The audio for this podcast can be downloaded at http://highedweb.org/2008/presentations/aps8.mp3
[Intro Music]
Announcer: You’re listening to one in a series of presentations from the 2008 HighEdWeb Conference in Springfield, Missouri.
Kyle James: Guys, I just want to go ahead and warn you, I’m from the South and have a strong Southern slang drawl, so if it’s hard to understand me or if I’m talking too fast, just yell at me. Wofford’s a small institution— about 1,400 students—so probably some of the stuff we do is a little bit easier to control, and I’ve got a little more control than some people. So just keep that in mind. And doteduguru—maybe you’ve heard of it maybe you haven’t—is a blog that me and Shane and a few other people write on, and 95% of everything I’m about to talk about is up on the blog. If you’re crowded and cramped and want to go see another presentation, I will not be offended. So what are we going to talk about here? We’re going to talk about building the argument, key performance indicators, understanding terminology, all things Google Analytics, some additional stuff, and then four rules that I have kind of made up. And if somebody could tell me five minutes before when time is up, because I’m not going to make it through all this but I want to make sure I get these four rules.
If a tree falls in the forest, and no one is around to hear it, does it make a sound? If you have a website, and no one is coming to it, does it matter? Think about that. And how do you know? Well, you use analytics. You want to do something; your boss wants to do something else. Where people are going and what people are doing on your site matters. It really can make a difference on what you should spend your time on. So, are you listening to what your visitors are telling you? They are talking to you— whether you know it or not—they’re subconsciously leaving little data bits; you know, tracking where they’re going on your site, what they’re engaging in, what’s keeping their attention and what’s not, and understanding that can really help you make good decisions.
So, how do you make decisions? Is it your boss who tells you what you need to do so you do it? Probably most of you, that’s the case. You know a lot of times they think we just go around and play on the computer all day. And yeah, we do that sometimes, but we also want credibility. I think everybody here wants to know that what we do is important and it matters. I was telling somebody last night that: Who else controls as much of your institution as you do? You could go set up a redirector button on your home page to a porn site. Your president could go make a speech that doesn’t have that much impact. (Laughter) We hold the keys to the institution. What we do is important, so making sure that they realize that is important.
So KPI—Key Performance Indicators—getting into a little bit into the business here, and it is easy to break that up into four categories. You’ve got sources, success, users, and content. Now I’m not going to spend a lot of time on this, but it is important that when you’re trying to decide what your key performance indicators are that you can break them up into these four categories and then dig in a little bit more in your planning process. So what’s important? What do we care about? If you’re an institution, number one, whether you’re in that office or not, students applying to your school and coming to your school is number one. That has got to be something that we all agree on that is important for your website to do. What else? What’s important? Maybe we spend a lot of money on videos; maybe videos are important and that’s something we want to measure. How about an interactive map? Isn’t it kind of important to know if people are using that? Maybe we want to know what people subscribe to the RSS Feed of our newsroom, sign up for email updates? What are the real important business decisions that matter to you? Defining those and assigning those, before you even look at any of the analytics. The data can be a mess, but as Zane told us yesterday, you’ve got to filter it and segment it before you can do anything. So, decide what’s important to you. Maybe it’s people downloading and viewing your view book—maybe that’s important. And we’ll show you how to track that.
So just going through some terms here: visits, page visits, absolute unique visits, number of new visits, traffic source, landing page; probably everybody is pretty familiar with that, so I’m not going to spend a lot of time with that. Those are kind of the understood terms, and then we’ve got these misunderstood terms. People think they know what they mean, but do you really know what "average time on sight" means? Pageviews. I’ve written a rant, which you’ll notice here on the left hand corner—I don’t know if you all can see that. All this will go up on the blog soon; I’ll put it up on Slide Share. But I have kind of set redirect links here, just teaching you as we go through here—and there are tons of them because I’ve written tons of this stuff already—will take you to the specific article on the blog with a tracking URL so you can see how that kind of works and you can use it for offline stuff like this. So pageview. What is a page view? Why is it important? I mean, if it’s a blog, one pageview per visit is probably all you care about because if it’s your engaged audience that come, read your content, and leave. They don’t need to go anywhere else because they read your blog. If it’s an institution website, is 3 good? Is 5 good? What’s good? Well, if your most important thing is getting a student to apply for your school and they can do that with in three clicks off your home page, and your page visits are five, well what does it mean? Throw it out the window; page views are irrelevant.
