Bummer! This is just a preview. You need to be signed in with a Pro account to view the entire video.
If MacGyver Did SEO28:59 with Martin Macdonald
Martin looks at how you can push Excel to do much more than you ever thought possible. In this case, build an index of the web to give you early warnings about index updates and get the full picture of your vertical.
Thank you very much, and good morning to everyone. 0:01 I am actually quite surprised to see so many people here 0:04 because I assumed that about two-thirds of you would still be hungover, 0:06 in the hotels, upstairs, whatever. 0:09 So congratulations everyone that made it, and the people that filter in throughout 0:12 the presentation if everyone could just turn around, point, and laugh at them. 0:16 That would be great. 0:18 So first of all a quick introduction, and actually before I go any further 0:21 this is huge. Last here—who was here last year? A quick raise of hands. 0:23 Right. You remember this stage last year? It was kind of like there to there. 0:28 Everyone was about to fall off. This is massive. 0:32 I could run up and down here. Anyway. 0:35 Quick introductions. He says. 0:37 There we go. That takes a while. 0:41 My name is Martin Mac Donald. I work for EAN which is the Expedia Affiliate Network. 0:43 I don't know if anyone's heard of it. I'll go into exactly what it is and what we do in a second. 0:48 If anyone ever wants to get in touch with me the four things that you see there: 0:53 Twitter, LinkedIn, Facebook, and Google+. 0:57 That's also kind the order of the amount of time I spend on each one of them. 0:59 So kind of try the top one first and work your way down. 1:03 If you make it to Google+ and you haven't had a response 1:06 well I've probably died or something. 1:08 The Facebook one's quite interesting actually because I'm Facebook.com/mogmartin, 1:10 and I used to use that name about up until about three to five years ago. 1:15 It was kind of tied into my past of working in gambling and poker SEO, 1:22 and there's a lot of bad shit around the Internet which is tied to that mogmartin, 1:30 so I've kind of stopped using it most places now. 1:33 But if you ever want to see some of the stuff that I used to do historically, 1:36 search around that keyword. 1:40 Incidentally I used to have Twitter.com/mogmartin, 1:43 but they banned me despite the fact that I was a beta tester for them. 1:46 Seriously guys? I mean that just wasn't fair. 1:49 So for the people that remember my presentation last year, 1:53 I've got one warning before we start. 1:55 This is very much going to be a black hat-free zone. 1:58 I apologize. I genuinely do. 2:00 I know you guys love it when we come up here and we talk that kind of shit— 2:03 the stuff that you could do and not tell anyone about. 2:07 That isn't what I'm going to be talking about today. 2:10 I've kind of got a different take on most of the speeches that I've done. 2:12 Normally I go through lots of actionable hints and tips and things that you can do 2:18 about your site, ways of link building, ways of content production, things like that. 2:21 Today what I'm going to do is slightly different. 2:25 I'm going to talk about a project that I've been working on for the last couple of months, 2:28 something that we're doing internally at Expedia—EAN, 2:31 and something that I think probably could have a lot of value to a lot of people in here, 2:34 whether or not they work for an agency or they're in-house. 2:40 There's a ton of stuff that this tool can be used for. 2:43 At the end of the presentation I will be discussing how we can work together 2:46 and make it all better. 2:49 So first of all for those of us who don't know who MacGyver is. 2:52 Does anybody not know who MacGyver is? 2:57 One. Yes! Excellent! Had I done this—I was fully prepared for about 80% of the people 3:01 to raise their hands because I was kind of assuming that anyone under the age 3:08 of about 35 probably didn't see him first time around, 3:12 so to those younger people in the audience he is that guy that 3:15 Patty and Selma have got the crush on. 3:18 The point in this is he was famous for getting out of scrapes and situations by 3:22 building lots of interesting stuff out of not very much. 3:29 Now that's kind of what I've done in this little project is built something that 3:33 really should never have been done in Excel because that was a really stupid thing to do, 3:36 but I got so far down the path I just thought oh, screw it. 3:42 I'm going to have to finish anyway because I put about two or three months into this now. 3:45 I kind of MacGyver-ed together a way of indexing Google 3:48 using two things: Advanced web ranking from Caphyon and Excel. 