Rand FishkinA.M: Sooner or later we believe, another concept will emerge in the content marketing and that is, content pollution or content inflation. What do you think about these concepts ?

R.F: I’m sold on the idea of an overwhelming amount of content that currently floods people’s social streams, subscriptions, and email. I wrote about this last year, noting that content marketers could become their own worst enemy. In the future, I expect the value of content marketing to primarily come from the few pieces and to the few sites that can stand out from the crowd and attract their audiences with remarkable, unique work, rather than those that simply publish regularly because they’ve heard it’s a good marketing tactic.

A.M: There are more and more players using black hat techniques using crowdsourcing. How do you think Google should address this bad issue?

R.F: I’ve seen some click farms and manipulative stuff in the realm of behavioral data, but from what I’ve seen it doesn’t last long, and Google’s doing a fairly good job of shutting it down by focusing on behavioral data they can verify. For example, when I did some experimentation in 2014 around queries & clicks as a way to drive up rankings, I noted that over the course of just the next 6 months, that became much more challenging.

If I were at Google, I’d be using pattern matching and machine learning to identify queries and clicks that come from real sources vs. manipulative efforts. Given the right inputs, I imagine this wouldn’t be terribly challenging and as such, I don’t expect it to work as a tactic for very long.

A.M: Almost the same as crowdsourcing, there are platforms enrolling bloggers to spread content and links. How will Google stop this practice? There is the case of PayPerPost, but that case actually made more platforms to appear.

R.F: They’ve been shutting these down pretty fast, and while new ones continually spring up, the vast majority are better at causing penalties over the long run than helping real sites achieve traction and marketing success. For the churn-and-burn spammers, this sort of thing will always exist in some form, but for real businesses (which Google is doing a better and better job of recognizing and rewarding)

A.M: What are the hardware requirements to build an engine to crawl the entire web? We know that Moz upgraded their hardware, we even saw an extensive story about this  and we loved the transparency offered. However, for a minimal scan, because there are many who dream to do it, what would be the minimum costs?

R.F: It would be very hard to imagine without, at minimum, $1-2mm of investment. Alongside the bandwidth, storage, and serving requirements, you’ll need an impressive team of software engineers (at least 2-3), and folks who are capable of building something truly useful and usable command very high salaries.

At Moz, we’ve been rebuilding our own index to be closer to Google scale, freshness, and features. I expect at the end of the process, we’ll have spent ~3 years, a team of 6-7 engineers (plus input from product folks like myself), and $5-7mm in hardware.

A.M: What are the yearly increases in websites from year to year ? How many new websites are available when Moz is “scanning” the web? What is the increase on a global scale?

 R.F: Hard to say, because the decay rate is also great and there’s a very large number of sites on the web that aren’t what we’d consider “active” or “real” (that includes things liked parked domains, one-page sites with almost no real content, and lots of spam/junk). We find a few million new domains each month, but what percent of those are “real” is a tough question to answer. In the future, I hope to be able to provide a better reply!

A.M: What are the hot niches, where there is not so much SEO and it is easy to rank a new online business. Of course all of the niches are crowded, but still, what are the niches where a SEO can have better luck ?

R.F: Ha! I wish I had a list, but that’s actually not something I maintain or look into. In general, though, fields where there are fewer technology-savvy players (skewing toward older, less-male, less-wealthy, less-first-world demographics) have an easier time.

A.M: You have OSE, therefore a ton of data about the whole web. What insights can you give to our readers about the trends?

R.F: Well, we’ve seen a great deal of adoption of search standards like rel=canonical, robots.txt, webmaster tools use, etc. over the last few years, suggesting that site owners and operators are getting savvier about SEO in general. We’ve also noted large increases in non-.com domains, but haven’t seen tremendous pickup (yet) of the new gTLDs anyone can register (e.g. “.blackfriday” or “.coffee”).

A.M: We’ve seen that SEOmoz dropped “SEO” from the name and became Moz. Also, MajesticSEO dropped (again) “SEO” from their brand. Why is that? Is SEO no longer important?

R.F: Definitely not. In fact, the opposite is true. I think both companies – us and Majestic – simply wanted the opportunity to have our brands grow to include marketing fields and data needs outside of the purely SEO field, hence the branding changes.

A.M: We want to make an interview about global SEO with Matt Cuts from Google. Give us a question for him.

R.F: *****************************************************************************************************************************