Reading over Digg, as I sometimes do, I found this submission on Building a Niche Minisite. The first few comments imply something scary to me.

…it’s SEO, which are the scum of the earth, just behind spammers.

[Above] comment is over generalized but SEO’s are known to hurt the internet experience time and time again. If SE’s didn’t have to fight SEO’s just trying to trick people into making them a quick buck imagine how much more useful a simple google search would become

[Note: I bookmarked that and wrote some notes a while back.  The comments I refer to seem to have been removed since then.]

Reminds me of a similar situation, going back a few years, concerning “hacker” vs “cracker”. Originally, hackers were those who hacked software and systems together. They were curious, inventive – not evil. But its a very negative label nowadays, with the inference that hackers are out to cause damage. When this trend started, old-school hackers tried to get the label ‘cracker’ applied to their more malicious counterparts, but failed to do so. Media reports were always saying “Computer hackers broke into such and such”, and the war was over.

I think those of us in the web industry are facing the same battle. To be a Search Engine Optimizer seems to be the same as being a dirty rotten spammer in some eyes. To my mind, this is just plain wrong. The true search engine optimizer works on the following principles:

• Ensure the site is crawlable by engines (missing pages, broken links, javascript menus – all of which are bad for other reasons too, such as limiting a visually-impaired user with a screen-reader from using the site)
• Getting a good balance with respect to keyword density on pages (generally meaning to remove the non-essential stuff from an article – always a good writing tip anyway! Remember English 101? Tell the reader what you are going to say, then say it, then tell them what you just said.)
• Setting relevant page titles, putting headings in the right places (Titles help when users bookmark pages, and headings are great for breaking up content into readable and digestible chunks)
• Making sure that the site has plenty of internal links (when you link from one page to another – good for your users if it helps them find related information on your site about the topic of interest)
• Getting inbound links from other sites (Helps promote your content so that others can find your pages. If its valuable content, there is nothing wrong with helping people find it!)

Some “black hat” SEO’s will ugly things, like comment-spam on blogs and forums (which counts as inbound links), creating gateway pages, using hidden text, and all sorts of other dodgy behavior. For example, in the digg article, one user points out that the submitter’s user account, and the first few commenter’s user accounts were all created at the same time (yesterday). The obvious conclusion is that the submitter created some fake accounts to vote for his article, raising the odds of it getting to the front-page of digg. In my view, this indicates that the submitter realizes his article lacks in value, so he must try to game the system to get the results he wants.

Sometimes optimizers use black hat techniques because they feel they’ll get a more immediate payoff. Other times, they use black-hat SEO techniques because its the only way to get a site ranked if it has nothing of value. The latter reason is very common, and is what really harms the internet experience.

So those of us working the Wild Wild Web will soon have to find ourselves new job titles, if we cannot find a way to help those on the outside to distinguish between black hat and white hat SEO.

The Sydney Morning Herald reports that office workers are turning to the web to find out the cricket scores.

SMH says Akamai (an internet content delivery firm) indicates that Australian news sites are receiving more than 126,000 visitors per minute – thats 70% increase.

This reminds me of the recent Melbourne Cup – I was looking for some info there, and couldn’t find much in Google.

I’m thinking that ‘special events’ create a special niche which has not yet been filled. Visitors are currently turning to news sites to find out whats going on. Maybe you can jump on board? Create a site for next year’s Melbourne Cup or Ashes Tour, and just quietly get some content and backlinks going over the next year. Then, come next events, you’re in a good position to get plenty of visitors for specific terms.

Just a thought.

Interspire is a software company who I bought some software from a while back for a client. They send me periodic email, most of which gets deleted. Their latest newsletter I liked, thanks to one article titled “The 12 Most Popular Headlines Of All Time”.

They give examples like “They laughed when I sat down at the piano – but when I started to play!“, and explain how this makes people want to read more – “what happened when he started to play? I assume she was really good and nobody knew?” etc.

The article is linkbait (designed to get lots of people to link to them, just like I am doing now) but its a good read, so they’ve earned it!

One of the sites owned by my employer (Tailored Consulting) is called Brand Boots.com. Today, Brendon tells me that BrandBoots is ranked #5 in Yahoo for the phrase “ugg boots” – and thats great news.

BrandBoots was a venture by Brendon and Mel to sell Ugg Boots (Ugh boots, Ugg boots, Ug Boots – however you want to spell it!) from an online shop. Over Christmas etc, they would sell LOTS of boots overseas. (Don’t forget, its hot here in Australia – but other countries can get snow, and warm sheepskin boots are great in cold weather).

