11
Big Brand & Corporate SEO Strategy – Rishil Drops Some Wisdom
View Comments · Posted by admin in SEO Strategy
This is a pseudo-guest post by Rishi Lakhani (or rishil to his online mates). Why ‘pseudo’? Because essentially I have copied and pasted a series of tweets he posted a couple days back which I thought were worthy of capturing in a blog post that won’t disappear after a couple of months (aka the lazy way to write a blog post!). Although these are just a series of 140-character snippets, they are by no means off the cuff comments. They are the product of years of experience as a top-drawer SEO strategist, and make a very important point about the importance of having patience, the right people and the right tactics and strategy for a large scale search campaign to bear fruit. This really resonated with me at this point in time having recently made the the transition from a small specialist agency focusing on SMEs to a global media agency working on big brands. I’ll leave the rest to Rishi…
“OK Lets get some stuff right. SEO isnt Just for Christmas. nor is it easy to turn it round in large corporate dinosaurs.”
“Big Brands have massive changes to culture, internal buy in, stake holder mentality before a decent SEO strategy can be put into play”
10:01 AM Jun 8th via web
“Anyone who does SME SEO should realise that their approach to SEO isnt the same as approach to Big Business – a lot more strategy needed”
10:02 AM Jun 8th via web

“If you havent dealt with large ancient CMSs you wont know what I am talking about. Some times it’s next to impossible to change culture…”
10:03 AM Jun 8th via web
“So although a Strategist would be politically better in Big Brands, a technically adept SEO may find better solutions.”
10:05 AM Jun 8th via web
“It took me and a team of 20 to shift a complete Search Strategy from £15 million in revenue to £40 Million. Over 2 years.”
10:07 AM Jun 8th via web
“So dear SEOs – when you see a big player responsible for a large brand and then see they have shit SERPs dont delude yourself.”
10:08 AM Jun 8th via web
“There maybe more than meets the eye. A Real strategist wont go for quick wins – I take sustainability over short term success any day.”
10:09 AM Jun 8th via web
“So taking it back to point- know your skills- and know who you work with complements you – dont try and get political if you havent got it.”
10:10 AM Jun 8th via web
“/rant/lecture off
”
10:11 AM Jun 8th via web
brands · guest-post · seo · SME · strategy
18
Give It Up! Social Media Edition – SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
Last session from SMX London folks! It’s one of the legendary “Give it Up” sessions, for social media this time. Let’s get stuck in:
Chris Bennet – 97th Floor
Chris is going to talk about 2 main methods he uses:
Infographics – yes everyone talks about them but they really do work
- What makes a good graphic?
- lots of data / info
- aesthetically pleasing
- complex ideas made simple
- Example: STD’s across America – shows hotspots of STDs per region and what the main STDs are – this was a challenge because the client didn’t think there was anything in social media they could do
- Example: Who owns the most servers in the world – - picked up by engadet, gizmondo and many other sites – several thousands of backlinks after only a few months, would have been boring as text.
- Example: What is a Petabyte? (same client as above) – another example where the client didnt think they could do anything cool / viral about their brand.
- Example: Where does the money go? US consumer unit expenditures – picked up by shedloads of newspapers and magazines as well as going hot on Digg and other social news sites – hit the nail on the head for people. Around 29,500 backlinks according to Yahoo
- One of the best thing about this kind of stuff is you’re getting a link profile that your competitors cannot possibly replicate (they can do similar things but not get the same links)
- There is still a problem with social news sites like Digg, Reddit etc – they don’t like commercial domains and sometimes domains get bans (you can ask to be reconsidered though, getting better at being more commercial friendly now)
- Tip: Kevin Rose of Digg is saying all banned sites will be unbanned – when this happens don’t submit your A grade content straight away – give it time and submit other stuff (incl old posts etc) and give your site some age / natural profile first.
- A way of getting around difficulties with getting commercial sites/domains to go viral – set up non-commercial dedicated domains - Example: Visual Economics – a noncommercial site full of financial infographics – has built up 50,000 backlinks (I think) and lots of authority
- An average infographic takes a couple of days. Mostly research time (we use students to do a lot of this)
Guest Virals / Blog Outreach – one step beyond guest blogging
- Guest virals = putting viral content on high profile sites on behalf on your brand. This is actually really easy to get big sites to host this (who wouldn’t want this kind of stuff) but ask for a link back to the client site – eg within the Embed code for the viral
- We’ve been successful with this on some of the biggest news sites on the web – they wouldn’t be interested in just a guest post (they have plenty of written content) but a well designed infographic
- Tip: reach out to people who have linked to / embedded / dugg etc your content and build a relationship with them, these are invaluable for getting them to help you with follow up content
- Track social success – analyse viral successes/failures to learn for the next time
Parks Blackwell – Range Online Media
- Having a pre-planned content strategy is very useful. Note: marketing messages don’t count! Think above and beyond and try and do engaging content above and beyond salesy shit.
- Align this with marketing calendars and PR initiatives
- Find someone in your organisation that can speak like your customers do
- Transparency. No-one wants to be a part of a community that’s not genuine. Allow feedback, talk/respond to people.
- Monitor.
- Make a decision before you start on what you want to achieve
- Understand the correlation between activities and metrics
- Be wary of changing metrics halfway through a campaign – can cause issues (you end up changing your tactics) and also can alter transparency
Good advice but fairly standard stuff about social media engagement – no real ‘give it up’ style tips in my opinion though.
Melissa Campbell – Distilled
The laws and guidelines in the UK that you’re probably breaking with your social media marketing!
- The UK Code of Non-Broadcast Advertising (CAP Code) means you can’t pose as non-client on social media sites
- People think virals are excluded from this – they’re not! You have to disclose commercial intent from a legal standpoint
- Example: Orlando Figues posted fake reviews on Amazon for his own academic books (the dufus used his university name) – then got caught and tried to blame his wife! (cheeky git, mine would have had my balls for doing that)
- Being deliberately offensive – eg 2 guys who did a white guy / blacked up guy minstrel guitar (poor Melissa’s slide’s didn’t actually embed the video but she dealt with it very smoothly and explained instead – nice one!)
- Melissa shares a slightly dodgy tactic which she asked we don’t blog/tweet about – so I’m keeping schtum
- But… this is government by a self-governing industry panel. The panel is NOT a police force! If you have a viral that you dont disclose as being for your company and everyone loves it you’re probably not going to get in trouble. It’s also open to interpretation and there’s a lot of vague/fluffly language in terms of penalties – they want to look at each complaint on a case by case basis and there’s no set penalty
- However the Digital Economy Act (2010) does cover copyright infringement. But there’s indication that this won’t clamp down too harshly on things like using copyrighted images from flickr/google images. Also Melissa doesn’t think this is going to clamp down on marketers as harshly as we fear it will – it’s mainly about illegal downloading type stuff.
- For further research Melissa has a few links to the legislation in her slides – watch out for the slides which I’m sure she will be tweeting out – so go follow her!
Good, snappy session this one. The stuff I loved the most were the infographic examples and publication techniques from Chris Bennett – this stuff is very useful to look at – you can literally copy the methodology and replicate the success, as long as you have a good enough idea / design of course!
Well that’s your lot from SMX Advanced London 2010 folks! It was insane (in the membrane) getting the full house of 10 posts up but totally worth it! Thanks to everyone for all the encourangement and retweets, I couldn’t have got through the marathon without you lot!
case studies · infographics · legal · liveblogging · smx · social media · tips · viral
18
Social Media, Search & Reputation Management: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
On to the penultimate session from SMX – no time for an intro – this was a mammoth session as you’ll see below! Get yourself a cuppa before reading through this stuff
Kristjan Mar Hauksson – Nordic eMarketing
What happens when bad news goes live?
98% of journalists go online every day. Kristjan talks about how many journalists simply grab and republish news / press releases (and blog posts I would add) and republish them as their own articles. Journalists are “thirsty for information” and they are looking online to get that info. (I would add that there is a hypocrisy about people talking about content scraping / spinning as blackhat, whereas journalists effectively do this all the time – but that’s another debate
)
Kristjan shows some amazing slides where he has identified logfile visits based on which company the visitor works from – eg someone from HSBC came in on this date to this page, someone from BBC came in on this page – awesome idea.
Think about “who am I influencing?” Is it the media, the end user/consumer, etc – and draw up a plan based on each of these.
The Proactive Approach
- have a plan, be ready with basic information and foress possible scenarios
- saturate the search space for brand terms, products etc – so you are ready to capture people trying to find out more when a problem or opportunity happens suddenly
- Be ready with draft PPC campaigns for unexpected / sudden events (eg ash cloud or others that can be predicted) – an airline should have a plan in place for what to do if a plane goes down, etc.
- What assets do you have that you can take advantage of when
- Think about what happens when people search for [your brand] + “bad service”
- Example – Icelandic travel client of theirs is expecting a really good summer as a result of the focus on Iceland from the volcano – curious people!
