I have always wanted to believe in "good" link building. I understand why the other stuff works, everything from link networks through paying webmasters to outright spam, but I want to believe that you can win with great product and real organic marketing.
Today, I'm going to walk through some examples of my current favourite form of link building, namely using your developers to build cool stuff to acquire links for the long-term.
As a web consumer, I love to see innovative companies building their brands by building amazing products. As an SEO, I want to believe that it's possible to get links this way, I love making the internet a better place.
Today, I wanted to run through some examples and thought-experiments of people doing this well. My role these days revolves around working out how we should be advising our clients and thinking about how we can build the teams and tools that enable us to help them. I spend a lot of time thinking about the future of the search landscape. As ever, a large part of the game relates to links.
I'm sure I don't need to tell anyone here the importance of link building. We all know it's important, but what "it" is varies hugely depending on who you speak to. The confusion and demand is one of the reasons we are running a set of day-long link building conferences in the next few weeks.
I'm going to share one of my favourite link building tactics: using your developers to build cool stuff.
Link building with product #1 (your core product)
If you can build things that are naturally link-worthy and that have a built-in ability to demand that people link to them, then it's no secret that you are on to a winner.
My favourite example of this is Dropbox who, through a combination of flawless engineering execution, built-in virality (through the free space rewards they give you for recommending them) and a friendly brand that is easy to evangelise, have gained links from some of the biggest sites on the web and gathered tens of thousands of links in a relatively short time.
The core brilliance of this strategy is that it doesn't actually rely on Google to reward them - these links drive sign ups and traffic and so, while they rock the link metrics, I bet they rock the true KPIs harder. The big downside is that not every business can be this cool.
Supposedly. I have to say, for me, I find it a lot more fun running a business that people want to link to than one that has to beg, borrow and steal. This is probably just me though - and is similar to my preference for trying to compete on quality rather than price.
There's a lot of money to be made being Walmart, but it's not my game.
Link building with product #2 (fun and games)
Whether you have a core business that people naturally link to or not, if you have some creative techies on your team, you can quickly build things that get links.
This is my "current favourite tactic" I mentioned above - alongside content-driven and design-driven linkbait and outreach-driven link building (all of which should be part of your arsenal) I like product-driven linkbait.
The key point here is that we are not necessarily talking about polished products. I would actually argue that constraints on time invested are critical to the success of these kinds of thing.
You are not necessarily looking to create something perfectly polished here - sometimes the rough edges are part of the attraction - some of the best results come from the output of spare time creativity or hack days / weeks.
They allow you to tell the story of how it was "built in just three hours", you get conversation about improvements as well as sharing it for coolness it's more likely to get links as it looks less commercial.
Crucially, you can do more of them. As with any linkbait, you are, to some degree, bottling lightning so you should always prefer 5-10 shots over one big punt that took 5-10 times as long to create.
I'm going to show you some examples of the kind of thing I'm talking about before I dig too deep into how (I think) you should do this for SEO purposes.
The examples that follow highlight different elements of success, though none was done for pure SEO purposes:
Exhibit #1: facelette.com
I came across the guy behind this site (Zach Holman, lead developer for Github FI) through a screencast he recorded about automating inefficiencies (possibly the most amusing geek-out screencast I've ever seen - there's a niche for you).
He is clearly a kick-ass developer which gives him a headstart when it comes to this kind of thing. I didn't say it was going to be easy.
Of course, this was all before he shut it down, with casual disregard for the links he'd acquired from gawker, gizmodo, techcrunch, gigaom, hacker news, business insider and many more.
Exhibit #2: justbuythisone.comReevoo powers reviews for large swathes of the UK e-retail market [fair disclaimer - they're also a client of Distilled].
They have a kick-ass dev team and they decided to have a hack week where they built something cool. The result was "Just Buy This One" - a site that takes the heart-ache out of decision paralysis for electronics purchases by recommending a single product in each category / price range combination.