Average time on site. We’ve got an example here: 10:00: someone came to your site. At 10:01: they go to page two. At 10:05, they go to page three. We don’t know when they exit, so how long were they on your site? Five minutes, because in analytics, there is no way to know how long someone was on that last page, because that is not logged. It’s logged when they have a call to the server for the first page—for the first time they visit a page—but once they leave and go somewhere else, we can’t track that. So keep that in mind that average time on site does not track the last page they were on. So if they only visited one page and bounced, we don’t have a time for them. Bounce and exit rates, these are both in Google Analytics. This applies to pretty much any tool that you use. A bounce is always an exit, but an exit is not always a bounce. A bounce means someone comes to your site and then leaves. An exit means they go to other pages through that. An exit is whatever the last page they were on. Now if you’ve got, say, Wofford, for example, has email that you can get to from any page on their site, which is on another sub-domain—it’s not tracked—so that always looks like a bounce rate. So kind of keep that in mind when you’re looking at specific pages that you might have a high bounce rate, but what could be causing that? Maybe there are other things that you want them to do that causes a bounce rate, but they are still engaged in your site.
Traffic types. You’ve got multiple types of traffic. Usually the direct none means someone stuck it directly in their browser. They bookmarked it. Referral means another site they visited. And then search, you’ve got organic; most of it’s going to be organic unless you’re running some sort of [] campaign or searching and marketing kind of stuff. That’s the three main types most people will see. We will talk about how can set up some other, more advanced tracking to segment that even more.
Benchmarking. Alexa, Compete, and Quantcast. They’re three different tracking sources. This is good for comparing how your site ranks to other institutions because you can go to any of these. This is kind of just pulling a report that Quantcast pulls, and it’s kind of cool to see the demographic data. Alexa is international; Compete and Quantcast are US-specific traffics, so I think that these are a little bit more reliable, a little bit more worthy. And just to kind of throw a plug out there: edurank.nucloud.com. Me and some friends are working on a little business venture we’re playing around with. It’s a free tool we set up, where you can go and rank your site. It’s very much still in alpha-beta, so if you find any bugs I’d love to hear about them. I don’t think the email report is quite working right. But you can go there and stick in your website, and it will rank—we run it through all three of these services and do some other stuff, and put out a rank for your site. This is only based on like 1900 .edu domains, and it gives you a percentile of how you rank from. I think Harvard is number one and it goes all the way down to whoever is 1900. It’s kind of a cool little thing that you can put in a decision-maker’s mind and say, "Hey look. Our sites ranking better then our main competition. We’ve got more people are coming to our site than our competition." It helps bring credibility to what we are doing. Google also has some benchmarking data in it, so just to be aware of that. I think you can even segment this out a little bit more based on institutional size. It gives you trends on how you compare against the industry as a whole.
Choosing analytics package. I kind of left an Easter egg in this one. Can anybody figure out what’s wrong with it? Exactly, yeah. I pulled the financial one instead of the analytic. Yeah, there are plenty of free ones out there: Omniture, Webtrends. There are a bunch that you can actually pay for. But this is just a bunch of analytic packages that are out there. I put Google Analytics in the center because I really like Google Analytics.
The ten-ninety rule. Avinosh, who is the web analytics guru guy at Google now, wrote this book, which I highly recommend. It really gets you thinking about what’s important, why is it important. One of his big things is the ten-ninety rule. It’s going to be tough enough getting buy-ins with this kind of stuff, but what he says is spend ten percent of your budget on tools and ninety percent on people. Because you can have a free solution like Google Analytics and pull lots of good data or a $100,000 solution like Webtrends or something or Omniture, and nobody knows how to use it unless you pay good people. So, ten-ninety rule. Who uses Google Analytics in here? A good many of you. Who has more than one profile set up for their institution? A pretty good many of you. Who has sub-domain set up tracking on there? A few of you. So here’s the tracking code, and you can see I have a little line in here that says Set Domain Wafford.edu. This allows you to also set track sub-domains in this main account. There are a few filters you have to apply to make this work right, and we will go through those. Another thing that you can do is track across domains, meaning that if you have a dot com that your admissions site is set up on—your main dot edu site—I think, there is a way you can do that too with just a little extra set up. So yes, you can track more than one domain in Google Analytics. Once you have the code installed, here are two cool little resources—cool little tools—you can use to make sure the code is installed correctly on your site. One is Sitescan and one is Wasp, a Firefox plug-in. Love it. Two little things that help you make sure. Hopefully most of you have a div tag in your footer that you can just dump this in. If you’re doing some of this advanced functionality that we are talking about, then it’s probably more important to get the tag up near the top instead of right at the bottom. Everybody tells you to put the tag at the very bottom, right above your footer or right above your closing body tag, and that is right unless you’re trying to do some of this advanced stuff—this advanced JavaScript stuff, because it needs to call the code before it runs the script.
Data Overload. Who is overloaded with Google Analytics and does not know what to do with all the data? I mean you’ve got all of this stuff coming in and you don’t know what to do with it. Segment and Filter. When you set up your profile, you can set up to 100 profiles in Google Analytics. That means that you can just set up a profile that tracks admission traffic that’s all off campus. You can set up one that tracks a certain sub-directory with only people that come from this city. Whatever you want. Set up as many as you want to; the more you set up the better you are because this really helps you to make decisions, because it’s so hard to make decisions about something. You’ve got this big ol’ bucket of data, and you only care about a portion of it. Figure out what’s important and start segmenting your data. With or without the www, make sure you get that set up right—we’ll talk about that in a second—set up your default page, you exclude your parameter. If you’ve got any kind of session IDs, anything funky going on, go ahead and get that cleaned out, otherwise you’ll have multiple instances of a page coming up in here. And set up Site Search. Setup Site Search if you can. Site search is awesome.
Just to kind of show you, we’ve got 25 or so profiles set up on ours. Most of these are all on our main domain, but you’ll see Admission Only Off-Campus visits. Admission, why do they care about visits on campus? That’s trash data; clean it out of there and only look at off-campus. Same for Alumni. We’ve got one for blogs. Athletic sub-domain traffic; I just want to track that in where I’ve got one that’s just a big one.
A few extra things to throw out. Set up a sand box account—set up a sand box profile—where you can test new filters, new things before you run them out to your main thing, because you want to make sure it’s working right before you do it. And set up a raw data one that does nothing but collect the raw data, because with Google Analytics, once you set up something, you can not go back and change something; it’s from there on. The standardizing data. You’ve got four things here. What’s wrong with that? They’re all the same page, but you need to make sure you have your analytics cleaned a little bit so you don’t end up with four pages telling you the same thing, because that’s hard to read. One of the standard filters is all lower case, and I think that this ought to be like a default filter that you put on every single one. There are a few exceptions, if you’re tracking campaigns in a certain way where you have capital lettering, I can understand people doing that. But for me, this is an always install it. Your www, make a decision and go with it. You need to have this set up right so you’re not getting your data wrong. I’ve seen site out there where if you go to one or the other, it breaks. It’s all about you user guys. It’s all about the usability of your users’ experience. If you go to www. whatever. edu and it breaks because you expect people to come without the w’s, then you need to make sure that that redirects all to one. Set up your 301. Remember to filter, because you don’t always want everything. You know you’ve got different shades of what you want to look at.
Sub-domain traffic. So this is just kind of a filter for if you only want to include sub-domain traffic. But remember you’ve got to have this top line of coding to get that. Putting all-domain traffic filter. This is the one you need to apply on your main domain. Once you set up that extra parameter here, this makes sure that you’re getting all these extra buckets of data here—all this sub-domain traffic into one profile. Exclude traffic filter. You can write regular expressions here and you kind of have to go to the post to find out how to do that. There’s a great little tool that Google has to tell you if you want to take an IP range and get a regular expression written so you’re including a range of IP addresses. This is great for all of us because we all have on-campus stuff. We all have an IP block that we have blocked off, so get rid of that with some of your filters because you don’t care about that data.
Directory filters. This is another money one. Different departments only care about their data, so why not set up profiles for them so that you can hand them just profiles of their data. Just Admission, just Alumni, the gifts or donation. You can give them smaller buckets, and it’s a lot easier for them to make decisions and look at it—and you to look at it—when you’re looking at smaller buckets of data instead of the big picture sometimes. Country filters. You can filter by country, region, city. If you’re a small community college, most likely you don’t care about anything but your city. Set up a filter that just looks at trafficking off-campus, you know, you can set up the IP filter so you’re not getting your campus, but then you’re also looking at trafficking your city, because that’s probably a segment that you care about. You can do that. With filters you might have to play around with them a little bit, because order is important— about which order they come in through. If you have any problems or questions about that, talk to me and I will run through some of that.
Full referral URL. This is just another little one that you can set up. Usually when it shows you referring sites, it only shows you the domain; it doesn’t show you the exact page it came from. Sometimes, you want to know that. And there is a user to find report in the system that you can use to get the big bucket, to get the exact URL they came from. A good example of this is, going back to the blog, I wrote a post about Davidson making it to the Sweet Sixteen and how, looking at their traffic rankings with Alexa and Quantcast, how it spiked business to their site. And it got picked up on one of their forums buried down in their site. I was like: who’s driving this traffic, with a hundred visits one day? So this can kind of help you to dig into some of that.
Tagging and tracking. So now we’ve filtered our data, and now we want to segment and track specific little bits of it. Destination URLs. This URL, you’ll notice, is nothing but tracking code. Anybody can do this in Excel, but here’s just a little report that you can go to HEW/URL Builder. It’s a Google doc spreadsheet that you can stick in your URL, the source campaign and it punches out a destination URL. This is great for setting it up in your email marketing campaigns. That handles what’s going on when they click through or not, what happens when they go to your site. That’s kind of important to know what that specific traffic is doing. You can set this up on social media sites, Facebook, set up those extra parameters so you know what those guys are doing when they come to your site. You can do a redirect; you can set this up on a campaign for off-campus. You sent out postcards that go to a short URL you set up that redirects them to whatever your landing page is with tracking variables on it. This is valuable and doing this in a spreadsheet allows you to know over time and you can share, because it’s Google docs, across all of your constituents, and they can look and see how we’re tracking this, how are we setting this up, what are we using to pass for variables. So it’s really easy doing this in a spreadsheet.
Tracking. Here’s when we are talking about the on-click function when I was mentioning earlier that sometimes you need the code at the top instead of the bottom. This is one of those gremlins. I don’t think that this shows all the data. I am having some trouble sometimes getting it. It just seems like it’s undervalued or maybe people don’t care as much as I think they do. But you can use this, and the perfect example that I use is setting up tracking on our view book. It’s a PDF file so how do you know if anybody looks at it or not? Well this kind of allows you to do that with Google Analytics. You can also use this for all kinds of other stuff; if you’ve got your homepage and you’ve got three links to your Admissions site or you’ve got three links to your Alumni section, set up each one of those and you can know which one of those links is driving the most traffic. Yes, you can do a heat map and look at it that way, but this is just another way to do that. You can also use this with video to bring in your flash data with Google Analytics. You can track to see what videos their viewing, how long they’re viewing them, and whatnot. I still think that this is a little buggy; they’re still developing it, so I don’t think it fully reports, but it is something definitely to pay attention to.
Tagging audience segments. You can tag a visitor when they come to your site with specific audience segments like a prospective student. You can call them donors. It’s great for email marketing campaigns that you drive them to a landing page that’s not on the site anywhere. You can tag them in a certain segment and then know what that specific audience is doing when they go around your site. It’s great for making decisions, but I’ll be honest, I have never used this before. But if somebody has, I would love to talk to them.
Reports. Now that we’ve segmented our data, now that we’ve filtered it, now that we’ve tagged it, let’s look at reports. Sitesearch, mentioned it earlier, I love this one. This one is so important. At Sitesearch, you can get all kinds of stuff from this. This is a usability test; here are your users telling you where your site is deficient. You can tell, in Sitesearch, what page they search from. Maybe it wasn’t always their home page; maybe it was from this page and you’re getting a bunch of people that get down to your Registrar pages and are searching because they can’t find the course catalog or something. That’s important. Digging in that data and knowing where that search started from, how many times they have to redefine that search, knowing what they searched, I mean this is keyword research. I mean they’re giving it to you; they are letting you know what they look for. And this is just the top ten that I pulled but you’ll notice that employment, human resources, and jobs are right there. This must have been the summer when people are wanting to come get a job, but it tells me people are trying to find the job board, and maybe I need to make that more apparent. Another example is that we tend to call them residence halls…your audience calls them campus dorms, guys. That’s a big one right there, and if you’re tagging your pages as campus dorms what you’re getting in search engines and whatnot, it’s also what they’re searching for, so make sure that you have your keyword set up that way. And maybe they gave up, they went to search your site and couldn’t find it and they bounced. They exited.
Key word report. Key word data, this shows you what people are going to Google, Yahoo, or whatever and searching for your site. This is really good for a long tail; you want to really dig down and know what pages are bringing in traffic. The first nine of these are Wofford, and then all of a sudden Betsy Cox novel sticks in there at number ten. What is that? It’s one of our faculty who writes books and obviously is bringing a lot of traffic to this site. It’s good to know these things so you can plan your strategy around what’s important and filter out what is not.
Content by title report. You can also look at the URL—I like the content by title; it’s a little more clean—but if you do any kind of SEO, this is important. Page titles are very important to SEO and you can quickly look and see which pages have good, descriptive titles that make sense to your user and are easy to read. So it’s good stuff.
Referring site report. Who’s driving traffic? Where are people coming from? Terrier fans, right here, is a big one. We’ve got some loyal fans that use forums that go and talk and that links back to the site a bunch. Facebook right there is number five. A lot of people come to your site from Facebook. Tagging and tracking is important because you can more filter, more define, this group. It’s a large group of people coming. This is a pretty cool one that I blogged about a few weeks ago, but the 404 Error Report Page. There is a special code you can stick on your 404 page in Google Analytics that’ll send you to a 404 page. There is even some stuff that you can do if you have a Google webmaster’s tools account set up to add some extra code to your page that will give them some extra options. But you can also set up a custom page with a link to your site map and find out what they are looking for and what sent them there. You can do this right there in this report, and this is money for student workers. Here, go look at this, and go make sense of that, and un-tighten these other domains, tell them to redirect this URL to this, they can tell you which 301’s to set up with this so that you can handle the bigger stuff. It’s all about your user, guys; make sure it’s clean and that they are not running into this kind of pages, and you’ve got to know about them before anything.
Now goal tracking is something, I admit, we don’t do well, but this is the bread-and-butter—this is the most important thing. Putting dollar amounts on conversions. We’re starting to get into this, but if you have a schedule a visit, well you need to set up a conversion go on that. You need to put a price dollar on that and figure out what it’s worth to your institution to get someone to schedule a visit. For us, if we get somebody on campus, that’s the best that we can do cause then we can sell them on what we are all about. Requesting information—there is another good one. Donors. If someone gives, are you tracking that? Are you really knowing what campaigns are driving what? Looking at some of this other stuff we just saw, you can look at how your email marketing is stacking up to your direct mail, to your web traffic, if you have social media and are trying to promote something through that, and splitting all this and then deciding what direction do we need to spend our time and money on?
Wait, there’s more. We have just looked at our site but there are all kinds of other stuff to track out there: newsletters, Facebook, TV linked-in, direct mail. The offline campaigns. Guess what? Some people still think that you can’t measure and track traffic that comes from offline, meaning TV commercials. Who all has seen those commercials that say, "Come to our site, whatever.com/tv10" or something like that. Guess what? They are tracking that, because they probably have a tv9 and a tv8 and a tv7 and a tv30. And they’re measuring each one of these and seeing which one is the most successful. When you send postcards, setting up a redirect kind of like I’m doing here and giving them a short URL that redirects them to a landing page that is specific to your campaign, and then tracking that, I mean, this is valuable stuff. This where we don’t look like a bunch of people playing video games on our computers all day, we are making business decisions for our institution all day. We are telling them, here is where you need to be putting money in. Print. You do postcards with links back to your site. There you go; you need to be doing this. Media. Maybe you broadcast your games online and then send people to a page to a new campaign. If you send them a short URL then redirect them to track some of that stuff—maybe even segment them to a certain group as TV audience, radio audience—you can really use this to help you better understand what you should spend your time on.
This is just kind of looking at two of our email templates, and you’ll notice that they are very link heavy. When we do email, I really try to push our departments to get them to the site. Whatever you do, give them a little bit— they don’t want to read tons of stuff—give them some bullet points and then get them back to the site. Every one of these has tracking on it; every one of these we know what medium sent it to them– we use Bronco–but in our email campaign, you can tell that we had this many click-throughs. Well, yeah you know they clicked through, but when they got to your site, what did they do? You don’t know, so you can get your report.
Here’s just an example of our homecoming email that we sent out. We tried to do a little bit of promoting around our new linked-in group that we set up. It’s nice, pretty, lots of links, and lots of stuff. Of course, we’ve got our data right here about the report, but then here, coming into Google Analytics and then tagging it, you can kind of tell there is a spike here. This is just a little tid-bit of stuff you don’t think about. Even three weeks later, people were still using that email to get to the site and search around. Who would have thought that? Most people think that you get an email, you read it, and delete it, but something like homecoming, if you send out early enough, it gets in their mind and they still remember this email, and they’re still using it to move around your site. This is an important audience; this is our alumni, and we care about these people a lot. Looking at where they go and what they do on the site can help us make better decisions about what is important for us to be optimizing. Maybe they care all about our blogs and don’t care about our videos. That’s important. Spend more time with the blogs and not the videos.