- A problem / threat is also an opportunity, look at what the opportunity could be from any threat
- Re tools: Yahoo Analytics offers a lot more detail than Google Analytics in terms of who individual visitors are – worth sticking on your site as it’s free.
- Claim your social media brand profiles on the main social media accounts – he mentions KnowEm on the slide as a way of doing this, and then start building a positive brand on some of these profiles
- Develop a clear social media strategy – he looks at the Habitat hashtag spam case as an example of bad strategy – they just wanted as much visibility as possible – but they pissed off the entire Twitter community in the process
- Use social media to open opportunities – being a company which engages with the public via social media eg Twitter makes you more interesting / sexy to mainstream media (articles/interviews/FREE PR) as well as guest blogging opportunities
- Real Time Search is increasingly important. Kev shows a SERP for “gordon brown” in google which shows up a tweet from none other than Stephen Tallamy talking about him but also a couple of really negative tweets about Gordon Brown
- Look out for Google suggested search – what comes up in suggestions when you start typing your brand? If negative phrases are coming up you’re getting a negative association before a user has even performed a search! How to deal with this? I’ve seen some ideas around doing a major PR push which gets people to search for a specific query, which then pushes down negative suggestions. A slightly more shady approach he’s come across is using services like Mechanical Turk to get people to search for various phrases via different IP addresses to generate volume needed for new suggested searches. Not sure if this is effective – mixed reactions – but if it does work this might not work in the near future as Google picks up on this.
- (bonus tip via Q&A) Regarding tools: “I recommend a little known tool..it’s called Google” – use ‘latest’ tab for recent/realtime results related to your brand keywords.
- Social media can be used in unique ways for reputation management and brand expansion -
- Social media has a shelf life - must be combined with long term / traditional SEO to have a permanent affect. Short term is due to what he calls the “Google Honeymoon Period” (not just QDF terms, this happens for all keywords Arthur says)
- How to use the ’social media channel amplifier’ – sorry but I can’t help you with what this is, he whizzed through this so fast I missed it. Some kind of made up name for a methodology.
- used sitelinks to take up more SERP space above the fold (no explanation of how they ‘got’ sitelinks – its not just a switch you flick!)
- Obtained a second listing for brand term – subdirectory
- Press releases – got into Google News results
- Linkedin Profile
- built up Wikipedia page
- Identify your keywords used to research you, other execs and your brand – dont focus on volume but impact.
- Analyse your keywords – check SERPS and rate the results objectively – Mikkel rates each result on the SERPs for each brand keyword on a scale from +2 (positive) to -2 (negative) and dumps the analysis into Excel. He suggests reading the Book Search Engine Marketing Inc by Hunt / Morans on p165 for the details of how to make sense of all the raw data in Excel (scoring method). You then get a score out of 100 for every keyword, every result and an overall score. This can then be repeated monthly to get an idea of how this is changing over time. This is very time consuming – but Mikel says (and I agree) that I haven’t come across a tool that does a good enough job of rating the sentiment of a webpage.
- Analyse the negative websites – find out how to minimise their negative impact. Look at: is the keyword competitive? What factors are making the site/page rank well? etc.
- Develop the right strategyies for your goals – is the strategy “damage control” (minimising negative impact) or “Tailgating” (building dominance for your brand reputation keywords).
- Clients often ask “can you remove critical sites?” – well yes you can says deMib – but there are no reliable methods – things that should harm your competitors may instead benefit them; also big brands will come back in very fast.
- What about hacking – well you could go down that route but its illegal and I dont recommend it! A side note: when was the last time you checked your robots.txt file? This is a common hacking technique -> they just block the entire site! (blimey – good point!). The other approach is pushing down negative listings by optimising.
- You can actually buy out the negative sites – many webmasters are up for this! (but use a third party to make the payment).
- Find another page on the offending domian (eg reviews sites, UGC sites, etc) about your brand that is actually positive (or get someone to create a positive page on that domain) and then build links to that to get it to overtake the negative page
- Manipulate the source – supply new ‘evidence’ to the author and try to convince them to change their perception of you
- Execute your strategies
- Monitor and benchmark your results
Kevin Gibbons – SEOptimise
Kevin starts by exploring the idea that Google is now your brand’s new homepage (2009/2010 idea). In fact for 2010/11 Kevin reckons Twitter.com is actually ‘your brand’s homepage’.
90% of consumers trust recommendations from people they know; 70% actually trust opinions from people they dont know; in contrast only 41% of people actually trust search results / PPC ads as a recommendation (taken from a survey – source TBC)
“Build brands not around products but reputation” – eg Virgin Atlantic SERP is filled with good feedback, their Twitter account is full of useful info for people. In contrast, Ryanair is the opposite – they haven’t bothered about building a positive reputation online – eg the I Hate Ryanair blog ranks on p1 in the SERPs for “ryanair”!
(I’d add that this is down to the fact that Ryanair’s product is crap, so it’s all related)
Analyse how much negative traffic your brand keyword generates? SERP visibility as well as negative keyword search volumes associated with your brand.
“What harm can social media really do?” – Kevin points at the United Breaks Guitars videos, which allegedly caused a share drop costing $180million for United Airlines!
Kev’s Tips:
Brilliant stuff from Kevin – to the point and some great tips.
Arthur Coleman – OnlineMatters.com
Arthur is not a PR person – they create and run sites for clients and takes a proportion of revenue. There’s a powerpoint snafu where Arthur realises “whoops… old version of the presentation”
Luckily he finds the right one and we’re back on track.
3 Key Takeaways:
Case study of a reputation management issue with a large telecoms company which is a client of his. The company had a particularly bad article about the company ranking in the SERPS which they wanted dealt with urgently. In order to deal with this they:
This helped to push down several negative listings, some even off page 1.
Arthur then looks at the SERP for his own name, which had a lot of other Arthur Colemans show up, including a convicted serial killer (or something like that). By building up his own social media profiles, eg Twitter, Sphinn, Seomoz, Kaboodle, Tagzania, Google profile. He now owns all 10 of the listings on page 1 for his name. We’re all very happy for him.
Arthur believes there is something going on which he calls “Author Rank” – a kind of authority metric for people / profiles. Google actually encourages you to build up your Google Profile in order to “improve your profile’s rank” – I guess he’s suggesting that more authoritative people will automatically get more profiles to rank without having to build those individual profiles up (kinda like domain authority for a person) – not convinced that this happens, I still think you need to optimise each profile – but interesting idea.
All in all I think this guy went through his presentation way too quickly and tried to give an overview of everything he knows in 10 minutes, meaning that (a) he talked really fast and it was hard to keep up (let alone liveblog – give a brother a chance!) (b) he didn’t actually go into much detail on the ‘how’, just the ‘what’.
Mikkel ‘deMib’ Svendson – deMib.com
Mikkel starts with a quote: “everyone has the right to free speech; but that doesnt mean anyone has to hear it”
6 Steps to Search Reputation Management:
An outstanding presentation from deMib – so much great info and ideas – way too rushed and condensed, but I think I managed to get most of it
(apparently it’s not a new presentation – he did the same one last year – but I didn’t hear that so personally I’m not bothered – dupe content ain’t as bad as its cracked up to be y’know
)
liveblogging · reputation management · serps · smx · social media
18
Top Ten Customised Search Analytics Reports: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
Looking forward to getting stuck in with a proper Analytics session here that will go beyond ‘look at your top referring keywords and their bounce / conversion rates’
. Standout speaker on this one for me should be Will Critchlow, self confessed Analytics and Excel geek ninja and always willing to share some advanced tips. I’m not sure how the “top ten” format is going to work – does everyone share their own top 10 or is it a collaborative effort? If it’s the former there may be a fair bit of overlap – there’s only so many reports that are useful for a search marketer. Let’s see…
REQUEST FOR HELP! I think I missed a lot from this session in these notes – partly because I’m flagging and partly because my battery started dying halfway through. If anyone has stuff to add please let me know and I will happily link to you in return (for what its worth)
Alex Cohen from ClickEquations
Alex (@digitalalex) starts by saying “Reports Suck” – the reason: creating reports gets you 0% ROI. Acting on them and optimising is the bit that actually makes you money. So his focus is on making the reporting take as little time as possible to keep time for analysis/action. Title: 4 ways that reports suck and what to do about them.
Replace the subjective with the objective. If you have some raw data/ metrics you need to ask “is it good or bad?” Go from metrics > goals > progress achieved to date > projections based on current rate of success.
Alex is a fan of bullet graphs – check out http://click.it/bullet-graphs
Top 10 Keyword Reports are Lazy. A lot of people use top 10 keywords as a report for clients, but they don’t tell you much. 6 of the 10 may be exactly the same month to month.