3:52 It leads me to this question. Everyone knows what the mother of invention is right? 4:00 Anybody. Necessity. That. No. 4:04 For most people perhaps. In my case total ineptitude. 4:11 I put this together in Excel basically because at the time 4:15 I thought this would be the quickest and easiest way of doing it. 4:21 I could have done it in PHP. I could have used different script. 4:23 But basically I was doing it as a proof of concept so I used Excel 4:26 because it was quick and easy, and I'm pretty much inept at most programming languages 4:29 these days because I don't really do it enough anymore. 4:32 The reason I put it together is because when I first started at Expedia 4:37 I was given a challenge. 4:41 I should probably explain exactly what it is that I do there. 4:43 I've got someone else from Expedia somewhere in the audience. 4:46 Jeff Slipko. Where are you? 4:49 Hey. There he is. So Jeff and I work together. 4:52 What we do is we help our partners who are people like airlines or online travel agents, 4:54 people like kayak.com, to improve their marketing. 5:01 So we've kind of got like a mini advertising agency within Expedia, 5:04 and we go at them. We help peripherally. 5:09 It's to improve their business, to do better, to drive more sales ultimately to Expedia. 5:11 You know? There is a little bit of selfishness there at the end of the day. 5:15 But everything we do for them we do for free. 5:19 Either Jeff or myself goes into these airlines or online travel agents, 5:21 looks at their strategies, works with them, tries to improve things, 5:24 but the challenge that we had was putting a dollar value 5:28 on the amount of help that we've given people, 5:32 and it's really difficult. I mean hey. 5:34 Even if you've got full access to Omniture or Google Analytics, whatever, 5:36 still putting a dollar value down to the amount of help that you do in SEO 5:40 is still difficult. 5:44 Imagine how hard it is when they won't give you that data in the first place 5:46 yet we still have to define why what we do is important. 5:49 So we had a challenge of trying to be able to demonstrate 5:53 the uplift that we have given partners without any data from them whatsoever 5:57 which was a bit difficult. 6:03 There are indexes of search results around on the Internet. 6:05 Searchmetrics is a really nice one. SEMrush, not so much. 6:09 But Searchmetrics is really nice. 6:13 I've always been critical of SEMrush, and I don't know why. 6:16 I think it's just the color schemes. I'm really—sometimes I see things, 6:18 and I just take an instant dislike to them. You know. 6:21 It's probably very good, but I haven't used it in a while. 6:23 The point is other people do this. Right? 6:26 Searchmetrics was the one that I looked at initially 6:28 to see if I could use that as our primary data set, 6:32 and they've got 100 million keywords, and it's really good for looking at 6:35 big sites and seeing how they've dropped. 6:39 It's good for doing initial market research if you're unfamiliar with the market. 6:42 The problem is though 100 million key words sounds a hell of a lot, 6:48 and trust me when you're trying to scrape that much stuff from Google 6:52 it really is a lot. 6:54 But the problem is for me in travel, I'm looking at affiliates who 6:56 I know are generating large amounts of traffic, large amounts of revenue, 7:00 and they don't even figure on Searchmetrics. They're just not there. 7:04 That 100 million keywords is to cover the whole Internet, and it's just not enough. 7:08 If it was 50 billion keywords we might be getting there. 7:12 So what I needed to do was come up with something 7:16 that would work much better for our industry. 7:20 Now you might be thinking well hang on a second 7:22 because there's these indexes as well. 7:25 Now just in case anybody's going to get the two things confused, 7:27 those are not the same thing. 7:30 Searchmetrics, SEMrush, and the one that I'm going to present to you in a second 7:33 are indexes of Google results and how quickly they change. 7:36 They are not indexes of links from website to another. 7:41 Big important distinction there. 7:44 We use APIs from most of those things in this project 7:46 to be able to define why stuff is ranking where it is, 7:49 but they're not the same thing. 7:52 So we came up with a little project called SERPwatch. 7:54 Basically what it does is it analyzes any updates. 7:59 So we have Panda 3.9. 