I can't agree enough with "the numbers don't lie." Too many people make assumptions about what is the most profitable for them. At the end of the day we're all in the same business. The business of arithmetic.

@Patrick: Correct me if I'm wrong Ari but that's a fairly new blog and that's why there's not many posts on it. The one article you saw when you clicked on the categories is because there's only one there. As far as the usability goes I don't understand that either. That's the standard Thesis format. Categories and blogroll.

Dave Dugdale
October 9, 2009 - 1:25pm

This is the first SEO post in a long time that actually shared test data they have done. I wish more SEO blogs would get back to the old days where people share test data like this.


October 9, 2009 - 3:04pm

CureDream never got into Twitter ever since I've been keeping a low profile; I think I was the first japanese superheroine investigated by the FBI.

Another character I play is on Twitter and he's found that that you do get some weird connections there. A venture capital funded competitor of mine tried to hire him: with a star-studded collection of famous C-level managers, $20 million in funding, and three years of 'stealth mode' development, they're getting less traffic and revenue than he's getting with 1.25 employees (not even FTE), a year of development, and maybe $3k in investment.

Now, they'd probably pay him a salary that's more than his project is making now, but he's got some ethics. He thinks business is about making money from customers, not ripping off a bunch of VC's. He's a software jockey, but he's excited about a unique business model that blows the door off Web 2.0. He'd rather have a chance to succeed or fail at that than go work for a bunch of time servers.

October 9, 2009 - 3:21pm

Great post, Ari, and it's amazing how many "search experts" fail to live by this mantra. A little scientific method can go a long way.

And now I'm off to track you down on Twitter...

; )

Tom McCracken
October 9, 2009 - 4:21pm

Good post, I'm a big fan of testing, it was refreshing to read a post like this.

Although I do think Patrick made a good point, change the tittle, remove the graph and first 3 paragraphs and this post is perfect.

October 9, 2009 - 6:50pm

I saw the chart before I read and sat there thinking "BULL!" and then started reading on and couldn't stop agreeing. I treat SEO like horsepower, no promises, just straight performance numbers. Great work dispelling the "he said she said" myths. I think there are too many people blindly following what others blog, post, etc and not enough independent thought. Everyone who tells you something has something to gain from telling you. Or in another form; Do you trust your competitors to give you the secrets to their success?

October 9, 2009 - 6:52pm

Ahh my bad! Thanks survivalguidedotnet & sorry @Ari...I was a bit in a rush, so please forget what I said about the usability of your site (probably hard to tell by my long post, but I just like to type a lot, I guess lol)

October 10, 2009 - 1:06am

"This is the first SEO post in a long time that actually shared test data they have done..."

Comments are very entertaining.

October 10, 2009 - 2:34pm

Constructive Criticism, remember :-P

October 16, 2009 - 4:32am

This post reeks of awesome. Great job, Ari. :)

Add new comment

(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.