Lazy data makes the long tail cry. (I’m not sure how the biology of the long tail analogy holds up with this statement – it sounds a bit dirty to me). Alex quotes his ‘hero; Avinash Kaushik in saying that instead of looking at this lazy data, look at what’s changed.
How to build a whoat’s changed report: Segment (campaign, keyword, ad group); Metric; Time period. He suggests trying out Juice Analytics’ plugin for easy “What’s Changed” reports: http://click.it/whats-changed.
Alex then talks about the problem with looking at “Impression Share” data. It’s much better to look at “opportunity cost”, ie what is the maximum exact match traffic you could be missing out on for each keyword and how much that would cost.
Finally Alex warns of the dangers of being obsessed with Quality Score – it’s not the be all and end all, look at how each keyword is actually performing for you, not just QS.
A very quick tool plug that looks like one to check out is the Click Equations Analyst plugin for Excel which looks like it outputs some cool reports.
Pere Rovira – Grupo Elisa Interactive (Spain)
Very laid back start from Pere – he spends the first 5 mins of his 12 doing things like slagging of the computer / having to use IE, introducing himself in a laid back way (I like the look of Hippochondriacs Tumblr blog which he showed – need to find the link…) and making geeky jokes. I love this guy, he’s like a Spanish Professor Frink
“If you can’t figure out an action after looking at a report, don’t use it, it’s crap”
You need to figure out how users search – he shows a matrix of 4 different moods of search behaviour which I need to pull out from the slides
Segment your keywords based on keyword modifiers – eg “download” is different to “guide” – look at data for each of these kinds of groups.
How good is my copy? Pere talks about looking at the CTR from your search results to gauge how good your title tags are – he really likes the new Google Webmaster tools interface which includes clickthrough data. He then analyses impressions and CTR for various different buckets of keywords.
Showcases a tool: SEOGuardian, which a friend of his has made. It looks pretty cool – analyses and visualises competitor rankings for different keywords. Looks like it’s in
Finally Pere talks about the importance of landing pages. Are keywords connected to calls to action on the page searchers land on? Is your searcher getting what they wanted when they searched?
Use CrazyEgg heatmap tool to segment whether searchers actually clicked on the keyword calls to action that they originally searched for.
Segment your referring keywords report in Google Analytics with the report of internal searches that take place on your site. Are people searching for the exact same keyword they used in Google once they get to your site? This is a great way of figuring out whether you are failing to give searchers what they were looking for.
The basic paradigm is to look at the actual people who are doing the search (he likes to think of searches as Woody Allen, which kinda fits with this dude somehow
) Really good presentation.
Tami Dalley from ROI Labs
Tami started by saying they are tool agnostic – they just focus on what works.
Regarding the default reports in Analytics tools, Tami says “don’t let other people decide what’s best for you” – ie customise.
She talks about the importance of using tracking parameters not just for your sources but for the content of your PPC ad copy that you are rotating (utm_content in Google Analytics) – this means your report can actually tell you which ad copy is working.
Dont obsess over Click through rate. It’s ROAS (return on adspend) which should be the metric you use to make decisions.
“Dig where the gold is” - look at demographic data and regional search data from comScore and compare this to your analytics traffic. Which areas are you underrepresented compared to the opportunity, and where are you overrepresented? (She uses US regions as an example… I’ve not used comScore but I wonder how good they are about UK region specific data and how accurate this is).
Don’t look at visits in isolation. Tami shows a ‘campaign stacking report’ from Omniture Sitecatalyst which basically looks at multi-touch tracking (ie which channels visits came through en route to conversion). There’s a useful looking query that you can use in Sitecatalyst to get this report which hopefully Tami will share in the slides.
Will Critchlow – Distilled
Will starts by reminding everyone how he has twice pwnd Rand Fishkin in their recent presentation-offs at SEOmoz seminars and shows his baby daughter wearing the “My dad beat Rand Fishkin” t-shirt – totally irrelevant to this post but I had to include it here.
Will shows up the quotes from SES about last click attribution – “its more about crediting the bar where you first met your wife”. The complex answer is “econometric analysis” – “I’m a geek but I still dont bother with this”. What you really need to do is try this technique: http://bit.ly/1sttouch – distilled custom external Javascript file to use with Google Analytics which allows first touch tracking! (AWESOME).
This uses custom variables to store the first visit referring source, landing page, keyword and one other which I need to check in his slides! (sorry). An annoying thing about Google Analytics custom variables is that there’s a 64 character limit which is very limiting in terms of storing data and makes multi-touch tracking impossible. Will suggests that we all give Avinash Kaushik a slap when we next see him to try and get this changed <- LOL. However Will also added later that while multi touch sounds fantastic in theory, the data you get is actually VERY boring and messy. First touch + last touch seems to be a pretty good compromise.
The keyword data is URL encoded so has all kinds of annoying things like %26 in the keyword report – for this Will has a scary looking Excel formula which I’m sure he’ll share!
Will actually shared his top 10 search analytics reports – they are below:
- Visits to Purchase – Nobody buys from you the first time they hear about you. (except the loser with the backwards baseball cap – funny slides from Will
) – justifies caring about. - First touch vs last touch organic vs. paid
- First touch vs last touch branded vs. non branded
- Search vs. Direct visits
- Econometric models – treats values as random variables. Estimate parameters of joint probability distributions of variables being studied. I HAVE NO IDEA WHAT THIS MEANS! (I think it’s something along the lines of treat everything, from number of links to number of kittens, as equal variables within your model to iron out bias)
- New links – ie referring traffic URLs – this is so useful because as Will has said before, the most useful links are the ones that send traffic.
- Keyword Length – tells you whether you are under or over performing on elements of the long tail.
- # of Organic Landing Pages over time – every page that has received >1 visit from organic search – this is a metric of indexation that you should actually care about (hmm indexed pages that receive no traffic are still important to know about – but I’ll let it go because this is all solid gold
) - Real-time reports – not really time to explain this, I’d like to know what he does to achieve this (one for the pub later maybe)
- Heatmaps – I’m not sure if I missed this or if will ran out of time
Apparently there was something about Google Alerts which Will mentioned – totally missed this so if someone can help me fill this in that would be great!
Every time I’ve heard Will Critchlow speak I’ve been blown away by the amount of value, detail and clever ideas (usually involving using data) that he gives – and this was no exception. I really need to look at his slides again which I’m sure he will make available so make sure you follow Will on Twitter
[update: Will has mercifully put his search analytics slides up now:]
Critchlow Will, Top 10 Analytics Reports
analytics · attribution · first touch · google analytics · liveblogging · multi touch · reporting · smx · tracking
18
Bullseye! Reaching and Closing Your Ultimate Customer: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
This sounds like an interesting one. According to the programme blurb:
this is not an analytics panel. This isn’t even a technology panel. This panel is about getting to the core of all of our planning, budgeting and endless hours of work – our specific and ultimate consumer. We dare say that this will be the most diverse and unique group assembled at SMX Advanced London this year – highly specialized marketers and researchers who will give attendees an intense look at the differences between how various consumers search and buy.
I’m intrigued. With a line up of speakers including the head of Google UK’s retail vertical Martijn Bertisen, Facebook’s Head of ‘Inside Sales’ (whatever that means – suggestions in the comments please) Josh Smith, Search Evangelist at comScore Eli Goodman and the Robert Barnard CEO of Decode (which specialises in working with youth culture). I’m hoping it’s more “Bullseye”, less Bullshit. (Actually I’m fairly sure it won’t be BS but I just had to say that
)
Eli Goodman – comScore
Eli starts with an intro saying proudly “we track what eeeverybody does online” [grins, then adds hastily:] “with your permission of course.” Heh. nice save.
He goes through some search user stats. Apparently while the rest of Europe has about 30% y/o/y growth of searches online, Spain stands ouat at about 60%. No idea why.
In Europe the average no. of searches per searcher are a lot higher than in the US. Eg UK – 140 searches / searcher (per how long I didn’t catch), compared with 108 searches/searcher in the US. Is this because US searchers are less fussy / pedantic or because European search results are less relevant / contain more crap/spam?
YouTube is far and away the 2nd largest search property in Europe after Google – dwarfs Yahoo and Bing. For Eli this is a justification of Google spending top dollar on Youtube.
Male vs Female searchers in Europe: Males search more everywhere (52/48 split in UK) but much bigger gap in Italy / Spain.
Bing is the only search engine where more females search than males. Weird. Is this because more women use hotmail (I’m always trying to get my missus to switch to gmail, she doesnt like change tho) and then ‘Bing’ from their inbox? (that sounds a bit dodgy).
Martijn Bertisen, Google
The first thing to make sure of is that you’re being found. If you can’t be found via search you’re not going to make sales. He claims that 25% of searches lead to a purchase. Really?
He shows some stats about how many billions of people search per day (3.6bn) and how widespread online video has become. He wants to focus on realtime insights and points to the tools they’ve been building eg Youtube Insight, Insights for Search, etc.