8:03 This is the same kind of stuff that Dr. Pete was talking about yesterday. 8:05 But kind of the difference is I'm going into much, much, much more depth here, 8:07 and you can too for your industry so that you can work at all sorts of 8:12 competitive intelligence, biz dev, all sorts of stuff. 8:15 Having this kind of index for your industry allows you to see 8:19 the strengths that your site or any other competitor's site has got 8:24 and focus in on why they're doing better than you or worse, whatever. 8:27 It efficiently allows us to look at any partner or anybody in the travel industry, 8:31 and that could apply to any of your industries as well. 8:37 It helps us develop further clients. 8:39 This does all sorts of stuff. 8:42 It's a great tool because the world fundamentally revolves around now 8:44 first going to Google, doing a search, doing—then moving on to the thing 8:49 that you want to do. People use it as their homepage. 8:53 Even if you know that you're going to SEOmoz, I bet most people 8:56 just type SEOmoz into the browser bar and then click on the result in Google. Right? 8:58 Most people do that. So that's a huge, huge information set, and we're not using it 9:01 for business intelligence at this point in time or certainly not enough, 9:07 so that was one of the things. 9:10 The last item on that list is it allows us to own the data set for the industry, 9:12 and we can then tell other people things about their sites that they weren't expecting. 9:16 That's quite a powerful thing. 9:21 So my original thought was that we could use it for infographics. 9:25 I hate infographics, but I've got to sell this internally 9:30 so I'm handing people like that buzzwords. 9:33 I put this together with the thinking that if I can tell you how much someone's site 9:35 has gone up since they implemented something or how powerful certain sites are 9:41 within the industry for certain destinations of certain countries. 9:45 There's a million different data points that we can use for infographics, 9:48 and all of them are covered in this. 9:51 Finding out information about your competitors. 9:55 This is an awesome tool for that, but on top of that 9:58 if you work in an industry where, for instance, from our perspective, 10:01 we've got—using this tool—we've got lists of every major travel site 10:04 in every country of the world. 10:09 If these people aren't currently partners of ours, we can use that as a biz dev list. 10:11 You can use this to look at competitors. 10:15 You can use this to look at people within your industry 10:18 that are selling the same products as you and look at how they're doing keyword wise. 10:20 There's tons and tons of business information that we're ignoring 10:24 because scraping Google to this kind of volume is something 10:28 that historically has been quite hard to do. 10:31 The most important thing, I think, for me is it allows us to position ourselves 10:34 pretty much as the authority within the travel market. 10:38 I can see someone from Skyscanner down here who is a great site in the travel market 10:41 looking and smiling. 10:44 Because I can use this to tell you what your rankings have been historically 10:47 for most of the keywords in the industry. 10:50 I bet you anything I could tell you what your rankings are 10:52 for more keywords than you could. 10:55 We'll do that afterwards. 10:57 So how does it work? 11:01 In essence it's really, really, really simple 11:03 because I just MacGyver-ed the whole thing together. 11:06 All it is is I'm using Advanced Web Ranking. 11:09 How many people in here use that as their primary rank-checking tool? 11:13 Not many. You should. It's brilliant. 11:18 I've been using it forever. It is my favorite of all the rank-checking tools. 11:22 Fundamentally if you want to use this spreadsheet, which I'll release at the end 11:24 of the presentation, you need that. 11:29 I'm not on an affiliate commission, but the owner of the company 11:33 has sat over there somewhere so he is buying me beers tonight. 11:35 What it does is it just gives us these raw, unformatted lists here. 11:41 So what I've got is the keyword, the search engine, the position it was in, 11:45 and then the landing page. 11:49 Then what we're doing is we've got lots and lots of different VBA macros. 11:52 See the beauty of this was actually no coding was done in any of this project. 11:56 All of this was just recording a macro of me doing stuff 11:59 and then reapplying it over and over again. 12:02 Not one single line of code was written in building this whole index. 