Adwords, SEO & Google Lies

by admin on November 28, 2011

Welcome back Rankers. Want to get right into this week. Question is coming in from myko54 who says ‘Does Google have plans to transcribe audio component of video to make it searchable?’ God, I hope not. The reason I say that is because obviously that would make a lot of sense, right? It’s sort of the next logical step for Google to go but they’re already doing closed captions on video and they’re trying to transcribe these things on the fly. If you just have a look at mine they’re absolutely terrible, they turn out as gobbledeegook, I don’t know whether that’s because I’m not articulating properly or it’s the Australian accent or it’s a bit of both. Makes sense for them to do it but they are going to have to become Apple like in their ability to understand different dialects and accents and all those sorts of things.

I want to talk a little bit to you today about Adwords. Many of you know that we do Adwords campaigns for clients but we only do them when they’re doing SEO so the idea is that we use the Adwords and the SEO together to have a more rounded campaign for a client. There’s a few things that you need to understand with Adwords, I’m not going to get into the mechanics of it today but just how it works in conjunction with an SEO campaign. It’s great because you can use Adwords to target specific phrases right now and switch it on and see what happens or just this weekend or a specific timeframe or a specific geographical region. So it allows you to target a lot better. One of the things that you are going to find when you get into keyword analysis is that Google tools will all tell you something different about what’s popular and what’s not.

I want to take you back a couple of weeks to our Gasp Clothing Gasp Melbourne experiment or blog posts and the shows we did back then and the sort of traffic we were getting from those phrases. Some of you may remember it was a major social media faux pas by an Australian fashion retailer. Everyone said, ‘Oh it’s terrible that’s the end of that company, they’ve committed social media suicide’. Anyway, it hasn’t hurt them and the hoohah from that social media episode has gone away but they got lots of great publicity out of the whole experience. We have a couple of posts on our site so we rank for some of those gasp keywords. We were looking at the stats a couple of weeks ago and looked at Google Insights which was telling us at the time that Gasp Clothing was more popular that Gasp fail or Gasp Melbourne. Now, you can see here that there is a spike around that time that I’m talking about specifically for Gasp Clothing. Gasp Fail in relationship to Gasp Clothing doesn’t even rate on the board anywhere, it’s insignificant in comparison. Gasp Melbourne shows as 38% of what the Gasp Clothing searches are. This is a world wide search for the last 90 days, if we go the last 30 it’s pretty similar except we don’t have any Gasp Melbourne data. Then if we have a look at Webmaster Tools and this is the search queries, your site on the web search queries that I’m looking at and because we were ranking for these phrases I can get data on them within Google Webmaster Tools and that will tell me the search impressions. What we’re seeing here is Gasp Clothing is 1000 impressions and Gasp Melbourne is 700 and this is only the last 30 days so it’s saying something completely different from Google Insights for Search. That’s why I would recommend that if you are going to do Adwords make sure you’re using Webmaster Tools and make sure you’ve got it set up properly. If you are already ranking for some phrases, you don’t have be ranking high, you don’t have to be ranking on the front page to get this sort of data. You can be ranking say 53 it says here for Gasp and it’s telling us there’s been 500 impressions. What you can do with Adwords is quickly test those numbers to see whether they’re accurate or not. Interestingly, we’re seeing here Gasp Melbourne 700 and Gasp Clothing 1000 in the last 30 days. If we go to the Google Keywords Adwords Tool and we put those same three phrases in there we see we get very very different numbers.
We’re getting told nearly 10,000 people a month search for Gasp Clothing, you gotta get into that, which of course is not true. Remember that the Adwords team, the Webmaster Tools team, they’re different teams within Google. Of course, they talk to one another but they’re totally different departments so they’re going to have different data. These numbers here are made up of an average of what Google Adwords thinks the total global 12 months search would be. So, it’s a monthly average. So, it’s obviously wrong. What I would do is use this Tool in conjunction with Google webmaster Tools if you’re already ranking for some phrases and you can start to see some impressions. The one thing here that is really outrageously inconsistent is the relationship between Gasp Clothing and Gasp Melbourne according to Google Webmaster Tools. Not necessarily so when you look at Google Insight for search but one thing that is consistent is that Gasp Clothing is definitely the most popular phrase out of those three so we’re seeing that across the board. So that’s what you take with you and you can then go and test.