But they had a lot of trials and tribulations, including:

• Dispute over the phrase ‘ugg boot’,
• Lots of returns when the material turned out to not be high quality
• Mel sacrificing her Garage to serve as a ugg boots warehouse!

And now they’ve stopped. Nobody was really putting any effort into growing and marketing the website, because there were just too many hassles. So now its over – but they haven’t closed it down.

See, BrandBoots.com is a quality ugg boot domain name. Its been around a long time, and its got a lot of content, and lots of sites linking to it. So what they’ve done, is put up a notice on the homepage:

Over the past couple of years here at Brandboots.com we have sold 3 different brands of Ugg Boots.

To the right is a photo of our packed boots awaiting pick up one day – that’s in excess of 80 pairs of ugg boots sold per day. But no more!

And whilst these ugg boots have been fine, we’ve stopped selling them completely.

No more will be sold of these other brands. The reason why is simple.
Because they are nowhere near as good as UGG Australia boots. Simple.

There ya go. They’ve decided, “if you can’t beat em – join em” and have signed on as an affiliate for Ugg Australia.

So now Brandboots.com can make money as an affiliate without anybody having to spend time on it!

In fact, Brendon tells me that it cleared several hundred dollars last month – and the only time invested was to remove the store a few months ago, and add affiliate links in. Very minimal stuff.

Perhaps you should consider affiliate systems instead of adsense/PPC ads on your website?

Jonathon Mendez has put up a great article titled “Choice Kills Conversion“.

Too often websites, landing pages and even ads decrease the odds of conversion by presenting considerations to users when they are already past the consideration stage and ready to buy.

Much like Columbus showing how to stand an egg on its end, this seems blindingly obvious now that its been pointed out.

Practical example: If you run an Adwords campaign for Ugg Boots then conventional wisdom suggests you create special landing pages for various keywords and phrases. So if I click an add for Jumbo sized Ugg Boots, then you’ve got a better chance of me buying your product if you don’t show me options to Mid-sized boots too – thats just going to confuse me!

So Matt Cutts has put up some video-blogs to Google Video recently, looking a little like he’s been kidnapped and forced to answer questions put forth by readers of his blog. And good on him, he’s got some great responses in there.

Read on…

• http://video.google.com/videoplay?docid=-9028425054136856586

At the end of this one he talks about having links in an option box. For example, on Tailored.com.au – thats the only way to get to our blog archives!! We’ve noticed that once blog entries drop off of the homepage, they tend to drop really fast in the SERPS. I guess this is why. (Don’t know why it didn’t occur to us earlier, actually!)

Given our fear of totally overhauling the site, I think the most elegant solution will be to create a file called ‘sitemap-archives.php’ or something, which scans for files matching certain patterns. It will identify the monthly-archives pages that blogger creates, and just make a little sitemap of them. (Or alternatively, scan for individual post-pages, and map those out grouped by date)

• http://video.google.com/videoplay?docid=3583760678227172395

In this video, Matt talks about webmasters having many sites on a single IP address. He notes that it is very common for this to occur. (I mean, most webmasters will have shared hosting. So your website will be on the same physical machine, with the same IP address, as potentially a thousand other sites)

Matt says that having 10 or so of your own sites would be fine, but having 2,000 would raise some eyebrows and get you in trouble. I wish he’d drawn a little more distinction there – I mean, would it be a problem to have 100 sites on a single IP? How about 200?

What I find most interesting is the way he mentions ‘different IPs’. He doesn’t mention C-blocks at all. So if your IP is 111.222.333.444 (I know, impossible), the C-block refers to the 3rd grouping of digits. So 111.222.333.444 and 111.222.333.555 could concievably belong to the same datacenter, whereas 111.222.333.444 and 111.222.311.444 should be totally different.

I don’t know if this is a myth or not, there doesn’t seem to be any ‘final word’ on the matter. But many people believe your sites should be spread across multiple IP addresses, and multiple C-blocks if possible, to maximise the value of cross site linking.

The interesting part here is the way Matt says ‘different IP addresses’, without even appearing to consider C-blocks. Is it a myth, or isn’t it?

We’ve been considering getting dedicated hosting at work, because we have a lot of client sites (100? 200?) spread across several different reseller accounts with different webhosts. I’d love to consolidate these onto a single machine, but I don’t know if our search engine rankings would take a hit because of this.