12:05 So what this does is it gives us—and this is how simple it is. 12:11 It gives us the keyword, the position, the landing URL. 12:14 All we want to do initially is just analyze what the differences are day by day 12:17 for each one of these. 12:21 We can catenate. 12:23 Richard was talking about this a lot yesterday. 12:25 We can catenate the keyword along with the landing page 12:27 and then do a VLOOKUP every day based on the previous day's data, 12:30 and then take all of the differences in that and drop it into a separate table 12:34 and keep it. 12:36 Excel is nowhere near big enough to start doing 50,000— 12:38 I'm scraping a large amount of data with this. 12:42 I'm actually scraping, for one of my data sets, 25,000 keywords 12:44 times the top 50 results so that's 425,000 Google result pages every 24 hours. 12:47 That's one of them I'm running. I'm running about six at the moment. 12:52 This can scale up very well, 12:56 but Excel is the wrong thing to do this in 12:58 if you're going to start doing lots and lots of queries on it. 13:01 So what I've done is I've worked it in such a way that instead of recording 13:04 all of the exact data, we're recording changes day by day, 13:07 and it's a much smaller table. 13:10 So what we end up with is the position it was in this day, the next day, the following day, 13:12 but we also know, and I'll explain how we're going to do this in a second, 13:17 we all know what the domain name was and what the landing page was. 13:20 Those two bits of information are pretty crucial to the rest of the analysis 13:23 that we're going to do. 13:27 This is kind of laying out how it's building the dashboards. 13:29 We take all of the rankings, an Advanced Web Ranking. 13:35 We're doing exactly what I was just saying there. 13:38 We're catenating the landing page and the keyword 13:40 so that we can then look up the difference between that and historical points of time. 13:42 We're then summing those changes and using the Optify click-through-rate data 13:46 which is one of my favorite ones. 13:52 Does everyone know what I'm talking about by the Optify study that came out 13:54 about a year ago maybe? 13:57 Nobody? Anybody? Fine. 14:01 Someone Tweet a link to it because it's excellent 14:03 because that's the model that this uses as well. 14:05 Then what it does is it takes the perceived traffic from Google ad words 14:08 by keyword, uses the Optify click-through-rate data 14:14 to give us an idea how much traffic that site is generating 14:18 from that individual keyword and then sums up every single site's traffic 14:21 in the data set and gives you a reasonably clear picture as to who's doing best, 14:26 who's doing worst, and everybody in between 14:31 and the different changes that they've got day by day. 14:34 The other stuff that it does if I go back a second. 14:39 The other stuff that we're doing is I've got lots of little add-ons for it 14:41 that allow you to finalize individual sites from that data set. 14:45 So when you're looking at say the top 50 sites in your niche, 14:49 you can then click on any one of those sites, see all the keyword data for it, 14:52 see all of the rankings for it historically, do the rank checking which I'll speak to 14:56 yourself about afterwards—reference to Skyscanner— 15:01 and what it also allows us to do is look at the reasons why that site is ranking 15:06 either across the whole domain or across individual keywords. 15:11 It's ugly by the way. 15:17 I'm really not very good at graphic design. 15:19 In fact I am shit at anything to do with graphic design. 15:22 I'm famous for producing ugly websites basically. 15:25 I'm also famous for producing ugly Excel sheets. 15:28 This is about as sexy as this stuff gets, but it's just kind of a very quick 15:31 demonstration of the kind of stuff that you can output in seconds. 15:34 Before we start looking at the kind of learnings we can get, 15:41 these are all of the different kinds of data that we're keeping at this point. 15:45 We have every domain that has appeared in any of these searches. 15:50 We've got every one of the landing URLs, all of the keywords, 15:55 and the perceived volumes in adwords. 15:57 We know the changes day by day so we can see who's doing well, 15:59 who's improving, who's losing space. 16:03 We're also taking all of the SEO metrics for each one of the landing pages 16:06 and each one of the domains. 16:09 There are a couple of different APIs straight in Excel as well, 16:11 and we're also taking all of the social metrics for both the landing page URLs 16:14 and for the domains themselves. 