The mistake that a lot of businesses make that we see when we take over their Adwords campaign is essentially they’re paying too much for a phrase. Here’s an easy example. On the weekend my wife was googling for a Dremel which is basically a small hand held drill that’s got lots of different attachments, no, it wasn’t for me, it was for her. We googled Dremel and we can see that Bunnings is finally doing Adwords which is great to see, Bunnings are probably Australia’s largest hardware retailer and if we click on their ad, hopefully it’s not costing them terribly much, we don’t actually have the word Dremel in that ad. The same sorts of things that Google applies to Organic they want to see similar things in Adwords and it’s about relevance. They will reward relevance in your Adwords. If your ad is really relevant to the search it is going to give you a higher quality score and possibly cost you less and also increase your click through rate. When someone does a search they want to see what they’ve searched for. I happened to click on that ad because it was Bunnings and maybe the brand, it’s got a good click through rate because of the brand. But I would say they are probably paying more that necessary because there’s a couple of things that’s happening here. When you click on that ad the word Dremel didn’t appear in the ad itself. When we look at the source of the landing page that we were taken to and it’s great that we’ve got a landing page, you see so many ads where they just drop you at the front page of the site or the home page. That’s really really bad because it’s totally irrelevant, some one’s going to click on that ad, they’re going to hit your page which is your home page which probably won’t have a lot to do with their search and then they are just going to hit the back button. Basically it’s a waste of money for you and Google’s going to charge you more if you do that sort of thing. So, you’ve got to set up pages that are very specific to the search. You can see here the page title is not specific to the search ity doesn’t have dremel actually in the page title. But when we do a search for that word actually in the copy on the page it’s right down here in “Narrow by Brand” so this page really isn’t relevant for my search. It might be relevant for that Ad but that Ad is not relevant for my search. When I search for dremel on the Bunnings Warehouse site there’s a whole range of dremel tools that are just not seen. Not that you can buy off this site anyway, it’s basically a catalogue site but I would assume that in some point in the future they are going to turn it into an ecommerce site. So that’s an example of an ad campaign which could be getting much better results than it probably currently is.

One of the other problems that Bunnings has with this particular ad campaign and of course the other morning while we were sitting in bed googling for a dremel I didn’t have my desktop with me because that would indicate I’d be at the desk not in bed, so of course I was using an Ipad, an IOS device, portable device, mobile device whatever you want to call it. I did the search for dremel got the same page that we see in Google desktop, then unfortunately for Bunnings if we click on their ad we’re taken to a page which is obviously set up when you look at the URL, it’s re-directing me to a mobile friendly page so I can search my store. But it’s got no relevance to what I was looking for so of course, I didn’t look at Bunnings the other morning, I just looked at that page and went “wow what a waste” and looked somewhere else. I actually looked whilst I was mobile at the Google shopping results, which of course Bunnings are not in because they are not transacting. It’s those sorts of things that can really hurt your quality score in an Adwords campaign and there could just be 5 or 6 of them and your going to be spending too much money and your not going to be getting the best bang for your buck that you possibly could. When we take over an Adwords campaign they’re typically the things that we have to fix and make things more targeted and those sorts of things.

How you use that in conjunction with SEO is that you make sure that if a phrase is costing you a bundle, you better make sure you’re ranking organically for that phrase as well. Whilst a lot more people are clicking on ads now than what they used to people still trust the organic results more than they do the ads. Just ask some of your customers Do you click on ads? Or do you click on organic results? See what they come back with.

They’re some of the basics of how you should be using Adwords and Organic search together. Hopefully that’s helpful and we’ll see you next week. Thanks very much. Bye.

convert this post to pdf.

{ 0 comments }

Social Signals in search

by admin on November 18, 2011

Welcome back Rankers. Last week, last Thursday in fact, was a big day here in the office. We had our highest traffic day for the website in a single day, ever. It was because of IOS5, what’s IOS5? It’s the new operating system for things like your Iphone, your Ipod touch, your Apple TV, your Ipad. The new operating was released last week, lots of excitement, I was excited, I came in here early got my Itunes, tried to download it and it said it’ll be finished in about 7 years and it was saying the same thing to everybody around the planet basically. It wasn’t able to be downloaded, it was just getting hammered. The Apple server was getting hammered because everybody wanted this piece of software. One of our writers in here, said there’s a better way to get it, you just go and do this and you pull this switch over here and you put that URL in there and I said Yeah OK. I wasn’t that interested because it sounded like too much hard work to me. Then later on in the day one of the other guys was trying it and it worked for him as well, I think it was Byron and he said “You know what, this would make a great blog post”. And all of a sudden we all just went “Oh yeah of course it would why didn’t we think of that earlier?” So Michael, the writer whose idea this was wrote the blog post. This is the page that got us about 650 views in about 36 hours, something like that. You can see that it’s only had 32 tweets, it’s only had 21 likes and it’s only had 2 +1’s if you know what I mean. I know what you’re thinking, e3 right but that’s it that’s all the backlinks it’s had that we initiated with. We didn’t put our backlinks on networks, we didn’t jump into forums and put backlinks in forums. Within 12 hours it was ranking #1 for download iOS5 or iOS5 download timeout which was the main problem. People were having problems with timeouts trying to get it. So the traffic actually came from, we got 155 hits just from Google itself being real time. We were ranking on the Saturday, so about 30 hours after the post went live we got about up as high as 5 for download iOS5 with no other backlinks except a little bit of social media here. We’re nowhere to be found for that phrase, we got to #1 for the timeout phrase but the download iOS5 phrase only got as high as 5 and then it disappeared. The only linking that we did was here. This is what you would typically expect with social media signals for Google. They’re time sensitive, they have a short life span unless people keep talking about it. If people keep talking about it and keep sharing it then Google will keep taking notice but it won’t if it’s transient in nature like this event was. People could download iOS5 after the initial rush and they didn’t have a problem with it. Incidentally, Google Webmaster Tools was showing that we had in this period 110 different queries and that’s not searches, that’s queries. It’s saying here 2,500 different searches for that particular phrase that we would have been ranking for. Admittedly some of them would have been on page 2 so we wouldn’t be getting traffic from them because no one goes to page 2. This is the spike of the traffic and what happened. This referral source here is Twitter being #1, Google being #2 which really surprised me because it was such a short timeframe and we didn’t have any backlinks so why would we rank for those things or those phrases. They were hard phrases to rank for but because we got the amount of referrals from Twitter I believe that sent a social signal to Google to say this is popular right now so you need to show it. There were only 32 tweets so all this traffic would have come from the followers of those people that tweeted and probably about 10 of those were from us. Just interesting to see what effect social signals have and that you actually can rank for something just through social signals. But if we wanted to keep that page on the front page of Google for download iOS5 we’d probably have to do more work with more pages inside the index inside our site linking back to that particular page referencing it.

