Hacker News new | comments | show | ask | jobs | submit login
What Every Programmer Should Know About SEO (katemats.com)
120 points by kateroth on Mar 5, 2012 | hide | past | web | favorite | 31 comments

Good article, I wanted to write something about it too(web developer with strong interest in SEO here). I'd say in the end it's more an introduction to SEO than it is advice for programmers about SEO. I'll detail 2 things she overlooked in my opinion.

For example, web developers have a fantastic opportunity to build links by getting involved in open source communities. These are extremely link-rich environments and by posting tutorials or creating small packages, you can find yourself with a lot of links pointing to your site. They will maybe not be relevant to your target keywords but will still be links from high-authority sources.

And I'd say the traditional way of garnering links still applies to web developers: get to know fellow developers, get interviewed on their blogs, exchange resources etc. So hang out where your community lives, whether it's on Reddit, StackOverflow, Hacker News etc. Get to know people, the SEO will follow.

This only works if your site has something to do with development.

Or are you suggesting posting development related topics to a non development related site purely to get those juicy backlinks (and thus good pagerank)?

No, not only. You can host software packages and tutorials on your product website.

For example, patio11 of HN fame has an article about Rails SEO Tips on his product website http://www.bingocardcreator.com/articles/rails-seo-tips.htm

Even though I'd say that he could have done a better job at integrating it on his website (it looks too much like an isolated page in my opinion - there should be a "Tips for developers" section), it does the job at attracting links and boosting SEO for his main site (also with A/Bingo, a software package he developed: http://www.bingocardcreator.com/abingo).

Actually he developed a blueprint of what could be done for startups regarding SEO: http://www.kalzumeus.com/2010/01/24/startup-seo/

And here you go, he got three more links for free :-)


That's sort of what I mean. For my work I build and run a number of websites which are almost completely unrelated to tech. These are service industry websites so there is no "product" to speak of.

I am looking for ways to improve pagerank, so let's say I was to write a few pages about how I built some of the trickier parts of these websites and hosted these on the websites themselves but perhaps did not link these through from the main pages (but did link back to the main pages from my new pages).

If I got links to my articles on reddit , HN etc then perhaps this would improve pagerank even though the pages are unrelated to the sites main content. So this would improve the search rankings for the websites on their main keywords as well?

Yeah, that's the idea. However personnally I would link from the home page ("Inside the business" chronicles). It doesn't have to be a big link, could be in the footer for example. I don't have metrics for that but I think pages that are not linked from anywhere on your website might look a bit shady in Google's eyes.

By getting links to your articles you'd improve overall authority of your site (also important to link back to your homepage and other important pages in order to drive PR back, as you point out).

By improving your website's authority you'll rank better for your main keywords as well.

Ok, you really don't need to know everything about SEO to do a non-terrible job of ranking. Do these things and you can get traffic with SEO.

1. Title tags that get clicks (not kw stuffing or plain exact match)

2. Write content that is unique and semi-frequent. (update once a week or so at least)

3. Acquire links.

#1 and #2 are the basics that you need to do to even have a shot to show up for a given query. #3 is what is going to get you on the first page.

You can do #3 by having great content and doing social promotion, using email newsletters to blast your content out to people so they can share it, or if you're in a boring industry and nobody likes you, you can buy or build links via the numerous link building services out there that can hit any number of link sources.

At the end of the day, natural links are viewed as better, but paid links can get the job done in some cases.

Given a certain bar of content quality, the quantity and volume of links is still the most important ranking factor on Google and Bing as far as I can tell.

I despise SEO, but oddly enough I actually agree with what the article says.

Possibly this is more for web-programmers than others, but these days that's quite a lot of people, and what the article says is succinct and to the point.

What you have to realize is that SEO is a logarithmic curve; when you've done none, doing a little bit makes a huge difference. When you've already got a decent site, well... a lot of effort for virtually no gain.

Everything except "Building Links" I would call UX/Accessibility if only to avoid any semblance of the shadier parts of SEO.

"Typical link building [...] can be through deals, partnerships, PR pitches, link exchanges – even paying for links."

A few of these techniques - link exchanges and buying links - sound like really bad advice, especially given that she is a former SEOmoz employee. She also omits the most straightforward way for startups to get backlinks => create content on your site worth linking to.

I've found that people hate to hear that when asking for advice on how to get traffic. It's even less popular than "eat right and exercise" as weight loss advice.

A lot of people take a more directly cynical view of content creation, rather than the more passive ignorance we have with regard to health. Our health is not something someone else can take generally, content is.

It is too easy to look at the rampant copying and dirty tricks used to pull in viewers for ad revenue and think "if you can't beat them, join them" or worse "I might be able to beat them, but joining them seems easier".

Mainly because unlike eating healthily it's actually a pretty broken concept.

Your search ranking is determined by your ability as a writer rather than the quality of your product.

Actually, a lot of those techniques still depends on having great content. From experience, many will be reluctant to link to a site that doesn't look the part, even if they get paid for it.