16:18 So it's kind of getting more and more complicated. 16:21 This started out as something that I said, yeah. I could put this together in half a day. 16:23 It's taken about two-and-a-half months. 16:27 So it then gives us this really ugly table here. 16:32 But all of this again is automatic. 16:36 Now in the same way that Richard was showing us yesterday morning 16:38 the functions that he's built, the add-ons that he's got for Excel that allow you to do this, 16:41 this spreadsheet essentially has all of them in it, 16:45 and it is using them to pull this data in. 16:48 The first one—now this is a great one because I'm sure everyone 16:50 has had this problem at some point or another. 16:56 I've got a great big huge list of URLs. 16:58 I want to extract just the domain names out of those. 17:00 It's an absolute freaking nightmare. It genuinely is. 17:02 The way I started doing it was I built a huge array of every one of the top-level domains, 17:06 and I then used that to pass and match it against the URL string 17:10 and take the stuff that was immediately before it, 17:12 but that gets buggered up by .co.uk and trying to pass and match on full stops 17:14 and things like that. 17:19 It also loses when you've got a large amount of subdomains, 17:21 so WordPress.com for instance. It doesn't really work for that. 17:23 There's an agency in the UK called iCrossing. 17:27 I don't really know anything about the agency. 17:32 They're quite a bit one, but I've never worked with them. 17:34 But they have got a really nice add-in which is available at that address, 17:36 and it's got a number of different functions within it. A couple of them are awesome. 17:41 It allows you to identify what the top-level domain is for any domain from a list, 17:44 allows you to pull the actual domain, the subdomain. 17:48 It's a really good quick and easy way of doing it. 17:51 That's the URL that has that. 17:54 Obviously this will be available for download afterwards. 17:56 I am using the same tools that Richard was talking about yesterday, 17:59 Niels Bosma's SEO tools, for the middle of it, and after Richard's presentation 18:02 I'm actually changing it to be his plugin for the rest of the stuff that comes from Moz. 18:06 So that's kind of a split of the data that we've got for every single 18:12 one of the URLs that are in this table. 18:17 This is the dashboard that we're building out of it. 18:19 So we can see on a day by day basis what the velocity of updates is. 18:23 Now this isn't real data because I didn't want to display it. Sorry. 18:28 This is normalized data with random websites. 18:32 What I've done is I put one big change in there, 18:38 but the way that this actually works is you see— 18:41 you would have seen the same on Dr. Pete's presentation yesterday. 18:44 What you see is a little line like that, and it kind of looks like something 18:46 you would have seen in ER with little spikes as stuff is changing 18:49 and going backwards and forwards. 18:52 But then every now and then you do see a great big huge spike, 18:54 and what we can do day by day is go in an analyze who won and who lost 18:57 on the previous day's algo updates. 19:01 Then the manual bit comes in where you really need to sit and look at those sites 19:03 and try and work out why. 19:07 But the difference is for the last however many years 19:10 whenever there's been an update like a Panda or Penguin or a Mayday or whatever, 19:15 the first thing that happens is everyone goes into a blind panic. 19:18 Everyone does. It's international sign of—it's terrible. 19:21 People start complaining oh, I've lost all of my traffic. 19:26 Why has this happened? 19:30 You then spend the next two or three weeks looking for other examples 19:32 of sites that have suffered from the same thing. 19:35 You then look for the commonality between them. 19:37 You start analyzing why. 19:39 By that point another update has happened. 19:41 You're fighting an uphill battle with that one. 19:43 If you're tracking this to a large amount day by day, that's the crucial bit, every day, 19:46 you can find out within 24 hours every site that's been penalized, 19:52 every site that's gone up, and you can be the first to react, to change 19:56 which can offer a huge, huge advantage. 20:01 Now if you're in-house, there is no reason whatsoever why you shouldn't be looking 20:04 at say 25,000 keywords for your industry and then analyzing it that way. 20:09 If you're agency-side this stuff is so powerful 20:13 because you can use it to pitch with as well. 20:17 Hey, did you know that on March the 9th this year, 20:19 you lost 25% of your rankings, but your competitor XYZ increased it. 20:22 This is why. This is how we're going to help you to do it. 20:26 This can really, really, really help. [applause] Thank you. 20:29 I'm not even agency-side now so I shouldn't really care, 20:35 but I remember when I was. This would have been a really, really easy selling tool. 20:39 You know? Being able to show people stuff like this and all of their rankings historically 20:44 across the entire industry, that's really powerful stuff. 20:49 Most of the partners that I go into and chat with don't know this data anyway. 20:54 They've got no idea. 20:59 The fact that we can tell this, and you can tell this about your industry 21:01 and your clients, this is so powerful. Why the hell are we not doing this? 21:04 I thought it was an awesome idea. 21:10 Dr. Pete obviously thought it was an awesome idea as well 21:12 because we both came up with the same bloody thing 21:14 and are presenting it at the same conference. 21:16 [laughter] But never mind. 21:18 We found out about 3 weeks ago when everyone was having— 21:20 we got like an email group of speakers, and when we saw each other's titles 21:23 it was like oh, shit. Here we go. Who's up first? He is. Damn it. 21:27 [laughter] So this stuff is really powerful. 21:32 Now this is where it starts getting ugly because as I say 21:38 I'm really not very good at designing stuff. 21:42 The open source version of Open Site Explorer— 21:44 was it the term Richard was talking about yesterday? 21:48 I thought that was a really nice looking dashboard. 21:51 The reason I asked you to email it me is because I was going to steal it 21:54 and copy yours entirely and just make mine look just like it. 21:56 You forgot to email it to me. Yeah, and I got drunk. So. 22:00 [laughter] It wouldn't have happened anyway. 22:05 But this is the kind of stuff. 22:07 As a report the query here was trip adviser pages 22:10 that are ranking for London-related content. Bang. Two seconds. 22:14 I've got all of them. 22:17 I can then use that to look at the different parts of the site 22:19 that I need to build content on. 22:22 For instance I know that Expedia.co.uk do really, really well at, I don't know, 22:24 hotels in Liverpool so I can use their domain and those keywords 22:29 to work out which pages they're getting all the traffic to, 22:35 have a look at the content and analyze it, and then do the whole repeat plus one issue 22:38 of just being slightly better than everyone else 22:42 which isn't something that happens very often to me by the way. 22:46 Now the other thing that I'm using in this is a tool called SocialCrawlytics.com. 22:51 Has anybody in the room used it? 22:56 One, two. You want to? Can you get access to it? 23:01 I will give you access to it. 23:05 It's a really cool tool. 23:07 It's built by someone in the U.K. who's a great SEO. 23:10 Really all it does is you give it a URL, it goes in, it spiders very, very quickly 23:13 every URL on that site, and then it checks for every social metric 23:17 for every one of the URLs. 23:19 Now it's got a nice front-end at this point in time, 23:22 but I was speaking to Yousef who is the person who built the tool last night, 23:24 and I've convinced him to put an API in so that we can use this 23:28 for this spreadsheet as well. 23:30 If anyone wants access to it I'll explain how in a second. 23:35 This is the kind of results it gives. 23:37 Now what's really interesting here is you've got every one— 23:39 I think this was Expedia.co.uk. It might have been Hotels.com. 23:42 No, it was Expedia. So you can see here all of the Facebook likes, the Tweets, 23:48 the digs, all of it. Right? 23:52 This took two or three minutes to generate for a site of that size. 23:54 It's massively quick. 23:57 What's bothered me for a long time is we've always looked at link building 23:59 and thought about well, hey let's just stop putting links into the home page 24:02 and category pages and build links to product pages and stuff that we really want to rank. 24:06 That kind of thinking hasn't really permeated down—certainly the black hat world anyway— 24:10 hasn't really permeated down to social yet. 24:15 This points that stuff out instantly. 24:18 So the delta there, the blue line on this graph, 24:21 is how social shares and mentions look like on large corporate sites on average, 24:25 large e-commerce sites. 24:29 They're really tall at the top, and it trails off really quickly and then disappears. 