The other interesting thing that happened this week which kind of made me think about that experience and what you’re up against with Google. This is an excellent interview by a guy called Eric Enge (I don’t know if that’s how you spell his name) if you go to stonetemple.com you’ll find this interview. He’s interviewing Google’s chief scientist, essentially, and basically what he’s saying is reiterating 2 things for me anyway, there’s a lot more in the article than just the 2 things. It’s a really good read, it talks about Google translate and everything. He emphasizes that Google is making up to 2 changes per day to the algorithm. This isn’t just tinkering with the edges of the website this is like to the algorithm itself and also he reiterated that speed is so important now. You’ve got to have fast pages, you’ve got to have fast delivery. In this article he brings out Google’s even going to the extend of trying to guess what it is that you are searching for and the most relevant page for that search and is trying to cache in the background for you. If you’ve got a slow website they’re going to ignore you, you’re not going to be included in that because it’s going to hold up the user experience too much. We’ve seen it time and time again, slow sites equals bad rankings.

The other interesting thing that came out this week. If you use Google Webmaster Tools regularly you may be a bit flummoxed. It’s always concerned, well not concerned me but thinking “Why are you telling me that Google? That doesn’t make a lot of sense”. It’s under this diagnostics area, under crawl errors. It says here quite clearly these are Crawl Errors and then it says here Restricted by robots.txt, it’s like well “Why are you telling me that’s an error?” I’ve gone through and restricted all these parameters – replytocom – if you’ve got a Wordpress site you’ll know why I’ve done that. But why are you telling me this is presumably an error. There was an excellent discussion that Matt Cutts was involved with last week and it was about how sites can get pushed out of the search results if they have a lot of auto generated pages from search and those sorts of things which Google doesn’t deem to be that useful to the user. One of the things that came out of the discussion is that this guy here says “I build webmaster tools so this confusion is entirely my fault”. The fault that he’s talking about is that what’s restricted by robots.txt is coming up in crawl errors. The reason that’s in there is it’s a report to show what’s being blocked and to correct it in case some of those things in robot.txt you didn’t expect to be blocked. For all of those who have asked me over the years “Why is Google saying this is restricted by robots.txt is an error?” It’s just purely putting the information in here that it’s found in robots.txt and it won’t crawl just so you know, in case there’s anything in there that shouldn’t be and you can change it.

Had a heap more to talk about this week but we’re just not going to have time. Thanks very much. See you next week. Bye.

Oh!! One more thing, Michael would kill me if I didn’t talk about this. In the site we now have a new category called Ask Jim. This is because we get a lot of queries via Twitter, obviously Facebook all of those things. If you have a question about SEO just ask on Twitter on jimboot or at stewartmedia or on the facebook page which is facebook.com/melbourneseo we’ll put answer up here or you can come and comment on any of these.

That really is it. See you next week. Bye.

convert this post to pdf.

{ 0 comments }

Webmaster tools data in analytics

November 2, 2011

Gasp Fail Real Time

October 12, 2011

Online Reputation Management

October 3, 2011

Google+ search

September 28, 2011

SEO Reality Check

September 19, 2011

Google & Backlinks

September 9, 2011

Google SEO Guide

September 2, 2011

Authorship Markup

August 28, 2011