The thing is: who you link to also matters and can harm your website if not done correctly. Linking to a site that is full of scraped content, AdSense and with a poor design is a bad idea, for instance. When you place a link to a site, you become associated with that site.

Regardless of your tactic, you still need to put the work in to build a good looking site with great content or you won't get far. In fact, if your prospective target is willing to link to a shoddy site, I would treat this as very suspect. Chances are the link won't be worth anything.

As an example the easiest way to get paid links is through directory sites (the cheap ones) and you have to be careful there, a lot will not benefit you at all if not to a detriment.

To be honest, just having worthy content is almost cliche. I prefer to think of it as a foundation to build on with a variety of strategies and tactics. For example, you write a great blog post (the foundation) and then go and post it on your twitter, HN and/or reddit (your additional link and traffic building strategies). Without the additional work, it becomes much like a "build it and they will come" mentality and the only time this could actually work is if you were already at the top of your game.

That is a great point - and I definitely wasn't encouraging buying links. I updated the post with your advice.

s/programmer/web developer

And even then it's a stretch.

Not sure why the author seems to assume that every programmer is even remotely involved in webby stuff.

Site freshness only matters for certain categories/topics. For other topics you could say the older the static content the better. Some of my best ranking/money making pages have not changed in 8 years and rank better every year, without doing anything.

Regarding ranking algos, you did not mention the Trust Rank (there is a patent G filed some years ago). A low Trust Rank is equally bad as a low Page Rank. I don't really know much about it, but afaik it has to do with the trust that G has collected on your site, while Page Rank is based on individual pages. If you link to "bad neigborhoods" (another important term, think casino sites, v*agra, etc.) your TR will go down for example, and its hard to get it back up. If National Geographic links to one of your pages, your site's TR goes way up.

And then there is the number of pages on your site. Google will only crwal a limited amount of pages from your site, depending on your Page Rank and distribution of incoming links. So try to keep the number of crawlable pages focused to the relevant pages (ie the ones you'd like to see in G's index).

After some time and if you have tons of links, you will see a "deep crawl" with a bunch of Gbots hammerings you server and downloading many more pages a day than usual.

PLEASE DO NOT READ THAT ARTICLE - it is bad for you and your business.

SEO is a bullshit industry. this means that 80% of everything that is written about SEO actually reduces your knowledge of SEO, that article is in no way different.

just pick a few examples:

    > A web crawler has to crawl your site (and every other site on the web)
   > Indexes all the pages ..
   > Organizes the indexed collection and stores ...
this one misses on of the most important steps of the google site processing pipe: Discovery is the process of discovering URLs to crawl. It is fueled by the sitemap.xml and crawling.

PageRank - is not - "another important part of authority" - pagerank is thoughtcancer. don't ever say that word, don't ever think that word. see my TC article here http://techcrunch.com/2010/07/07/startups-linking-to-your-co...

    >Ensure pages render without JavaScript enabled
sadly google parses and to some degree renders JS ... badly. this tip is good, but leads to the false conclusion that google does not parse/render/"see" JS.

    > you should also pay attention to how you construct links and anchor text to internal pages on your site.
internal link anchor text is mostly irrelevant.

    >your own internal links use relevant keywords in the anchor text.
it's also some of the worst things, as the impact in minimal to not measurable.

    >So if you load a bunch of text dynamically with Ajax, or JavaScript, create a non-JavaScript version that will show the same information.
outdated, google now even does onload-triggered POST requests if necessary. js is still bad, but the story above is only half the story.

    >Put them in the URL (and even better if your domain name has your keywords)
oh god, not again. these are the SEO URL rules

    #### landingpage url rules

    all URLs on an a webproperty must be according to these rules
    listed in priority!

    1. unique (1 URL == 1 resource, 1 resource == 1 URL)
    2. permanent (they do not change, no dependencies)
    3. manageable (1 logic per site section, no complicated exceptions)
    4. easily scalable logic
    5. short
    6. with a targeted keyword phrase
further down she says something about 404s and it sounds like they are bad. 404 are great. when pages are gone, they are gone - 404 or 410. if the move then 301. but do not redirect gone pages to friendly 200 pages. keep your site simple and your redirects rules as well.

    >Additionally you want as many links to contain appropriate alt text.
links don't have alt text, images have alt text. in an image is linked, that alt text substitutes the anchor text.

    >But SEO isn’t just about building a great site that is crawlable and has the right content – it is also about converting users.
no, that is the job of the product, onpage marketing, conversion optimization. SEO is really just about getting - targeted - visitors. (she has clearly read too much SEOmoz)

    >the most defensible SEO strategy is links,
no, it's not. that is just something that SEO agencies tell you to sell you their services and useless tools.

disclaimer: i work as an SEO - i hate SEOs. my biggest success is 62 M visitors a month. i don't read SEO blogs.

You just took five minutes of my time telling me how bad something was and not one second offering an alternative. For all the bad things your critique basically just boiled down to "not it's not" without offering proof or links.

Do you see something wrong with this picture? As a reader all I've got from your comment is angry guy doesn't like SEO and thinks all SEO blogging is for shit.