• http://video.google.com/videoplay?docid=3583760678227172395

He answers the question, does the webspam team use Analytics data at all in their algos. He seems a little evasive to my mind. He’s clear that the WEBSPAM team doesn’t use that data, but seems unusually uncomfortable committing to saying “they don’t factor into the SERPs at all”. To my mind, Google either does or soon will use Analytics data to affect your sites ranking.

So this is a bit of a double-edged sword. Perhaps if your analytics show you have visitors arriving at your site from all sorts of referrers (backlinks), and are getting lots of traffic, you’ll get a boost. But the opposite may also be true – a site getting only a few hundred visitors per month may be considered “unimportant” by google, and thus dropped in the SERPs.

Of course we hear a lot about ‘is it a good idea to let google collection *that much* data about us? Well, I guess it depends. It could be good or could be bad. It would seem prudent, however, to limit the data we hand to google on silver platters until we know for sure one way or the other.

The Analytics software is based upon Urchin, which has been around for quite a while. I’ve never seen Urchin in action, but I imagine it’ll be pretty handy. No doubt Google has made many customisations to improve the software for use in Analytics, but I’m sure the base software would still be worthwhile. If we were to get dedicated hosting, we could also get Urchin to monitor all of our sites, as opposed to relying on Analytics.

This is a new issue, and one that appears to be especially important to Google. What makes google trust your domain?

Well, it seems like there are a few factors contributing:

• Age of domain
• Incoming links (quality, not quantity)
• Historical information (have you been spamming, do you have a history?)

The most important factor, by far, is the age of the domain. That is, how long has the domain been registered for? Has the site been up that long? Have you been slowly adding content and growing the site?

MSN and Yahoo don’t seem to weigh this as highly as Google does. With highly competitive terms on Google, you can throw a lot of weight behind your ranking if the site has been established for a while. A good timeframe seems to be at least two years!

Here’s a cool tool:

Here’s a except from Jim Boykin, on a cool seo tool:

Have you looked at this tool yet? How old is your site compared to the top 10 sites? How many backlinks do you have compared to them? How is their quality of backlinks compared to yours? What the history of people linking to you (what words have people been using to link to your site?) How good of a resource is your site? Can you site attract natural backlinks or are you 100% relying on your SEO for backlinks? How much unique content do you have? Do we have to pay everyone to link to you, or do you have a “natural” reason why people might link to you? Keep looking at that tool with the phrases you’re targeting, and keep asking yourself these questions. An SEO is only part of your solution – how much are you helping?

All are good questions. This site can help. Take my employers website, tailored.com.au – as an experiment, Brendon took a stab at optimizing a page for ‘gold coast accommodation‘.

Please note that tailored.com.au has absolutely nothing to do with that keyphrase. Yet its achieving an okay result for a highly competitive searchterm (Pro Tip: Use google trends), somehow. This is because (on my machine, right now) tailored is fluctuating between a pagerank of 6 and 7, depending on which datacenter I query.

So what does it all come down to?

A trusted domain, and solid backlinks (which are the most reliable enabler to achieve a high pagerank/ranking).

On Page Factors – what does that even mean?

Well, its everything on your page. Lets look at the basics first…

• Title tag
• Meta tags
• Heading tags
• Keyword density
• Meaning HTML markup

Thats the short list, now lets look closely…

Title Tag

Simply put, this is the [title] tag of your post. It is, as the tag suggests, the title of your page. Its what shows up in the blue (conveniently named) title bar at the top of your web browser. When a visitor adds the page to their favourites, thats the ‘suggested’ name of the shortcut. Search engines will use it as the heading of the result snippet they show searchers.

You should absolutely used a few keywords in the title tag, no questions asked. But make sure it makes sense too.

Pro tip: Its my experience that you’ll get a better response from search engines if you phrase the title as a question. This won’t help your rankings at all, but does seem to provoke a better response when people see that question in their search results.

Meta Tags

First, what is it? Lets try: [meta content="Describe your site here, using lots of keywords" name="description" /]

There jury is out on the importance of this one. Meta data used to be very relevant, until people figured out that thats all the engines looked at. You know, back in the day. The engines quickly realised that people could write anything in there, and it wasn’t always accurate – so they stopped relying on it. Do they still use it? Well, that is the question!

Google will sometimes show the meta description in the results page, but will also sometimes show a snippet of actual content from the page. Popular theory (and it looks this way to me) is that Google uses the description until it has a chance to index the page content, at which point it uses that instead.

Pro tip: Don’t keyword stuff your meta data. That would mean a meta description of “cards vegas poker blackjack gambling online”. Instead, try something like “Poker.com is the best online casino, with secure transactions and honest games”. Meta data might not count for much, but every little bit helps! If you employ questionable tactics like keyword stuffing, you’re just going to set off alarm flags.

Headings

The heading tags! See how the word ‘headings’ is big and bold? Thats a [h3] tag. Simply put, its a level 3 heading. Headings indicate importance of the page. But its vital you keep your page well structured! You should have one h1 tag. Then a couple of h2 tags, with h3 and content grouped underneath.

Basically, look at your page with the CSS turned off. (You are using CSS, right? Of course you are, have a cookie!) If it still makes sense, and is set out like one of those boring research papers you had to do back in school – you’ve done it right.

Keyword Density

Experts debate about things like optimal density of keywords. All it means is that if you are optimizing for the word ‘poker’ – a 1% density means that for every hundred words on the page, you use the word poker once. Experts debate about a 3% vs a 6% density. The theory is, a higher density is better, but only up to a point. Just write naturally! If you’re writing about poker, you’re going to achieve a decent density anyway. And the principal of the engines is that you shouldn’t have to spoon-feed them content – write for your users, and the engines will pick you up and appreciate it.

Meaningful HTML Markup

This part is important. For example, a lot of editors will bold words by putting [span style="font-weight:bold;"] around the words. And sure, it looks bold. It IS bold. Or they use the [b] tag, for shorthand. This is no good. What they should be doing is using the [strong] tag. That tag has meaning to the engines!

Same for [i] vs [em] for italics. Use the meaningful tag wherever possible (strong or emphasised), because the alternatives are just presentational – they do not add any weight or meaning to the words, which is what your intention should be.

Incoming links are very important to your site’s SEO strategy. The basic theory behind search engines is that the engine will pretend to be a user on a website. The ‘user’ starts clicking links – any links. Eventually, they’ll cover all the sites that have links to them. So the more links to your site, the quicker you’ll be found.

Once search engines started getting huge amounts of pages indexed, they needed to figure out a way to rank the page – to determine how important it is, and what search phrases the page would be relevant for. Engines use a bunch of factors for this, but links are an important one.

The more incoming links, the better

See, the more links you have, the better – the search engines will find you faster. The more links you have, the better – search engines think “wow, this is a really popular site, everybody links here – it must be important”.

Quality of links is also important. A link from some crappy geocities site won’t be as good for you as a link from a really popular website. That is, a website with lots of links pointing to it – because now the engine thinks that Site A is important, so when Site A links to you, the search engine will give it more weighting.

Additionally, links from sites similar in topic to your site will do you the most good. If I have a site about foot fungus, and I link to a site about gold coast holiday acommodation, then the search engine won’t see a connection, so it won’t give it as much weight.

Thats the popular theory anyway. Makes sense.

So lets talk about types of links. You have:

• Reciprical links
  This is when you link to Johnny, and Johnny links to you. A simple link-exchange program. This type of link is very common and very popular, because it works. However, it could be better… search engines recognize this behaviour as artificial, so its not weighed as highly as purely natural links.
• Link Circles
  Whatever the technical term is, this is when you link to Johnny, Johnny links to Sally, and Sally links to you. Its kind of a technique to get around the issue of engines recognizing the link as artificial, so in that sense, its better. However, its not ideal either – unless all three sites are exactly equal in the engine’s eyes (will never happen), then one site will be getting an advantage over the other two. Its still of benefit to all sites, but not equal.
• One way links
  This right here, is as good as it gets. You have a link coming to you, and you don’t have to ‘trade’ anything for it. (Unless its a paid listing of some sort, but that usually looks the same to a search engine.) Its kind of like the other site saying “Yep, this site is great and I really recommend it”.

So one-way incoming links are the best ones you can get, but don’t turn down the others either!

A word on outgoing links

Seems to be a few schools of thought on this topic. Should you link to other sites if you don’t have to?

Some people say that every link on your page dilutes your pagerank a little. That is, the more links you have, the less each link counts for. This makes sense.

Some go further, and say that when you link out from a page, it actually lowers the pagerank of that page, transferring it to the destination pages. Therefore, they say, you should avoid linking to other sites like the plague! This one doesn’t make so much sense to me.

Popular opinion also seems to indicate that if I link to a site about green bananas, then the engines will count me in the ‘green banana’ community – therefore, I can improve my own ranking by linking to other sites related to mine. This makes a certain degree of sense to me. I don’t believe that this technique will raise my pagerank at all, but I do think this will help the engines classify my site. For example, if my site is all about ‘court’s, I could be talking about the law, or basketball. Theoretically engines would figure this out from the other words on the page (it gives ‘court’ the right context), but I can help them out by linking off to related sites.

Helpful Hint:

A good way of getting incoming links is to submit your site to free directories. This directories are often held in good esteem by the engines, and a few links from these sites can go along way. Whatever you do though, don’t wind up on a link-farm. This is a site with nothing but links to other sites – The distinction is mostly that a directory will group links, seperate them on page, provide a description of the site – and a link farm just grows links. I haven’t heard much about link-farms in recent memory actually, come to think of it, so perhaps this isn’t an issue anymore. But there was a time when one would be penalized for using link farms.

I’ll post up some directory resources soon.

A word of caution…

I’ve been reading some discussions recently, and these discussions have been confirmed by Matt Cutts on forums and blogs. Basically: you don’t want a million links from crappy little sites. You’ll be much better off to have a dozen links from high quality sites. I guess Google is viewing sites with a gazillion incoming links as artificial, for the most part – those links have to be part of an artificial and planned campaign.

Its worth noting that the number would have to be pretty high to hurt you – I think we’re in the realm of those damn viagra-vendors who spam blogs and forums all day!

Since I’ve started writing this four-part series, I’ve discovered a few other blog entries. One topic is especially important to my last few paragraphs.

Text Link Brokers Says: (On the topic of Recip links vs One Way Links)

So, to answer the question which is better? The answer is neither. You need a blended approach to link building – they can’t all be one way inbound links nor is it in your best interest to do all reciprocal links. You need to do a little of both.

To expand on this quote: Rob is saying that when you have the choice between a recipricol link (you link to me, and I link to you), make it look natural. He points out that newly established sites will typically only get links from other sites with low rankings – but as they’re established, they’ll get better links. So when you first get your site, exchange links with crappy sites. After you’ve grown, get links from better sites. He umms and ahhs a littel about buying links, but recommends buying a link in Yahoo directory. (Shame, its about the most expensive listing you can get.)

What he’s getting at here, is that if you have a site with no recipricol links, and 500 incoming links – well thats good, right? Thats what you should be aiming for!! Except that realistically, the engines will figure you have paid for these links, and penalize you. So look natural… get a combination of one-way and recipricol links to your site.

Had a little discussion in the office today, regarding SEO. We came up with this shortlist:

• Frequency of content updates
• Incoming Links
• On-page factors
• Trusted domain

There’s more, obviously. But these are the major factors affecting search engine placement. Lets discuss in more detail. This is post number one of a four-part series, in which I’ll go over these four tips.

Frequency of content updates

Its a general rule of thumb that the more often content is updated, the better. This is the major reason blogs tend to score so well – the engine will recognize that content on the web is often time-sensitive, and that the most recent information tends to be the most relevant. This doesn’t apply to all fields, of course.

For example, an article on the difference between cow species won’t be using any information uncovered in recent times, and thats not likely to change anytime soon.

So how can you apply this to your site to get improved rankings?

Running a blog on your site can be advantageous. Its easy to update, thus hopefully you get content more frequently. (Whether the blog is on your homepage like http://www.tailored.com.au is a whole other debate). I’d suggest a blog somewhere on the site (with priority placement, so the search engines realise it is important and check it over frequently), with short posts (say 300 words max?) that summarise a few key points, then link off to an article that covers the topic in more depth.

The article could be a brand new one, or could be an old one. You could also do what I’m doing here, and have a !more tag, so that most of the entry is only shown when you click to ‘read on’. With wordpress, this essentially creates a new page for my post (the article), and just shows a snippet on the homepage. This also reduces my chances of dupe-content penalties. (Another reason I don’t like blogger – if you write a large article as a post with blogger, it will be published to the Homepage, to the Archives, and to the Individual page – which one will google class as the ‘original’? Its anyone’s guess.)

The idea is that you recycle old content. Take that article from last year, change a few sentences, add an article, then write another blog post linking to it. The engine already knows its there, hopefully – but this would reset the clock, pointing out to the engine that “yes, this information is still relevant and exciting.”