By powering it off reviews they are uber-confident in, they felt happy to back the recommendations up with their "room service" guarantee that they'll take it back if you don't like it. This has a different set of cool features to highlight:
- Obviously high effort.
- Looks beautiful.
- Genuinely useful (I bought a TV via it).
Just to prove that you don't necessarily need to be an amazing developer (or even a developer at all, really), here's something I built during a Distilled hack day.
The focus of our day was on promoting a side project of ours (Hire Marshal - a tool for easily building job application forms and collaboratively reviewing applicants).
We did a bunch of serious things including a demo video and a revamped homepage, but at lunchtime we hit on a silly idea.
We were brainstorming about how annoying recruiters are. We must get a dozen cold calls from recruiters a week trying to sell their services. They are remarkably hard to get off the phone - we just can't seem to get the message across that while we are hiring, we don't want their help.
Somewhere in the melee, someone came up with the idea to create a form to screen them - kind of like an application form for them to apply to be a supplier of recruitment services to us. We thought we could make it look like something we had thrown together (shouldn't be hard!) but make it evil.
If you go take a look and start trying to fill it out, you will find that it starts adding questions as quick as you can answer them, that the questions get more and more stupid (before it loops round) and that if you try to submit it, you get a warning to "answer all questions".
It was just a bit of fun and an excuse for me to learn my first jQuery. It was a good job that it didn't actually have to work as not only am I not a developer, but I didn't really know what I was doing. Nevertheless, two hours later, it was live.
We got some amusing stats for something that was only really an afterthought:
- Over 50,000 visits in the first few days.
- Over 900 people stayed on the site for more than three minutes.
- 14 people continued filling in questions for more than half an hour!
- Traffic from ~120 different domains - that's more links than I normally build in 2 hours...!
- Over 1,300 visitors to Hire Marshal from the tiny attribution link at the bottom of the page (this surprised me greatly)
Now, this wasn't done as a link building play really - it was just some fun - but there are obvious routes that we could have gotten more link equity out of it.
How should you get links from all this?
I guess step one, that all of these pretty much ignored, is that if you are doing this as a link building play, then your top priority is probably to launch on your primary domain.
This can limit the scope of what you can build though (how would we get recruitmentcheck.com on hiremarshal.com?). So if you can't do that, your options are essentially threefold:
- Just link back to your main site - depending how you see it, this is either the safest option or a really time-consuming way of getting a single link... I am quite taken by it simply because it's kinda how the web is supposed to work (and it can drive traffic - see the example above).
- Place a copy on your main domain and either set up rel=canonical or 301 redirect (after the initial buzz has died down). This could be effective, but I would not want to see a site have a large proportion of its links coming this way.
- Just rel=canonical or 301 to your money pages. This is about as shady as my recommendations are going to get here - not the route I'd take - but could definitely be effective - especially short-term.
The best bit, though is that this kind of link building brings long-term rewards. If you do it right, these sites continue to get links off into the future without you having to lift a finger. I get alerts about people tweeting about RecruitmentCheck all the time months later.
How can you get started?
I mentioned above that you don't need to be a developer. I'm certainly not - I've learnt just enough to be dangerous. If you are thinking of cobbling together something for the first time, you might find inspiration from the story of 7bks (built by Tom at Distilled) - he wrote up how he learnt to code and built the site in four weeks (HN discussion) alongside a job.
However, as facelette and justbuythisone prove, if you have a developer or a designer around, you are likely to build something cooler by an order of magnitude.
In summary, then, my top tips:
- Just do it - keep the projects as small as you can imagine and churn 'em out to get a feel for what might work
- Get devs and designers involved if you can - if you can't, imagine what you can create yourself
- Think about how you are going to get links from it.
- Jump on news and trends if you can or general memes / popular likes and dislikes if you can't I hope they help you to make cool stuff to make the web a more fun place and get some links while you're at it.