I apologize for being so blunt, but this is not a very useful comment. Perhaps the goal here is to establish how much you know, but even then it really fails to convince the reader. It'd been great if it had been a bit more constructive. Obviously you believe you know what you're talking about.

I second that. Pointing out what a good SEO strategy is and how it opposes the points described in the article would have been helpful.

If somebody interested in improving their SEO skills should not read well-known SEO blogs then what does the OP suggest?

I personally follow SEOMOz and a few established SEO and link building blogs, but would be interested in hearing other suggestions.

i'm telling you the same thing i told the CEO of an international publishing house: read a book instead!

Search Engine Marketing Incorporated

http://www.amazon.com/Search-Engine-Marketing-Inc-Companys/d... [not an affiliate link]

sadly, nobody ever does.

You're telling us to read a 4 year old book on technology that changes on a monthly basis?

Lots of SEO involves guess work. Google changes its algorithm constantly.

That's the reason nobody reads books on SEO. It's the same for web development.

>You're telling us to read a 4 year old book


>Lots of SEO involves guess work.


everything is measurable, if you can't measure it, don't do it.

>on technology that changes on a monthly basis?

if you don't have a model of how SEO works, you can not improve on this model, even with new information. that books provides a - sustainable, working, applicable, employable - model.

>That's the reason nobody reads books on SEO. It's the same for web development.

... (words fail me)

Not sure why this article offended you so much. It was all rehashed info but it wasn't that bad.

Just because Google can somewhat crawl and parse certain JavaScript doesn't mean it's good for SEO. I think you would be doing a disservice to your company to act like you can just load your site with client-side loading content and act like Google is going to retain your content. I've countless anecdotal examples of content hidden behind JavaScript not being found by search engines. In fact, it's a celebration whenever Google announces they managed to crawl a common piece of JS (Disqus, FaceBook)

Internal links are not irrelevant...What on earth would make you think this? It's not hard to prove that internal anchor text passes the relevance to the linked-to page by using a made up word to link to an otherwise orphaned' page to rank.

Regarding 404's....Let's pretend I have a great piece of content that has amassed hundreds of links. Marketing has deemed that the content is no longer worthy of being present on the site and has removed the page without consulting you. You have the option of letting it sit there dead with hundreds of links pointing to it, or you can 301 redirect it somewhere to pass the amassed link equity. Yeah, you go ahead and keep your 404.

Finally, links are just something that agencies tell you to sell their services??? Are you suggesting that gaining good links will do nothing for your site's natural search traffic? Ok. Sure.

ad ">Just because Google can somewhat crawl" see my response above "this tip is good, but leads to the false conclusion that google does not parse/render/"see" JS."

for the other points: i have a very special niche, i deal with companies that go through 1,2,3,...x SEO consultant, agencies, companies that spend an unbelievable amount of resources, dev time, money just to get SEO right - and their traffic still sucks. they have the most optimized internal link anchor text you can imagine, tens of thousands hardcoded redirects (no joke, i have seen the .htaccess) to "funnel that linkjuice to the most valuable page" and spend a huge monthly budget for linkbuilding and the most "sophisticated SEO tools" you can imagine. their URLs were just "keyword optimized" and now they are working with a company "to keyword optimize our content". after five sentences with their CEO he starts talking about PageRank.

i have to cut through all of that bullshit and establish a manageable, scaleable site that consists of

  * a startpage
  * sitemap.xml
  * (interlinked) landingpages
and sadly from a SEO point of view: yes, you need links - but not the links SEO agencies tell you you need, and not with the methods SEO blogs recommend.


I don't think you read SEOmoz with statements like that.

"disclaimer: i work as an SEO - i hate SEOs. my biggest success is 62 M visitors a month. i don't read SEO blogs."

Blogs are a great resource (especially for SEO). I would also be interested to see what your spending budget is (with enough money, it doesn't take that much skill to bring lots of users to your site) and how you are getting traffic to your sites.

If it's all organic (IE: not paid), then I will be impressed.

i only do organic/unpaid unbranded. i mostly only work with the companies internal resources.

>Blogs are a great resource (especially for SEO).

blogs yes, SEO blogs, no (with about 80% probability) - if you want to justify a bad decision that did cost your company thousands and thousands of revenue, just google it. there is a very high possibility that a ("high reputable") SEO blog recommended it at one point.

The article is very good. I actually coded a project where my entire goal was to reach number one on google for the search term "male female translation." It took about a month but I eventually made it to the top following most of what was mentioned in the article. I just had to figure it all out via research, trial and error.

Honestly, I would expect any experienced web dev to have this knowledge as part of what they bring to the table, especially for a front end dev.

The article is not really stating anything new and there is already any number of similar articles out there basically documenting the same thing. In fact, given her past job experience as VP Eng for SEOmoz, I'm a little surprised at how little insight there is in the post. (Honest opinion, not trying to be snarky).

Getting links is still the most important part of SEO. She barely mentions this.

It's also the part that is least relevant to the developers who are actually building sites, which is who this article is targeted at.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact