I think the first thing any startup should know about SEO is that due to the ease with which someone can add SEO to their qualifications we currently have a world where every second person is an SEO expect. If your going to outsource your SEO(and its probably not a core competency so you probably should) make sure you spend a fair bit of time investigating any potential consultants. Also speak with them, discuss and then research potential strategies they suggest. If it sounds dodgy, Google probably agrees and you will be risking a lot for a small chance of gain.
I would not outsource something that could account for 70 percent of my traffic. SEO is hard and there are few companies that can help you rise above the competition.
Another one: if you're writing a bunch of beautiful linkbait articles like Priceonomics make sure to host your blog at priceonomics.com/blog not blog.priceonomics.com so your primary domain (that has all the real content) benefits from the linkjuice. Google treats subdomains as essentially entirely different domains.
This is incorrect. Google treats each page as a different page (even on the same subdomain). PageRank is per page and not per subdomain.
If anyone wants to learn most things about SEO, I highly recommend "Getting to Know SEO" by Andre Kibbe, it's the best book I've read so far on the matter: http://rockablepress.com/books/getting-to-know-seo
The published PageRank algorithm is not the last word on Google rankings. GP is correct. If you do not know this after reading a book on SEO, strongly reconsider whether that book was adequate to your needs.
GP said: "Google treats subdomains as essentially entirely different domains." This is false. If a domain has a small number of subdomains then their authority is shared thus making it false that priceonomics.com would benefit from using a folder instead of a subdomain for their blog, their only subdomain. Fell free to recommend me a book that says and proves otherwise.
You could argue it would make sense to keep their blog as a folder like they do for /jobs and /about for coherency. I would agree.
You're sort of moving the target here. I care very little about scoring Internet points, but I care very much about HNers not making decisions which will damage their business, so let's reiterate: it is categorically not the case that Google does not have a notion of domains. Page A and Page B, given that they are on the same domain, can easily influence each other's rankings even without a link passed between A and B, by separately garnering trust for the domain. This informs several SEO strategies which will work such as e.g. publishing OSS on your domain to get link juice and other indicia of trust which you then use to rank things that real people (and your bottom line) actually care about.
As to whether blog.example.com and www.example.com share trust metrics, sentiment among SEOs are mixed. Consensus estimate among SEOs who I'd trust with my business would be close to "blog.example.com and www.example.com are treated as at half-arms lengths for the purposes of accumulating link juice and virtually identically for the purpose of awarding link juice." This suggests that you want as few subdomains as reasonably possible for publicly visible content which will either attract links or which you want to rank. For similar reasons, I'd suggest most startups would require a darn good reason to have more than one toplevel domain.
Google has said not to worry about this issue, but that a very common answer from Google when probed about any specifics of their algorithms. Google will also say things like "Quality content is 99% of SEO" which if it is true is true in the sense that Santa Claus is true -- perhaps a good message to tell children but I would not base my ability to feed my employees' family on the likelihood of finding a jolly fat guy if I went to the North Pole to check.
I can't provide a cite for this right now, but I'm reasonably sure I've seen SEOMoz staff suggesting that it's a bad idea to keep your blog on a subdomain, for exactly the reasons Patio11 suggests.
(Also, whilst he's being modest here, Patio11 is something of an authority on SEO himself.)
So I moved my blog.mintek.com to mintek.com/blog after using HubSpot and switching to a custom wordpress install.
I can't tell you that all the advantage was from the url restructuring, but it had a marked impact. Having tried similar things on test sites, I do think it makes a difference.
And that is what SEO is about… testing assumptions because there is no definitive answer. Even Matt Cutts isn't going to tell you everything 100%.
Matt Cutts seems like a good guy, but there is an unavoidable conflict of interest when he gives advice like this. Google's job would be much easier if everyone believed the only way to rank is to have popular, useful content.
There's some matrix-style SEO going on in this post. Not only did Priceconomics.com get great keyword-rich anchor text links from that guest-post article, but they are also driving links to the article, creating more SEO value for the links that they got. Nice work! Watch and learn people...
"So how do you get good inbound links? Well, the key to really nailing SEO is producing such great content that other sites on the web can’t help but link to you."
And if you're a local plumber who, other than a lack of incoming links, provides great service? Social mentions of that great service (nofollowed on Twitter, for example) aren't going to help the ranking.
With startups and side projects, isn't the end result of this sort of thing that people have to create blogs/guides/content that is outside their core focus?
My impression is local plumbers don't care about SEO. They don't care about scale.
They care about making money. They don't need SEO for it, they need to offer a good service.
Links with a no follow attribute do have some value, it won't do anything for PageRank but it's social proof. I haven't done the test but others did; it pushes someone higher in the search rankings. Hundreds of no follow links were needed. I'm not sure that's always the case. Let's hope not because my local plumber is never going to get that. Nor does he care.
That being said, I think Google has its own internal Klout system.
"With startups and side projects, isn't the end result of this sort of thing that people have to create blogs/guides/content that is outside their core focus?"
You're right. But it has to somehow relate to what your start up is doing, otherwise it won't get you a lot of traffic, nor will it do a lot to push you higher in the search rankings.
My point with that last line was that it can just distract from what we should all be working on. For small clients of mine, it's not a successful pitch (spend hours writing a blog, it might help with Google) and can often favour directories and so on.
Knowing a fair number of local tradesmen and professionals with a keen interest in marketing via the internet, I've observed the answer to be paid traffic.
If you're genuinely good at a much-needed trade, ROI-positive conversion funnels for tightly-optimised keywords are much easier to set up than for most Web entrepreneurs.
The author lays out the basics well and anyone building a business on the internet should know at least this much. With these practices in place when you need to hire a professional for SEO optimization you will already be half way there.
I find the hardest thing about outsourcing SEO is that it really does start with the content, and you shouldn't outsource that. It is important to identify your keywords and focus on using them properly when writing this content. Good content, with a fine balance of writing for bots and humans, is very important. Good SEO copy is not the same as how we were taught to write an essay in college (or High School.) It should not be as verbose and takes a little getting used to writing in that style. Use keywords like you would salt, enough but not too much, and in everything.
The author makes some easy to implement suggestions. However perfecting these methods takes time and skill. Growing traffic organically through SEO alone is tough. Kudos to those that do it well. However search engines are unforgiving on seo abuses.
Though these are solid points I disagree writing compelling content is enough to garner links. First for some niches its rare for people to link to you, second by default you are a nobody.. Unless you already got tons of influencers who are subscribers or followers, most compelling content end up in a black hole.
I know google tells you differently but in most cases you have to be active and pimp your content. Like what Priconomics is doing here.
This article was written for bots just as much as it was for humans. Check out the relationship of the links and keywords. The string "price guide" is found 10 times, and always as a direct anchor text or near a link to the author's site.
That's not a bad thing, these guys are doing it right. Listen to their advice.
A lot of SEO knowledge is about knowing what not to do. Don't create copies of the same content across multiple pages, and if you do, choose a canonical version. Don't exclude important directories in robots.txt, but do exclude wasteful pages like calendar pages (often there is a new page every day from Jan 1, 1970 forward).
If you're a Jim's Plumbing site, you should be okay following your advice. If you're a giant site with many pages to index, you need to focus on structure. The flow of pages is important, and it's more than setting up breadcrumbs and pretty URLs. This kind of problem needs to be looked at when building a site, you can't just drop some code in after you've launched.
