December 19, 2012 at 11:10 am
Guy Levine is the CEO for Return on Digital in Manchester and London.
In his talk Guy explains about SoLoMo (Social Local and Mobile), where mobile is about smartphone and tablets and local is about the local listings, however, although you can also get local listings on your phone as well as on a desktop, they both need to be treated differently.
He continues by saying that websites need to be mobilised and not miniaturised because people want different things. He describes 3 different types of mobile web and categorises them into the good, the bad and the ugly:
Traditonal SEO cannot be applied to apps to get it noticed but the app store can be gamed. However, if the app makes most people use it once and then delete it then it’s a big waste of time, including the dev cost. The other problem is the big investment involved to market the app, so for example, in order to get the app to the top of the app chart that app needs lots of downloads, to get those downloads means spending lots of money on good old media, but this can be avoided if the app has a massive user base. The final point is that apps aren’t really owned by the developer/producer since they are pushed through a platform. If that platform decided that it didn’t like your app for some reason then the app can get pulled from the store, so choose your platform wisely. A good way to do this would be to check Google Analytics to see how many people are visiting your site or blog on a certain platform, which should give you a good idea of which platform to develop for. A good tip is to use remarketing to advertise to regulars any good pieces of content you have on your site or blog, or even your mobile app!
This type of mobile web is the type that has a url with an m at the beginning i.e. m.facebook.com. It has a few issues like, having to build a new website or building a different version of the product database since they need a new dataset to drive it – which means you now have two things to maintain instead of one. If the mobile site has similar content to the main site then there is massive duplicate content issues and the possibility that Google doesn’t know which to index or which to use as it’s primary result. Two strategies that can be employed to solve this is (i) to noindex the mobile site or (ii) use the rel=canonical tag to push the important mobile pages to point them to the main site. The good thing about web apps is that they can be added on to a current web strategy, but it does require a separate seo strategy because of the two separate sites.
Responsive is where one site changes it’s appearence, via CSS (Cascading Stylesheets), to suit the type of browser/device that is accessing it i.e. is it a mobile, tablet or desktop browser/device. The good thing about this, especially over web apps, is that there is only one site, which means only one seo strategy and no possibility of any duplicate content issues.
Guy then starts to look at keyword research and says that mobile users are going back to head search terms because of the difficulty of typing long tail phrases on a phone or they can tend to use auto complete. A suggestion to get those auto complete terms was to use Firefox’s user agent switcher to switch into the Apple/iOS mode and scrape the auto suggest, you can do the same thing with a tool called Mozenda.
On content, Guy mentions that using the old content won’t work since mobile users are looking for shorter bursts of information, they also look for reviews and social endorsements. It’s useful to have your social media icons as big as the buy buttons as that tends to help increase conversions since people are looking for more of a social interaction with their mobile device. The content needs to have a strong entertainment ethic because mobile users tend to look for enetertainment rather than to solve problems. As with desktop sites, mobile users need to be given calls to action. The best is click to call, unless the user is on a straight eCommerce site, and try to make it a streamlined experience – so no long forms as you are dealing with fat fingers and not mouse clicks.
Guy then looked at upcoming trends in 2013 and identified two:
This is going on all the time, like being at a bus stop having to wait for that late bus. Does the site fulfil the requirements and/or needs of that user in a 15 second time window?
Non-geeks are now interested in knowing data like how much electricity is used in the home or Klout/cred scores, so it would be beneficial to add something like that to the site.
If you are stuck for more ideas on mobile then Google has a questions and answer website where it gives a pdf of ideas at http://www.howtogomo.com
On local, Guy points out that just having your business details (name, address, phone number etc) on your contact page just isn’t enough, you must now put it on all of the pages through the website via the footers. He mentions to get reviews on local listings, for example on Trip Advisor, smaller hotels have been beating bigger ones because they have more reviews – these tend to work better on mobile too. Also look at getting links from .ac.uk sites from their location context pages since universities have a really good local presence.
Finally Guy summarises what to do for a mobile strategy:
Question from Neil Walker – owner of Quaero Media: When you are talking about “would you include content for mobile” you suggested the canonical tag and the stuff I’m reading from Google is that as long as you’re targeting or redirecting the mobile bot to the mobile site and desktop bot to the desktop site you shouldn’t need that. So I’m not personally experienced much with using those type of tags so I wanted to know your thoughts on it.
Guy: That was a fantastic question for someone who knows how to write HTML 3.2. Here’s my view, maybe it’s not the view of my company in case I say something a little naughty. I think as we know Google guidelines and Google reality kind of vary a little bit so what I would be doing is we’ve seen great results in doing it that way. I don’t think anyone would say that we’re a bad agency by saying that maybe we are a couple of months behind the guidelines anyway by the time we’ve implemented the results. So all the stuff we’ve done up to now, we’ve done it that way, but I’d just say keep an eye on it
Question from Unknown [If you are the person who asked this then please get back to me with your information – Thank you]: Just about the responsive stuff, obviously a lot of businesses especially ours that were early adopters of mobile so we weren’t able to do it at the time and now we are stuck with a mobile site that is on a separate url, it’s not even a sub domain. In terms of seo, what is the best way to tackle that because we have unique content across both but we don’t want to fight the same search space and my understanding of the listings are we’ve got our main site [then the mobile?]. How would you approach that?
Guy: So the interesting point I picked up on there is fighting the same SERP space, I think that’s the wrong way to look at it, because I think that’s like owning the SERP space. If you have two of your sites that are ranking especially if they are on different urls, then that has only got to be a good thing because all of a sudden you are pushing one competitor out the bottom. I think that this is a business problem and not an seo problem. The problem is that do we take a hit and merge everything into one site so that we have one site to play with going forward? Can we afford to do a great job on two sites separately? And which ever site you choose, there’s consequences. It’s a bit like changing your www address or changing your brand, sometimes your business just needs to do it so that you can move forward and grow. I think either way you could get away with and I think either choice would be the right choice, you’ve just got to decide which one that you want to do. But I remember when I went to my first search conference, one of the speakers was complaining that one of her competitors had 8 listings and how that was one of the worst things in the world and it’s like
January 21, 2012 at 4:56 pm
This week I helped out with a problem concerning a photography client’s new website. Originally, upon checking his links, I found that he had a review site which was not mentioned on his old website. This, as you can imagine, would be good for potential clients to see as a testimonial page.
But, here’s the problem. Do you risk upsetting the client by copy/pasting said reviews onto his website or causing a potential duplication issue which could lead to a rankings drop?
The answer to this is to use an iframe, but they themselves do present a problem if the site in question doesn’t immediately show the reviews straight away. Mainly since external domains in an iframe cannot be interfered with as that presents a security risk for that website.
Our client’s review site had that problem. It had a large header and then a member profile box before you got to see any reviews. So what I did was to extend the size of the iframe to something like 1240 by 768 and add scrollbars:
Next I encased that with a div, set the width and height to be smaller than the iframe, hid the overflow and set it an id
Then I set the body tag with an onload attribute and named it with a function name that would be my scrolling function. And finally I wrote said function, as follows:
The scroll function is just a simple script that grabs the div and automatically scrolls 100 pixels down the page. Because of the onload attribute in the body, this will set the iframed EXTERNAL domain to a position that you want to present to the user when the page is loaded. The best thing about this is that you cannot scroll back up to the top of the external domain’s page, since you set it to be 100 pixels down!
Here is the code in full:
Here is the demo:
January 12, 2012 at 10:47 pm
I’ve been a gamer for as long as I can remember and I’ve enjoyed seeing how video games have gone from an indie environment to the maturity of a few years ago, although not todays gaming because of the over reliance of “safe” developement whereby every game seems to be another run-of-the-mill first person shooter and the endless chirade over DRM (Digital Rights Management) and its implementations that cause gamers grief everytime a new and even more draconian version rears its ugly head!
As my gaming years progressed I started learning programming and did try making a game from scratch. I fondly remember, in my college years, trying to write a text adventure game on Turbo Pascal, just using IF..THEN..ELSE statements until I realised that I was getting nowhere, so it got binned. It wasn’t until a game programming competition on the Retro Remakes forum, many years later, that I actually managed to write a game using GLBasic.
During this time, with access to an open source handheld – the GP2X, I started to get into indie games. My first indie game was Cave Story, a metroid like platformer that was so lovingly polished I couldn’t take my eyes off it for hours! Since then I’ve played quite a few titles including Payback, Cthulhu Saves The World, VVVVVV, Rock of Ages, Portal and Tobe’s Vertical Adventure. There are so many titles I still want to play like Frayed Knights, Crayon Physics, Limbo….just to name a few.
Those of you reading may have seen some of these titles already but some of these indie games you won’t know about, and that’s the problem I’d like to address to the indie gaming community. There are a lot of games out there that people would play if they knew about them. Unfortunately, as Jay Barnson has already written about (see point 6), not all indie developers probably know anything about marketing their product to gain more traffic.
So here’s 5 marketing ideas to help you get just a little more exposure:
December 14, 2011 at 12:13 am
There are a few problems that plague the seo community, one of which is the snake-oil salesmen and the other is the same old regurgitated posts about how “content is king”, “paid links are bad” or even good old “SEO is dead”.
Coupled with Google’s QDF (Query Deserves Freshness) and the seemingly endless Panda updates, many SEO companies have wrongly assumed that in order to rank they must bring out new and unique content frequently. However, they forget that content is only king when they are writing for a niche market or the keywords aren’t as competitive, otherwise you’ll have bring links/traffic to that article in order for it to rank well within the Panda entrenched SERP field.