24:31 Spammy sites look like the red one because they bout 5,000 Tweets 24:35 or likes or something. They pointed at their homepage, or they pointed at their 24:38 category page, and that's it. It collapses. 24:40 It's a cool tool. 24:44 If anyone in this room wants access to it and 10,000 free credits to run hundreds of reports, 24:46 that's what they need to do. You need to Tweet Hi @Ysekand. 24:51 You need to include #MozCon. I'd appreciated it if you included me as well. 24:55 I Tweeted that earlier so if anyone's too lazy and can't write that down, 24:58 I mean that's not really much price to pay is it? Tweeting that. 25:03 I Tweeted that earlier, so just reTweet me. 25:06 Yousef will get in touch with you, and he will tell you how to log in, 25:08 and then you can start using this stuff yourself. 25:11 I've got four minutes left before I'm going to get kicked off. 25:14 I need to move onto kind of the pitch. 25:17 There's the recap as to everything we're doing at the moment 25:22 with this really simple Excel sheet. 25:25 But what I want, and I've really got an ulterior motive for being here and saying this, 25:30 what I want to do is release all of this into the wild, make it open. 25:34 I'll stick it up on a domain somewhere or something like that. 25:41 I am absolutely sure—because the longer I work on this I keep on thinking 25:45 of other ways that we can use the data and other reports that we can do, 25:48 other things that are interesting to me. 25:51 People come to me from other parts of the company and ask me stuff, 25:54 and I build reports for them out of it. 25:56 There's going to be so many different things that we can do. 25:58 If I release this publicly and put it up in something like GitHub 26:01 where we can edit it as a group, would people in this room be interested 26:04 in working with it and improving it for the community? 26:08 Anybody? [applause] Excellent. Excellent. 26:11 The next thing is I would really like to engineer this in such a way 26:16 where you guys can all check your own stuff in your own industries, 26:23 and everyone can upload all of the data to a big combined data set somewhere 26:26 so that we could have an open source version of search metrics. 26:31 Does that sound like a good idea? [applause] Yeah. Excellent. 26:34 Right. I will at some point in the next seven days register a domain. 26:38 I'll think of a name for it. 26:45 Stick it up there so that you can download all of the assets. 26:47 I'll put some original data sets in there. 26:49 If anybody wants to have this working with something other than Advanced Web Ranking 26:52 I'm sure we could do that as well. 26:56 I will put everything up onto the web. 26:59 If you follow me I'll Tweet about it at some point probably late next week, 27:01 say where it's up, and we can get this started. 27:05 We can build our own index of the web that we can use to look at changes instantly. 27:07 Thank you very much everybody. Any questions? [applause] 27:11 [Female] Thank you. That was great. 27:21 Way to start the day off with a bang. 27:23 [Martin MacDonald] Any questions anybody? I see Katie's got his hand up. 27:27 He's not allowed to ask questions. Hello. 27:29 [Male] Thank you for the presentation. [Martin MacDonald] You're welcome. 27:32 [Male] Is there any solution for local results like Google Places? 27:34 [Martin MacDonald] Yeah. 100%. So I'm not just doing SEO results. 27:39 I'm also doing all of PPC on a state-by-state basis in the U.S. 27:42 So it's using Advanced Web Ranking to scrape Google with. 27:47 I originally started scraping Google directly, but it was a pain in the ass 27:50 whenever anything changed. It just wasn't easy. 27:54 If you use AWR it gives you a 1,000 odds different search engine definitions. 27:57 Is that right? Two thousand. Two thousand odds. 28:03 So that's like, for instance, Google, PPC, U.S.A., Texas. 28:06 It goes down to that level of granularity. 28:11 You can use any of that data in this. 28:13 So I'm using the U.K., Spain, France, all the rest of it for localized versions. 28:16 [male speaking] [Martin MacDonald] Google Places? 28:21 Yes. Yes. Sorry. The guy that wrote the software sat over there, 28:27 so it's quite easy for me to check while he sat in the room. 28:31 But yes, it also does Google Places. 28:33 Any other questions anybody? 28:37 [Male] What's the catch? [Martin MacDonald] What's the catch? 28:39 The catch is you have to follow @searchmartin. 28:42 You do anyway right? So you can have access to it. 28:45 Any other questions? No. Thank you very much everyone. 28:50 [applause] 28:56
You need to sign up for Treehouse in order to download course files.Sign up