Story/case study re a young champagne brand looking to launch a campaign using a celebrity. The problem is which celebrity to use. Starts with a simple search “US socialite”. Then chooses video. You can see people like Kim Kardashian. He then moves over to the Wonder Wheel feature (which he loves) – gives associated terms to what you typed in. Martijn says “I dont even know who Kim Kardashian is, maybe I read the wrong papers”. He then googles her name and pimps out the wonders of universal search a bit. The goes to her Youtube channel and wonders over how many millions of views she’s had and how many subscribers she has (23k). ”Wow, we might be on to a winner here”. (No, Martijn, trust me, she aint no winner). He then uses Google Insights for Search to look at demographic segments of who is searching for her. Then he moves to Google Ad Planner to really drill down into the demographics. Combines “Kim Kardashian” with “Paris Hilton” and looks at males over 55, which in this case is the client’s target audience. Ad planner says there are 43,000 users in this bracket interested in these people, and tells you which sites they visit. (This is all pretty cool, I’ve never really played around with Ad Planner before).
I enjoyed this little ’story’ from Martijn. Yes essentially it was a plug for Google tools but the way he presented it was useful and showed you how you can go on a journey with them drilling down to get more detail and make a decision on your strategy.
One tool he mentioned is “Insights for Audiences” which is something that is in beta at the moment – one to look out for.
Mal from Wordtracker asks a good question – does the volume data include all searches or just a proportion? Martijn replies that the search volume data from Insights for Search uses all searches as a data source, other tools ag Ad Planner use sampled data.
Josh Smith – Facebook
Josh starts by talking through some stats on how widely used Facebook is – didn’t catch the stats but you know the drill – eg “World has 6bn people, facebook has 8bn users, we run everything” etc. (not me you dont facebook! hahahaha)
Facebook is no longer just for young people / students / teens – all age groups use it now.
He then walks through the facebook ad demographic planning system. You’ll have to excuse me for not reporting this, I’m glazing over slightly. The basic message is “we know everything about our users. Whether they’re wearing odd socks today. What time they’re going out tonight to cheat on their wife. Who they voted for. You can target the exact person you want to advertise to.” If you want more detail please go check out their sales stuff.
Robert Barnard – Decode
We’ve spent 16 years figuring out how young people / generation Y ‘think’ (I just had a mental image of a bunch of marketers hanging out in a youth centre on a council estate. And then running away.)
Shows a picture of Lady Gage who got 2 million hits on her Youtube vids, and OK Go, who got 50million. Emi just dropped OK Go because sales were low, and Lady Gaga’s got so much money she has made a gigantic turnip hat entirely out of diamonds. OK Go basically couldn’t sell records because they didn’t create a sustainable, consistent brand for young people to latch on to.
The average number of online friends that Generation Y (15-29 yr olds) have regular contact with (>once/month) = 43. (I guessed 2000 and made a tit of myself). In fact 35% have less than 10 online friends. So dont believe the hype in terms of hundreds of friends and the reach of getting to young people online.
Shows a great graph of a timeline from high school > after secondary school > singles > couples > parents and how number of friends (online, face to face and phone friends) changes. I’ll try and link to it at some point, basically online friends takes a U shape – highest during school, then goes up again at the parent stage. (Our moderator, Misty [real name], points out that for most of us geeks we didn’t actually have that many friends at high school and have the most right now
)
Q&A Session
- Q for Martijn re google’s tools: you demo’d US stats, are they available for all countries
- Ad planner doesn’t just look at sites in the Adwords Content Network, it goes wider
- Facebook Q: If you target a user based on a keyword they used, how long would it target them after they used that word? Answer: they are tagged with that keyword for 30 days after using it.
- What’s the clickthrough rate for Facebook ads? Answer: it varies a lot. If you target cheap Lady Gaga tickets ads to Lady Gaga fans, the CTR can be incredible – if you do general stuff its a lot lower. But make sure you run a lot of ad variations and compare the conversions based on ads + demographics.
- (i missed a bunch of other questions because I was arsing around on Twitter. sorry)
- I asked Martijn from Google a question (didnt miss this one!) about personlised search results and whether / how Google is gathering Facebook style demographic data on search users and whether this will be made available. Martijn just said “this is s a really tricky subject re privacy” and then claimed that they only personalise search results for logged in / signed in users. Really dude?
(I did give him a chance to correct himself, I guess there’s some blockages in the communications pipelines there)
consumers · liveblogging · research · smx
18
Proving Social Media’s Value: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
This is probably one of the big unsolved problems in digital marketing right now. Social Media marketing is still in its honeymoon period where everyone likes the look of it and wants it ‘done’ on their site – much like my 1yr old baby neice is ‘drawn’ to all kinds of random stuff like phones, glasses and anything shiny, only to grab it for three seconds and promptly drop it again. They love how certain stand-out brands seem to actually have turned their customers into fans and evangelists for their brands. But in terms of proving the value of social media ‘activity’ it’s all still remarkably fluffy given the amounts of budget being spent on it. For me the brands and marketers who make it through the recession (and make no mistake, it is hitting digital) are the ones who can genuinely prove its ROI and benefit to the brand. So I’m looking forward to hearing from some great speakers in Social Media such as Mel Carson who’s done a brilliant job with social media marketing for Microsoft AdCenter (have a look at their B2B social media white paper) and one of the big cheeses at Econsultancy Linus Gregoriadis. Who knows, someone might mention that social media gets you links, which for me is surely one of the biggest returns in terms of leading to actual sales/conversions via SEO.
Linus Gregoriadis – Econsultancy
Linus will be looking at the business impact of social media, what metrics can be used to measure it and whether it’s possible to determine an ROI from it.
According to their forthcoming survey, 64% of companies say they cant measure SM ROI as much as they would like to.
65% of companies planning on increasing their social media spending in the next year, with 56% planning a 20% increase and 15% planning to double their social media marketing spend.
Linus takes us through a number of other statistics taken from the report, including social media sites / channels used. Twitter is now the #1 social media site at 84% (Facebook is 78%, Linkedin 52%, Youtube 48%). Last year twitter polled only 48% and the year before it was 3%.
There are some other stats relating to what companies say they use these channels for, eg branding, product promotion etc. I’m guessing not many would admit to using social media for “spamming the crap out of unsuspecting customers”.
On to how respondents talked about how they track ROI from social media. 52% use clicks/visiors, 24% use position (branding) and 22% dont bother tracking at all. Number of sales (22%), leads (21) and Value of sales (14%) are notably low down the list.
Linus quotes Jim Sterne: measuring followers on twitter is the same as measuring hits on a website. Its interesting only because it’s a number, but it’s not actually useful. We need to find a performance indicator using a combo of factors.
Key takeaways:
- There’s no simple formula for measuring value from SM like there is in SEO or PPC
- It needs to take business objectives into account
- Search marketers can’t ignore social media; but it’s not a sub discipline of search marketing
Charlie Osmond, FreshNetworks
Charlie is going to share some examples of social media tracking.
The problem with social media ROI is that companies act like 16yr old boys – they treat it like a series of one night stands, a sale = pulling
(paraphrased from Gary Vanyerchuk apparently)
Types of metrics: web analytics data, community health metrics, outreach metrics.
Example 1: TM Lewin - Off The Cuff community
High range designer shirt manufacturer. FreshNetworks created a community around this (NOT a social network, more a place for content – eg “how to iron a shirt properly” video
- 22% more time on site
- Ppl who enter the community are twice as likely to convert to a sale
Example 2 – Vets Now – emergency vets
Built a community where vets could discuss problems and issues together of strange emergency cases. Part of the aim was recruitment.
- In first 6 months they’ve doubled the vet database and saved £20k costs in recruitment
- One vet had an unusual emergency with a dog in middle of night and didn’t know what to do. He sedated the dog, uploaded a picture and asked other vets for help. They immediately came to his rescue and the dog ended up surviving. This leads to the soundbite “social media saves dogs’ lives”
- 60% of vets sign in every single day, 95% sign in weekly
Example 3 – Butlins
Helped them run auditions for Redcoats on Youtube – instead of closed interview type auditions, it was publicly viewable to the world.
- 3 week Redcoat campaign was a great offline RP hit – drove a reach of “53m” (whatever that means in fuzzy PR language)
Example 4 – Jimmy Choo
“the coolest social media campaign” – mailonline
Raised a shedload of money for an aids charity in South Africa.
Set up a photo community – ‘you, your shoe and 72′
- 3 weeks from sign off to launch
- Compared to facebook (v big profile), the community was 4x more effective in driving sales (the community was much more appropriate for asking people to buy than facebook)
- Community visitors spent 2x more time on the main ecommerce site and twice the
- Overall ROI for the campaign was 20x more than the ROI of their online advertising
They also set up a 4square campaign involving women running around London chasing a pair of Jimmy Choo shoes that was moving from place to place – nice idea! <- lots of PR benefit from this .