As an example I once wrote an article on how to get the Windows 98 operating system to run on Qemu, which, at the time, had very little information to help people to get this running. Although the how-to article is a little outdated, it still gets me some traffic to this day.
Think of it in terms of Chess – the King, although the most important piece on the board can only move 1 space in an 8-way directional path (up, down, left, right and diagonals). This king represents your content, whereas your Queen – the most powerful piece on the board with the ability to move any number of spaces in an 8-way directional path – represents your link building efforts. When the board is full of opposition pieces, the king is potentially under threat but if the opposition has less pieces, it weakens the threat imposed on your king.
This also brings forth another point surrounding the issue.
If this content is to be written, who requires it?
A few months ago, this was addressed by Yousaf Sekander when he said that many SME’s (specifically referring to tradesmen like locksmiths and plumbers) were not required to produce good content in order to rank well in the SERPs, however, with Google’s enforcement of these practices, this prospect seems increasingly so.
December 4, 2011 at 10:14 pm
Neil Walker is a Chief Technical Officer for Just Search and he oversees SEO, PPC and R & D.
In this talk Neil discusses how to forecast for SEO, he shows that Google
November 20, 2011 at 11:44 am
Kelvin Newman is a Creative Director at SiteVisibility, specialising in SEO R & D and producing quality links, they are based in Brighton & London.
In his talk, he starts off by saying that there are 3 things which influence link building:
And then goes on to use psychological studies to show how people think and react to stimuli and tie this into link building.
The first study mentioned was the candle problem, whereby participants are given a box with wall tacks inside, a candle and a box of matches. Their task is to fix a lighted candle on the wall, however, most participants couldn
October 18, 2011 at 10:47 am
Barry Adams is a Senior Internet Marketer for Search at Pierce Communications in Belfast.
In this talk, he was showing how it was possible to optimise for Google News. Although, it is dominated by all the major newpapers, you still can get some useful off-site seo use out of it. The best news is that Google News results do actually have a high Click-Through Rate (CTR), however, unlike Google itself, it doesn
October 17, 2011 at 3:55 pm
Sebastian Marek is a Software Architect at Plusnet, which is now part of the BT Group.
This talk was aimed soley at code and design metrics and had begun by showing that the only valid measurement of code quality was WTFs/minute. Sebastian elaborated that there was no magic metric to determine code quality and the only justifiable way was to read the code and decide for yourself whether the code is good or bad.
Good design quality metrics are not necessarily indicative of good designs. Likewise, bad design quality metrics are not necessarily indicative of bad designs – Jdepend Docs
Simple metrics consist of:
Although, lines of code mean nothing if no knowledge of the system is known.
Sebastian showcased a couple of PHP scripts that measured metrics, such as phploc and php code coverage – which had an interesting metric called CRAP (Change Risk Analysis & Prediction), that measured how likely code breaks if the code was changed. A low number would indicate low risk, whilst a high number indicated a high risk.
He explained about cyclomatic complexity (CYCLO), which was the amount of decision logic in a single software module, and said that the problem with the first version was that it didn’t take the logical operators (&&, || ) into account, so CCN2 was born and from that catch was then added.
The problem with CCN in its entirety is that it doesn’t take into account the nesting complexities within control structures. NPATH solves this.
Twitter for Sebastian Marek: @proofek
October 11, 2011 at 5:23 pm
Ian Barber is the development manager for the Virgin Group digital team in London. He occasionally blogs at PHP/ir and is a regular conference speaker.
In his keynote speech at PHPNW11 Ian started to talk about how amateur skaters had improved their craft, over the past 10 years, by watching videos of professional skaters who had videoed their stunts for amateur fans to watch. Although the professional skaters didn’t show their mistakes on the videos, it still inspired their followers to attempt to get the stunts right. He went on to explain how some of the technology produced today came from academic papers written from the 1970’s. The reason that it had taken so long is that there was a lack of computing power to put those papers into practice.
“The best way to predict the future is to invent it” – Alan Kay
Ian talked about how every great innovation was built from previous work, for example Facebook is built from the work behind Friendster, AIM and the Harvard Social Networks; PHP was built from C, which was built from B and finally BCPL. In all these instances, the people behind those innovations were building them for personal preference and not to change the world. It was these small changes that can have big impacts. Unfortunately, small changes can also cause dogma to appear, but this should be ignored – a good example of a company who ignored dogma well is Google.
“The whole field had suffered blinders; in some sense search really did need two people who were never tainted by people like me to come up with that shake up” – Amit Singhal, Bell Labs
He mentioned that Amazon and Google had built their applications from scratch and then wrote papers about their findings – it was from these findings that Apache’s Cassandra was created.
Ian closed the keynote speech by encouraging the audience to do whatever idea they had, but to know the history behind it.
“Play your part to take the next step” – Ian Barber, PHPNW11
Twitter for Ian Barber: @ianbarber