Exclusive Lunarpages Interview with Rand Fishkin of SEOmoz.org

Author: Joe Whyte

When I wrote to Rand Fishkin I didn’t expect a response let alone such a detailed interview. I ran into Amy’s office like I just had won the lottery “OMG Rand replied!”

She stared at me blankly and said “Who is Rand?”

I almost fell to the floor. Let me tell you who Rand is..

Rand is the founder and CEO of SEOmoz, a search marketing consulting firm based in Seattle, WA. Rand’s best known for his work on the SEOmoz.org website, a resource that serves tens of thousands each month in the field of SEO/M. Like all good businessmen, Rand dropped out of the Univ. of Washington’s business school to pursue his dream of running a small company and has never looked back.

1. As a professional Search Engine Marketer, what is the FIRST thing you look for when YOU are considering a hosting company? What are the MUST have features?

• 24-hour customer service via phone to someone who can work directly on the server. I’m happy to pay for the time if they need to reboot the box or trouble-shoot or diagnose problems, but uptime is the biggest issue – going down when your site is getting hammered is a terrible problem, because it’s the time when you have the best opportunity to pick up new links. Your 15 minutes of fame on the web (whether via a site like Digg or Slashdot or through other media) only lasts 15 minutes and if you’re server can’t support the traffic, you’ve wasted much of your marketing efforts

• Notification of downtime in advance. If the box or network will be down for 15 minutes late on a Sunday night, I need to know on Friday morning or before.

• Root access – for my staff, it’s critical that they can access the root to install kernels, updates, etc.

• Hardware quality – running lots of tools and applications means eating up a lot of RAM and processing power; the box has to be able to undergo some serious tasks while interfacing with databases, fetching and delivering web pages.

• Geography – believe it or not, with developers, being within 500-1000 miles of our Seattle, WA office actually makes a difference in the time delay during site creation and testing. A few extra 10ths of a second on every upload, save and test can save hours (and developer frustration, as they’re notoriously impatient) over the course of a development timeline.

• Relationship – I need to have a hosting company that knows us personally, answers emails quickly and has staff who recognize our site and our names. The personal connection is what influenced my last two hosting moves and I haven’t been disappointed. Once we had VPs and CEOs who I’d met in person (and whose cellphones were in my cardfile), the quality of service we received was incomparable.

2. What is more important when considering a web host? Customer Service, Reliability or Reputation?

I’d put customer service first, reliability second and reputation last. A bad reputation might be from a few particularly noisy customers who had a bad experience, rather than a true indication of quality. Reliability is very important, but I’d say it’s outweighed by customer service – being in touch and available is the biggest priority among the three to my mind.

3. Do you prefer Windows or Linux hosting and why?

We’ve always used Linux because my developers are familiar with it. I do like the fact that Linux has such a huge development community toiling away to make it better for free. Windows is probably better for some enterprise-level organizations, but for us, saving the extra licensing fees makes Linux the obvious choice.

4. When you first started doing SEO do you remember what type of hosting account you started with and what type of hosting account would you recommend to new SEO/SEM’s and Internet Marketers?

It’s really more about the type of site you’re running, the traffic levels you expect to get and the level of customer service you need. If you have a web-based business in the affiliate sphere and are seeking out small levels of profit, high-price, high-level service hosting doesn’t make great sense. There are plenty of packages for under $50/month that can work (with a managed service and control panel). Once you have a larger, more important, business critical site, it pays to have your own box and a high level of customer service. We have fantastic plans that are around $150-200 per month, and for such a small extra fee, it really makes sense for almost every non-entry level business to upgrade.

5. Can you give a little SEO advice to our 100,000 clients about how they can get better results?

The best SEO advice I can give is to tell folks to do a bit of homework on the field. You can learn a lot of the basics by visiting SEOmoz beginner’s guide (we’ll be releasing an updated version in the next 30-40 days). If you don’t have time for even that, I’d recommend you at least pay attention to the basics of SEO, which are:

• Research your keywords – use Google, Yahoo!, Wordtracker and others to see what people are searching for in your field, and incorporate those keywords in your title tags, meta description tags and H1 tags.

• Ignore keyword density – the search engines haven’t used it in years and it always looks like spam when sites cram keyword after keyword onto a page

• Make friendly, crawlable URLs – see 11 best practices for URLs

• Make sure that you target only one or two keywords per page – you don’t want to compete against your own site in the listings.

• Create a great looking site – the kind that people love to visit, are impressed by and show off to their friends; this kind of viral marketing will have a remarkable impact on your SEO efforts.

• As with looks, go for functionality, too – high quality content that people want to read, tools they want to use, anything that makes a person jump into their email and tell 3 friends about your site is what you’re looking for.

• Don’t ignore usability – read Steve Krug’s book; Don’t Make Me Think. It should be mandatory reading for everyone who develops on the web.

There’s a ton of more intricate, complicated parts to the practice of SEO, but starting with the basics will give you, at least, a solid footing.

Do you think Digg.com will officially ban SEO related articles and domain names?

Digg probably won’t “officially” ban any type of content (except adult-themed material), but they will certainly ban domain names (I wouldn’t be surprised to see SEOmoz go on their list, as lots of our readers submit our blog posts and articles). Digg is, by nature, anti-marketing and very anti-SEO. They think that everyone who calls themselves and SEO is out to spam them, and with all the evidence they have, it’s going to be tough for those of us who are legitimate to convince them otherwise.

Do you think Wiki search will be a strong competitor for social search in 2007?

No – wiki search will not be a strong contender. I really don’t think people realize how incredibly hard it is to build a great search engine. Microsoft has incredible resources, very, very smart people and a fiery drive to get up to speed, but even after two years, they’re way behind Google. It’s not about the approach, either, Google (and to a slightly lesser degree, Yahoo!) really have done almost unbelievable things to get where they are in terms of providing a good search experience.

Do you like Web Positions Gold or WebCEO?

Nope. I build my own stuff to get data from the engines – it’s far more customizable and valuable for anyone who’s a professional in the space. I also found that the accuracy of the programs was frequently intolerably poor.

What is the best way to draw traffic for 2007?

It will probably continue to be the major social web portals and news aggregators. It’s hard to beat 25,000 visits from Digg, 30,000 from Slashdot, 10,000 from TechCrunch and a few thousand more from del.icio.us and reddit. Many companies who’ve launched this year have done exactly that, and in my opinion, it remains the best tactic.

How do you feel about foreign search engine optimization?

Google has said it’s cracking down on web spam in Europe, but for now, the kind of spammy link building tactics that were effective 4-5 years ago in the US are still highly effective in most non-english language engines. In the next 4-5 years, I expect these markets to mature, search to become more popular, and spam to become less effective. The trends will most likely mirror a lot of what has happened in the US, with some regional quirks. Asia in particular will be very interesting, as places like Taiwan and Korea have a chance to use search far, far differently than what the English-language world has done (more social, more web2.0, more interactive and with far more user options).

Thanks Rand!