Awesome talk from Charlie, LOADS of great case studies compressed into ten minutes, clearly these guys not only have great ideas and how to execute them but they’re confident in showing what benefit the clients have had from it.
Ian Strain-Seymour – BazaarVoice
Ian shows a Forrester diagram that looks like spaghetti junction starting with “Eyeballs” going into a funnel, going through a crazy maze and coming out with “Buyers” and “Contributers” on the other end. Firstly I hate it when marketers talk about getting ‘eyeballs’ – aside from being a bit scary it also makes me want to gouge mine out and hand it to them so they go away. Also that chart reminds me of that insane army ‘infographic’ that makes things more complicated, not simpler.
Um…oh right – stop ranting Jaamit.
Key metrics for “Voice of customer” (that phrase again from yesterday):
- Reviews, coverage depth, helpfulness, comments
- Questions
- Answers
- Blog / forum posts
Ian says that measuring how much you’ve put out there eg number of tweets etc is for losers as it “doesn’t actually show whether you’ve been successful in getting people to actually visit your site, there’s no ROI”.
He then goes through a number of different metrics to use including a ‘cost per click’ metric for social media, conversion rates, returning visitors etc. Maybe it’s his reporting style but this isn’t giving me anything about tracking social media value – it’s just a bunch of metrics that could apply to anything. For example he flashed up things related to purchases and returning visits – but how have you determined these came via social media? What campaigns were they based on? All a bit fluffly if you ask me.
Mel Carson – Microsoft Advertising Community
“It’s social media marketing, not social media”
Mel talks about the ‘Learning and Earning” social media white paper, which started as an email, then a document then a full blown white paper. He encourages us to “Bing” ’social media white paper’ to find it, but “for the few of you still using Google you’ll be able to find it that way too”
Mel deals with the social media not for Microsoft as a whole (2-3 mentions per second!) but just Microsoft Advertising.
They launched their blog & forums in 2006 – the “official voice and calling card” of MSN Adcenter. They used it to ask for feedback and feature requests, sign up beta testers, and generally help advertisers to help themselves by answering their questions (this also led to service cost reduction
Constantly looking at new platforms and technologies both for their blog and in general in terms of how we engage.
Microsoft didn’t expect us to need to go out to 3rd parties – eg other blogs, Sphinn, forums etc – this took them by surprise as they had thought it was all about selling the product.
Channels they use: twitter, facebook, flickr, youtube. But Mel says don’t just jump in and say ‘we need to be everywhere’ – ask yourself why you want to use each service, and look at the cost / time input required and whether it will be worthwhile.
We’ve also started seeing the value of working with PR teams, eg to help write blog posts, and particularly in creating content and RSS feeds for press contacts (we don’t always know how to write for them)
Impact on search – if you search for “b2b social media white paper”, they own seven out of nine listings in the SERPS
We use lots and lots of measurement tools. Eg buzz / sentiment tracking and analysis has been very useful to show Microsoft the value of what we’re doing
Twitter has been great for getting feedback. Mel flashes up a few example tweets from the community including a couple of tweets from Paddy Moogan which made me grin.
7 Value Indicators that We Track:
- Measure growth and translate as ‘reach’
- Think about how you’re lowering costs – every problem that the community solved instead of needing a call to the call centre is money saved. (Apparently American Airlines saved $10m this way!)
- Information and empowerment leads to increased spend
- Internal education leads to external evangelism
- “Earned media” – Windows 7 Launch = 221m impressions (hmm I wonder how much of this was ppl slagging it off)
- Think ’social media marketing‘ – be disciplined (dont just fanny about) and get other depts to think about it
- What’s the Return on InAction – if you’re not there / present / visible, what would that cost you?
Q&A Session
- Charlie: Freshnetworks have just done a roundup on different monitoring tools. Scoutlabs is simple, Nielsen Buzzmetrics is fully featured. Free report will be put out by them so make sure you look out for it.
- Linus: mentions a bunch of tool names but also says Econsultancy have a report on social media monitoring tools
- Mel: repeats that they use loads of different tools, depends on your goals. He mentions a new tool they’re working on called LookingGlass which allows you to overlay different kinds of data – eg sales data with mentions – sounds interesting.
- Charlie: when measuring the growth of your community, dont forget the 1:9:90 rule – out of 100 people there will be 1 who really engages and contributes heavily, 9 who contribute a bit and 90 lurkers. So it’s not just about raw numbers
- I asked a question about why hardly anyone talks about links being one of the outputs of social media that leads to real ROI via SEO / rankings – why did no-one mention this and do you have case studies for this? Mel said he did mean to mention this and they have had this benefit; Charlie says he did mention it (sorry!) and that this was part of the increase in traffic and conversions driven by social media. I still dont think anyone has really grasped this important outcome of social media properly.
Overall a good session, Charlie’s being the highlight.
liveblogging · roi · smx · social media
17
Keyword Research: Beyond The Ordinary – SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
Woah. On to the final session of day 1 at SMX London. My brain currently resembles some rather watery scrambled eggs so I’ll try my best to give you the highlights.
Christine Churchil, KeyRelevance
Christine is going to be looking at a technique to mine data and examine long tail keywords. This technique relates to the concept of an editorial calendar which is used in the publishing industry. This is basically a timeline of editorial stories planned out for upcoming issues of a magazine / newspaper – used to attract advertising.
The same concept can be applied to keyword research. You can use it to plan which keywords you will be targeting with all the stories scheduled for the future (ie you need to start with a content plan I guess). Look for forthcoming events, seasonal trends, launches / publication dates etc.
The point is you can be doing the keyword research ahead of time (tools she mentions are Adwords KW tool, Insights and Yahoo Buzz) on topical events and start writing content / producing videos / strategising ahead of time. Example (US-specific) – Renaissance Festival (RenFest) which features a bunch of total dorks dressing up in medieval shit [I may have loosely paraphrased that].
Basically what she’s saying is ‘plan ahead for topical events with related keywords’. Not really an advanced keyword research tip, more an SEO planning tip. ”It’s so painfully simple that some people just don’t get it”, says Christine. [Erm..would those be the same people who go to Renaissance Fairs?]
She talks about using Twitter trending topics to look for keywords. [Um, yes, but that doesn't quite fit with planning ahead does it?]
One final tip which I think is useful: create ‘evergreen’ content for topical terms all year round so you have a page waiting to capitalise.
Barry Lloyd – Microchannel Technologies
Barry’s talking about “Finding Great Keywords” using broad and phrase match. Let’s see if this is ‘beyond the ordinary’.
Um – not really. He says:
- use kw research tools (duh),
- both free and paid tools
- Use analytics – very important (really? tell me more!)
- Find keywords that actually convert (never heard that one before)
The dude then goes into PPC mode and starts talking about broad match buckets putting things in exact match keywords. Not sure he realises he’s in the SEO room. I guess the basic point is, use your paid search campaign to help you identify keywords for your SEO campaign. However I think I might just be tired and cranky, Rob Hammond next to me says he got some useful ideas out of it so sorry if I’m being harsh!
Sam Crocker – Distilled
Sam’s presentation is entitled “Becoming a KW research Ninja” – sounds much better Sam! His entire presentation is based on a quote from Tom Critchlow cracking the whip and saying “I don’t want you to work harder. I want you to work smarter” (just outsource all your SEO Sam!)
Sam is going to go through a number of kick ass tools and how to use them smartly, including Xenu (there is an awesome ‘Xenu kicks ass’ graphic that I want to get!), Mozenda and others.
First Ninja Tip is Google Suggest. It’s a great way of finding long tail KW opps. But it would take all day to go through all your keywords and what google suggests, and it would take years. Enter the magic that is Mozenda. Take an excel spreadsheet of your keyword list, upload it to Mozenda and stick in the Google suggest XML output. I’m not entirely sure how to get this XML output from Google Suggest but this tip sounds awesome. From the Q&A Sam confirmed that he is going to share the URL to get XML output for Google Suggest in his slides – look forward to that one. update: here’s the link!
Ninja Tip #2 is using Mozenda to trawl through user generated tags (ie folksonomies, what people use for bookmarking) like Delicious and Hulu (in the example of a TV show like ‘glee’). Again not entirely sure how to set this up (I’m assuming there’s an RSS feed you can get to, but another great one to look into.
Ninja Tip #3 is ‘the Popularity Contest’ – essentially look at other sites (incl competitors) which produce lists of the ‘top pages’ or even ‘top searches’ and then use Mozenda to go forth and gather data using these lists. This combines keyword research with competitive analysis.
Other Ninja Toys and How To Use Them:
- Xenu Link Sleuth – crawl entire websites and look at the URLs (assumes kw focused URLs)
- Microsoft IIS – like Xenu on crack – use to crawl entire sites and not break your computer / server – also check out SEO Toolkit
- Mechanical Turk. No time to go through results, no problem. 24/7 workforce for hire
- 80 Legs – similar to Mozenda but more user friendly? Sam himself needs to check this one out more (me too)
- Your noodle - use it! its the most important keyword research tool.
Great stuff from Sam, Ninja skillz as expected from a Distilled SEO warrior in training
[Update: Sam has published his slides along with a little blog post about keyword research - see below!]
Crocker, Samuel, Keyphrase Research Beyond the Ordinary [Final]
Andy Atkins-Kruger – WebCertain
Andy focuses on international SEO and will be looking at international keyword research for this session.
Similar to the point he made at SES, the main point with international KWs is you can’t just Google Translate them. ’Casseroles’ in English refer to the food / meal / recipe. But in French ‘casseroles’ are the actual dish that you cook them in.
Andy also warns about Google’s KW research tool – he has suspicions about whether the search volume data is sometimes just plucked out of the air. [casseroles] in French / France using the Adwords tool shows
The length of the long tail varies with different countries. People assume that English language has the longest tail. But in fact, German and Dutch languages have the longest tails because of how words are constructed. So you need to look for more long tail opps in these languages.
Q&A Session
- Someone asked how to get keywords included in Google Suggest. Sam says he remembers a blackhat test that some people tried (but it didn’t really work) – he’ll pass a link around later. Andy talked about Google suggest being based on searches being related to one another (I think he meant one keyphrase repeatedly being searched for after another) being a factor as well as just raw search volume
- A question about KEI (Keyword Effectiveness Index) and how to measure the importance of keywords. Andy is not a fan of KEI, partly because it doesn’t actually tell you whether a keyword converts or not. A better way of looking at it is how many competitor sites are using the keyword in their page title.
- Do you use semantic keyword indexing tools for keyword research? Andy: we prefer to use humans, they work much better
- Christine: try using the Yahoo keyword tool as well – while Google gives you alternative phrases, Yahoo’s one also includes phrases /sentences with your keywords ‘embedded’ in them
- Kristjan from Nordic eMarketing (the moderator) pipes in with a reminder to use Bing’s research tools which also include age and gender demographics, and also Yahoo Analytics which apparently gives you more detailed (demographic?) data to drill down to from your keywords.
Time for a swift pint! I’m utterly knackered, thanks for sticking with me folks. If all goes to plan I’ll be back tomorrow
17
Linkbuilding Outside Of The Box: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
Time for the session I’ve been looking forward to the most from the conference, hoping for some juicy linkbuilding tips. No time for an intro – straight in.
Andrew Girdwood – BigMouthMedia
Andrew kicks off by talking about how the linkbuilding industry is killing the web – “The Inconvenient Truth of Link Building”. Basically there has been such a flood of shite content, press releases and scraping just from the point of view of getting links that the web has been polluted. Links are supposed to be a vote of confidence, but this has been muddied.
Andrew proposes a methodology for filtering out natural links. Start by asking whether the link is on a trusted site or not, if not, the link should be devalued. If yes, ask “is the rate of growth for links to the sites natural?”, and also “has the link been ‘built’?” – his argument is that if the link has been ‘built’ ie manufactured rather than naturally acquired it shouldnt be counted by search engines. Personally, I think the onus should be on search engines to be able to figure out what counts as a vote and what doesn’t – if you don’t like how sites are ranking based on certain ‘built’ links then that’s fine but they’re not going away anytime soon. I think a lot of built links are perfectly natural and editorially given based on the content and the effort required. Just as a lot of ‘natural’ links are manipulated via offline relationships, PR and backscratching and have little to do with editorial approval. I think Fantomaster would probably say “who cares? if it works it works, leave the rest to Google to figure out”. :)
In terms of natural signals Andrew is a fan of Facebook like buttons as well as other social media sites and Google Reader likes/recommendations – these are things that help people vote on content which puts it in front of others. I think the suggestion is that these are becoming better signals than links. My reaction is that these can be gamed just as much (if not more) than links. Social media ‘likes’ also dont give you the same long term editorial vote as a physical link (better for QDF IMO).
Definitely food for thought from Andrew here! I think I’d like to hear more about exactly what he means.
Dixon Jones – Receptional
Dixon starts with a plug for Majestic SEO. He is going to go through some ways of using the tool. He shows the cumulative comparitive backlink tool as a way of analysing the competitor landscape in a niche.
Dixon demonstrates a way of visualising anchor text within Majestic – a raw dump of anchor text of backlinks in an Excel is not all that useful. Going from how Avinash Kaushik has talked about using visualisation tools, he suggests dumping a list of anchor texts into TagCrowd to get an idea of what the key themes and words within the anhchor texts are.
Dixon then talks about reclaiming lost links by pinging your backlinks and determining which ones are 404ing or have 302 redirects. Not sure whether this function is a feature of Majestic or whether he is using something like Xenu to crawl a list of backlinks (looks that way).
Re Quality vs Quantity links – Dixon thinks Google still hasn’t figured out the correct balance of filtering out ‘bad’ links (in agreement with Andrew) but there are very few links to any site that individually provide a lot of link juice. These are the ones to look for when analysing competitor backlinks.
Don’t just look at the backlinks to the homepage or entire site – look at backlinks to individual inner pages – this is a great way to find golden links to competitors.
Good tips but nothing really ‘out of the box’ in my view – just how to use Majestic for competitor link analysis.
Q from Rob: is there a correlation between backlink domain diversity or number of Class C IPs and rankings? Dixon: it doesnt take a genius to figure out that thousands of links from the same Class C IP are basically controlled by the same country, so it is a big signal.
Kelvin Newman – SiteVisibility
Kelvin is going to share 17 (yes seventeen) Ways to get links from university and government sites. This is exactly the kind of thing I was hoping for from SMX. Kelvin stresses that there isn’t inherent benefit from an ac.uk or .gov but of course these tend to be
- Reach out to bloggers. This is a great tactic in itself, but realise that there are a lot of academic blogs hosted on university sites. Follow/read them,
- Offer a student / staff discount, then ask to be mentioned on their website (personally I can vouch for this technique as one of the best ways of building authority links to a site).
- Give the university some good press – they all have press departments looking to showcase good things people are saying about this
- Participate in a scheme – eg industrial placement / sandwich schemes where they provide an intern / graduate. These were then featured as a case study on their site. So you get business benefit as well as link benefit from this
- Sponsor a student event (it’ll cost peanuts to do) – get a link back
- Deliver a careers talk – it’s hard to get a job as a graduate, so they love guest speakers from every sector. As well as the obvious benefits to your business of attracting graduates, you can also get links from their careers site and events calendars
- Advertise a job via their careers service – get a link back to your site – again this clearly has a business benefit
- Become a case study for their business department – eg MBA courses and business studies degrees can analyse the company -> links
- Get boycotted! A little bit of bad publicity doesn’t hurt much
– eg Sussex Uni (where I went as did Kelvin) has a reputation for being quite activist - Consider launching a community website – government sites tend to link to these – eg local organisations – secondary link juice this way
- Set up a charitable website – it’s much easier to get links without commercial content
- Some government sites do have business directories! They have the highest editorial standards of any directories I’ve ever seen, but they are out there!
- Put on an event – the more local the better, again one of the few commercial sites a government site will link to
- Start a campaign (it helps if the council agreees with the cause) [on a side note, great image of an I <3 Myself badge - actually all the images are top drawer as always with Kelvin's slideshows]
- Job websites are one of the few types of purely commercial sites that government websites will link to
- Run For Parliament (LOL) – this could be an outrageous PR stunt, many authority sites including government ones will link to
- Tip 17 was missing which Kelvin blames on the birth of his daughter (fair enough man, you’ve just delivered your weight in gold there). But he does say he’s working on an encylopedia of linkbuilding which will have a few more. Cant want to check it out!
A Question from Rob to Kelvin: you might have problems eg getting your university radio site to link to your viagra site… Kelvin agrees and makes the point that all of these techniques are not purely about getting a link – they have to be more than that and have some kind of viable relationship element to them. Good point, well made.
Really impressed with this one, well done Kelvin! I wonder whether UK universities and government sites are going to be flooded with approaches from SEOs pimping out their clients now. I doubt it, because (a) people tend not to follow up on linkbuilding tips, and (b) most of these require some actual effort to build a relationship / do some graft to get the benefit, that rules out a lot of people. If you’re one of the remaining few you’re on to a winner.
[Update: Below are Kelvin's slides, complete with cool pictures. I could of course have embedded this above the notes, but of course you wouldn't have bothered to read them then!]
John Straw from Influence Finder
John is showcasing a new tool for finding influential link sources, Influence Finder, which SEOchicks has exclusively scooped earlier today – check it out! The basic idea behind the tool (which frankly looks pretty awesome) is to take the raw data provided by backlink tools to narrow it down to what are the most influential links within that backlink profile.
The first thing the tool does is to collate lists from different tools (starting with Majestic API) to get the maximum data set, and then verify which still exist.
Then the tool tries to identify which of the links come from blogs, the idea being that blogs are strong places where influential links can be found. This is done using a decision tree, which has a 94% accuracy of determining what is a blog and what isn’t looks pretty clever.
Then the list of blogs are sorted to filter out which are active, ‘alive’ (ie not auto feed bots) and not lapsed bloggers. Lots of scientific method involved here.
The results? They reckon the algo the tool uses to identify linking partner sources is very good. He shows some examples of an analysis on Econsultancy who wanted to know which are the relationships that are most worth building up, ie the bloggers that are more influential.
Also he says that MyDeco have tried this and the linkbuilding that used to take them 1 day now takes 1hr with this tool, apparently.
I think this is a fascinating tool and one I look forward to having a play around with at some point. The important thing to note is that this isn’t a way of finding authoritative links – you can do this with metrics like PR, mozRank etc. But this is about finding influential bloggers to build relationships with. It’s something that we SEOs (and indeed social media marketers) do manually – it’s a time consuming but very important process. If it does as good a job as it seems it could revolutionise the process of identifying blog partners.
Apparently social media influence tracking is on the way but not yet part of the tool. Another mention for Nichola Stott who asks about the “salmon protocol” – personally I have no clue what this is, it sounds like a Monty Python sketch to me. That girl is like a walking encyclopaedia, seriously.
Final Tips:
- Build relationships (eg bloggers)
- The links which are harder to get are the most valuable
17
Leveraging Digital Assets For SEO: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
Rob Sheppard – Ask Jeeves
Rob represents Ask.com which was the first to implement universal search amongst the search engines. He starts by stating he’s not an SEO expert so won’t be going into the optimisation stuff much himself. Fair enough. iving an overview behind the concept of including non-text / non-standard results in the search – they found that their users really liked them – more relevance for some queries, freshness and variety of content.
While universal means there’s less space for your regular rankings to show up, there’s also more opportunities for added visibility. He shows examples from Google, Yahoo and Ask for the query “how to mend a bicycle puncture”. All use video results on the SERP, all from sites like Videojug. He says a key issue to deal with is the fact that very few authoritative sources of video keep showing up. He then looks at an election SERP, looking at results for Brown, Cameron and Clegg in Google, Ask and Bing all of which show a variety of different universal results, including biographies, news, images, video etc.
He talks about why users like universal search – this is based on polling and user data of Ask’s searchers.
How can I make the most of opportunities from Universal Search? He goes through the standard basics: upload to the big sites that actually get shown (youtube, Videojug, etc), use RSS feeds to your content,
There are a pile of text heavy notes in the slides – hopefully these will be available later and you’ll be able to see it for yourself.
Social media sites are taking universal search from a ‘nice to have’ to a standard feature of SERPS
Key tip: we believe the most important factor in search is answering questions. So always try and answer questions in your content – look at Q&A sites like Yahoo Answers, etc
Question from the floor: if you have lots of videos and a video sitemap, should you also upload them to Youtube to get them ranked in Ask Jeeves? Rob answers that if you have thousands of videos it’s probably not worth it; we’ll try to find the content on your site. But it’s definitely a good idea to be visible on youtube as well.
Ian Strain-Seymour – Bazaarvoice
Ian is going to talk about ‘customer voice’. He asks who knows what this means, a few reluctant hands go up. It sounds like a fancy word for ‘reviews’ to me
What is the voice of a customer:
- authentic conversations
- you and the customer
- B2B with supplier
- engineered fake conversations between shill accounts set up by marketers (OK, he didn’t really say that, I made it up – just checking you’re still paying attention)
He talks about customer reviews in microformats. They’re still in very early days and its worth getting into from this stage if you have reviews on your site. He displays the reviews shown up from a search for “nikon d5000“.
He then mentions using realtime search. “If you have a strong following in social media, put out questions to them about your products and let Google do the hard work”.
To be honest I’m losing motivation to keep up with covering this talk. He’s basically saying “look at this, you should do it too”, speaking very fast and not really adding much we don’t already know IMO. At one point he said “if you don’t understand you’re welcome to leave the room”. I think this was a joke, but no-one laughed. As far as I know, no-one left the room either.
A Question about twitter leads to a gem from Rob Kerry: some people are trying out leaving HTML links in twitter streams – sites that have twitter streams on their sites then conver these into links! Awesome.
(someone) – SundaySky
Sorry, I didn’t catch this dude’s name – he is a replacement speaker from the guy billed to speak. He starts by looking at heatmaps on universal SERPs and the fact that videos in blended results attract people and get more clickthroughs. Has anyone not seen this already?
He mentions that getting a video result improves your chances of ranking well – I’m not sure if he’s suggesting that if your youtube video ranks in universal search there is a knock on effect on your standard result… have submitted it as a question but would be interested to know if anyone has found this.
He looks at an example of Overstock.com using video content on their product pages. Goes through some tips for using video content:
- Each video should have its own URL
- Submit a video sitemap with compelling thumbnails – don’t leave it up to google to choose your thumbnails
- Keep video section as frequently updated as possible
- Add as much video content as you can – more = better chance of ranking
- Have a dedicated videos section on your site
My question got asked re whether he is saying that having a Youtube video indexed actually helps your standard organic result. He says his gut feeling is yes
Rob Kerry – Ayima
Rob is talking about Google News and how it helps with SEO.
G News is a short term way of getting into the Top 10 for your most competitive terms.
Also, if G is starting to take more notice of clickthroughs to particular sites, surely clicks on News results can help your sites’ overall CTR metrics. This also applies to personalisation – if you can get quick visibility / clicks via news results
Rob goes through some technical requirements for getting into Google News – s
Images, Images, Images – these are so important with Google news – each article should have a related image with standardised size, 300×250 or (?) -ie not a specific size, just the same size every time. You can get visibility just through an image not a headline and this often gets the most clickthroughs.
Multiple authors are required to get into Google News to weed out smaller / less authoritative sites. An automated / greyhat example of how to fake your way in is use services like Textbrokers, getting housewives etc to rewrite latest news stories to generate content for your site. Also fake author names are FINE
Q&A Session
- Question about getting rejected by G News editors – can you then get reconsidered easily – Rob says you definitely can – one example based around the fact that G News editors are based in the US so you can take advantage of that. He’s replied to them basically saying “How dare you, don’t you know that this is actually a really authoritative site in the UK” – that seems to have worked and got it reconsidered. “So you need to be Billy Big Bollocks with them sometimes”. The other more corporatey speakers look a bit uncomfortable at this point. Legend.
- There are some questions about building links to assets like images and videos to help them rank. There is general agreement to this. I’d suggest speaking to @NicholaStott about this – had a quick chat with her about this very subject over a fag at lunchtime and she definitely thinks that’s a winner. She’s also written a blinding post about this on State of Search. (there – how to get into liveblogging notes without actually speaking at the event!)
- Question about whether number of views to videos / channels affects visibility. It certainly does. Rob then talks about using fake IPs / accounts to generate fake views and how the price of proxy servers would go through the roof if this worked. Dixon Jones says proxy servers are generally very underrated. This greyhat exchange between Rob and Dixon causes some more uncomfortable shifting from others on the panel – LOVE IT!
- If you have a well ranking page and you want to add a video , should you create a new landing page for the video or put it on the existing page. The guy from SkySunday (sorry I will get his name eventually) said put it on your existing page, unless you can’t for some reason.
- There was a question about Google News: (a) if you use news sitemaps can you get around needing a 3 digit code in the URL, (b) do i need to separate News content from Blog content to get it indexed. Rob answered (a) no you still need them and (b) yes separate them otherwise you might only end up in Blog search which is crap. I disagree with Rob on both answers! You can override the need for digits in news URLs using a News Sitemap (I’ve done it and been accepted without a problem), and (b) Blogs can and frequently do get accepted in Google News (eg Fresh Egg blog) and I think you’re missing a trick by excluding all your blog content this way! (see, I don’t agree with everything Rob says
) - I submitted a question about services that offer fake views on videos to ramp up their power. Do they work and how smart are the video sites at detecting these. Rob mentioned Subvert & Profit which offers exactly this, including star ratings etc. However these are very risky and do leave a footprint. Digg has recently banned such a network and claim to have a very mature algo for detecting this. A better one would be to enlist one of many social media professionals who have strong networks to help increase views etc – much less risky.
- A little blackhat reputation management one from Rob: you can flood out negative mentions of your brand on Twitter using a network of fake twitter accounts and basically just drowning the negative stuff out with noise
Awesome session. Mainly because of Rob’s (and Dixon’s) contributions, but the dude from SundaySky had some pretty useful video optimisation tips.
liveblogging · local · smx · universal search · video
17
SEO Ranking Factors in 2010: SMX London 2010 Recap
View Comments · Posted by admin in liveblogging, smx
The first SEO session of the day and we’re on to the question that will never die so long as search engines exist, ranking factors. A great lineup for this one, including blackhat legend deMib (who I’m assuming will give us the real deal, non-Google-approved shizzle on whats helping him rank); Rand Fishkin who will no doubt be referring to the SEO Ranking Factors survey (although I hope it goes beyond the frankly ‘bored now’ question mentioned in the conference blurb, “how much does that H1 tag really matter”); Rob Kerry who I know will be interesting and talk from his extensive experience of working on some of the most competitive niches out there; Will Critchlow who’s moderating the Q&A but I’m sure will pipe in with some interesting observations.
Rand Fishkin – SEOmoz
Rand starts by saying he’ll touch on the ranking factors survey but has got a lot more to share – good stuff! He mentions Bill Slawski’s post about the ‘reasonable surfer’ patent granted recently, and stresses the importance of reading that. Whereas SEOs have been using the original PageRank Random Surfer model as the basis for our work, which says that every link on a page has equal weight in terms of PR passed. However the Reasonable Surfer model takes into account how likely a link is going to be clicked in attributing weight. According to Rand his ranking tests, links higher up on the page do pass more weight than those lower down. More generally it seems to be the case that links that are more prominent also correlate to better rankings. I think it’s easy to get carried away with this – I still believe on the whole a link is a link and just because a link is tucked away within a paragraph at the bottom of the page it doesn’t mean it’s less important than the massive flashing ‘click here’ banner at the top. But it certainly is interesting that Google *might* be considering this as a factor.
On to Twitter data influencing SERPs, a really important issue IMO. According to his observations, links from twitter certainly do affect ‘Query Deserves Freshness’ (QDF) results, given that both Google and Bing have a data deal with twitter now – in spite of the ‘nofollows’. Rand mentioned this at ProSEO last year but I think the size of the twitter link graph means that it will continue to be important to investigate/discuss.
Rand then moves to how the social graph and particularly Facebook ‘like’ data might affect SERPS. At present, as we all know, it’s not taken into consideration in ranking models. But he speculates that this could happen, potentially from Bing who have an investment in Facebook.
On to some correlation data about the importance of putting a keyword at the front of your title tags. The stats in his correlation analysis clearly show that it has a big impact. I think most SEOs would agree that this is the case from experience. ”A while ago some dude said it doesnt matter, just put your brand at the beginning – I think his name was ‘Rand Fishkin’”
However I still think there are cases for it in terms of clickthrough rates for big brands.
Ahem. We come to H1 tags. Yep, they’re not as important as people think. That is all.
He then touches on the dreaded Keyword Density. The data shows that while putting keywords on your page helps rankings (duh), if you look at keyword density % as a fixed causation factor you are an idiot. (or words to that effect)
Rand then shows up some data related to “Latent Dirichlet Allocation” which is about analysing which phrases are related and relevant to others. Um… to me it seems a bit like “Pantene shampoo science” but that’s me being flippant and dismissive… one to investigate in future I think
[Update: Rand's slide deck now added! Note how I got you to read my notes first - hehehe...]
SEO Ranking Factors 2010 SMX London
Rob Kerry – Ayima
Rob starts by talking about the Death of the External 301. It seems in January Google clamped down on smart affiliates using cross-site 301s on affiliate links and stopped them passing value. He uses the example of Top10 Broadband which suffered but recovered. The key message is, if you’re moving sites, don’t rely on 301s to save all your link juice – build up the link profile to the new domain with fresh linkbuilding,
Rob then talks about how Google also seemed to improve their link filtering back in January – a lot of sites’ links were devalued where there was not as much of a natural link profile – overoptimisation of links and particularly over-focusing on homepage. You need to make sure the signal:noise ratio in your link profile is good, and make sure there is a good proportion of brand/natural links across the site. ”Deep pages need to justify their existence by having external links to them” – fully agree with this comment! So often people only linkbuild to their homepage and they dont realise how much they are losing out – deep links rule the SERPs, it’s a fact.
Rob then talks about the “MayDay update” – take this with a pinch of salt! People like to come up with new names for so-called updates – in his opinion this could just be an example of what he has talked about above, ie lack of deep links affecting long tail rankings. However he does say you can’t rely on domain authority alone any more. You need to have specific content and pages related to specific terms with links coming into them. This may be the issue for people.
Mikkel deMib Svensson – deMib.com
Great opening slide – “Get Rid of the Crap on Your Site” – LOL. There are two common types of crap on a site: ‘code junkyards’ from lazy or incompetent developers who leave opportunities for hackers; and malware. He talks about how ignoring cleaning /securing your code can completely slam your rankings.
So how do you secure your website? Demib goes through a few steps you can take but sadly this particular liveblogger completely missed them – I will try to update later! (sorry)
Site speed has always been a factor – the faster your server responds, the faster the engines crawl your site. Completely agree with this – check out my previous post about site speed and SEO…
Does the ‘quality’ of code (ie validation) matter for rankings? Demib says, No, emphatically. Google can’t ignore a site just because it doesn’t validate – they don’t even validate themselves.
He then moves to the importance of clean code – don’t give Google unnecessary crap to wade/crawl through before getting to your content:
- .net viewstates are a total pile of crap! Most people don’t realise how much this can harm you (I totally agree). Demib has printed out a typical viewstate from a site’s code and stuck the pages together – it’s about 10ft tall in total! But there are ways of keeping this to a minimum – see this post for example.
- Keep all JS / CSS external – we all know this but do we do it? good kick up the arse here
- Minifiy your code – it can make a big difference
- Get rid of HTML comments – they just bloat your page
- Remove those ridiculous, unneccessary meta tags!
(while this is going on there’s some discussion about .net viewstates and Mark Cook comes out with a lightning quick solution of how they hide viewstates from googlebot (a bit of ‘whitehat cloaking’) on the Further site – great stuff and I love how twitter responded to SMX in realtime there!).
Q&A Session
- Question about the ‘MayDay’ update as well as Google’s new interface. I missed Rand’s answer, but Rob Kerry makes the point that it’s not necessarily a negative – a change in the algorithm is bad for some but an opportunity for others. It comes back to the point that you cannot rely on domain authority alone any more – need niche content, and perhaps this helps the specialist niche sites. (Interesting – are we seeing a corollary to the brand update taking place?). Demib warns of the dangers of always chasing the latest update or tweak because when it changes again you could lose out. Stick to the fundamentals. Regarding Google’s new layout Demib reckons this might be a win for Adwords – essentially because people are unfamiliar with the new organic results layout they tend to click on sponsored results more. Really interesting theory and I wouldn’t rule it out.
- A question about UK vs US SERPS. Rob says that in google.com it’s homepages that tend to rank due to the sheer authority of US based sites, tends to be more deeper pages in the UK.
- Regarding QDF and realtime search. Demib makes the point that a lot of smaller language / country results have much smaller datasets and you don’t see this happening as much as you do in .com. Rob gives an example of the “credit cards” SERP in the UK where you get a news results onebox with US news stories!
- Question about 301s and canonicals for Rob Kerry. Are canonicals now the replacement for using 301s? Well the most important thing is to try and repoint old links rather than using 301s. However failing that canonical does seem to be the solution of choice at the moment. It’s no coincidence that Google launched cross-domain canonical tags in Dec 2009, and then devalued external 301s in January. For Rob this is an example of Google taking away one solution and replacing it with another, Google-approved one. He has noticed it working and in fact a lot of sites are getting link juice via canonicals that don’t really deserve it. Demib doesn’t really like canonicals – he’s seen Google treat correctly used canonicals wrongly. However Rand has seen canonicals
- Question regarding future trends…responses below
- Rand: its insane that we tweet so much (good) content, but don’t use that content for SEO value. We should be capturing this as webmasters.
- Rob: If you value your short tail traffic, target links to the root of your domain (and vary the anchors a lot / include lots of Brand links) and don’t change your domains. If you value long tail, focus on deeper pages and deep linking.
- Demib talks about how so many sites are missing a trick by having stale pages / content and never updating it. Make use of UGC / pull in (or even scrape!) content from other sites, use dynamic elements – whatever you can to get your pages refreshed.
- Rob Kerry: at a past conference in the “Give It Up” session I mentioned the idea of bringing in affiliate links to your own domain and then using rel=canonical to your money pages – ie pretend it’s like a session id. A few months later at SES a big site said they had done this and had excellent results without having done much link dev.
- Good Q about rel=canonical: How important is it that two pages are identical in content? How far can you go with variations in content? I believe the answer Rand gave was that it doesnt seem to matter too much – Google really strongly values this (pls correct me if I’m wrong on this)
The end of an excellent session, loads of useful info. I think Rob Kerry’s thoughts were the highlight (as I predicted), I definitely need to come back to them!
Thoughts, analysis and rants on SEO, search and internet marketing from Jaamit Durrani.