This is a tid-bit. We sent out an email before Christmas last year to prospective students and alumni, kind of promoting a top ten of the year with a big push to our videos and a big push to our blogs. I think we did top ten blog posts and a top ten videos. Surprisingly, all of the alumni clicked on blogs and all the students clicked on videos. The blogs were all directed toward perspective students. Parents read the blogs, students like videos, so if you can make video blogs, which I think is a plug for the presentation this afternoon about that, merging the two has some value. Would have never known that unless we just tested it and measured it.
Blogs. Here we go on blogs. Two of my favorite things for tracking blogs, and to me, these are defaults. If you are not using Feedburner and Sharethis, do it. Feedburner gives you all kinds of good data. How many subscribers you have, what those subscribers are engaging in. Sharethis is great for your social media. What are they doing once they share this content through what resource? Most of them are still doing e-mail. E-mail is not dead. This is for doteduguru. I just went and pulled some data through blogperfume.com/feedanalysis. It’s free open-source thing that can give you some additional data about your RSS feed—you know, you’re subscribers—what are their habits? Are you growing? Are you staggering and whatnot? Sharethis is showing how many page visits, what they’re doing in your site, how are they sharing it, what’s your most popular content that’s getting shared, just some additional data.
This is the Ohio State Social Butterfly; I don’t know if anybody has ever seen this before. They’ve got hundreds of social media, social bookmarking, social networking sites. All this stuff. What do you spend your time on? What’s important? What do you measure? And I throw Twitter in here because I’m a Twitter dork. It’s not really used by students; it’s still a geek toy. Probably Facebook, LinkedIn, and MySpace are important. You should be doing these if you’re not already. Setup a landing page, setup a tracking URLs, and get them pack to your site. It does not hurt anything. If you have a student worker, they know this stuff better than we do. They are perfect to be setting this up and monitoring the campaign. Facebook’s great for perspective students, and great for current students. Our Facebook page—it said we have about 1,400 students—and we’ve got about a thousand fans of our Facebook page. We just started a LinkedIn for an alumni network; it’s also good for current students that want to connect with alumni. Maybe there’s a current student that wants to be a lawyer. He can go get on the LinkedIn alumni network, find a Wafford alum who is a lawyer, and make a connection to him. Maybe it’s valuable to him. That’s another resource that takes no time at all to set up; you’ve got to promote it, and it’s out there and you don’t have to worry about it. MySpace, I think we kind of all are getting more and more to where we do not look at MySpace. Just my personal opinion. College students use Facebook; kids that don’t go to college use MySpace. That’s kind of how I feel, but we still have a presence on there though.
You can monitor these things, too. Finding out what’s important, this is just kind of looking at the growth of our Facebook page. We’ve also got groups set up for alumni. I think we’ve got class reunion groups now too. Our school mascot has a profile that people framed it, and Athletics uses that to push out promos to games and what not. But you could see the growth of it. It goes over 960 here, but like I said, we have crossed over a thousand. Facebook does a good job of giving you some of that. LinkedIn, not so much. This is a Googledoc spreadsheet that I’ve been putting together. Each of these is weeks so it is about two months of data. I think we just crossed 200 this past week, but it’s still continuing to grow. If you do not have good measurement tools, spreadsheets are still great. Figure out what your KPI are, what you really care about the metrics, and still set them up in a spreadsheet. Then you can do custom reports that are relevant to what you care about. With Google docs, it’s easy enough to make an image and put it on somewhere.
Video analytics. I’m not really going to talk about this a lot, but you can see YouTube gives you all kinds of stuff. Popularity, demographics, videos most being watched. This can help you make decisions. This fraternity bid day had the most views of any of them we put up there. Maybe that helps us make a decision in the future about what kind of videos to produce.
Alright, so here are my four web analytic rules to live by:
1. Always be testing. There is always something you can test. There’s always something you can be playing around with. AB tests are great, if you’re there. You can measure two things and see which one’s working, and the one that loses, toss it out and stick something else in there. Always be testing something.
2. Don’t get caught up in the numbers; look at the trends. It’s hard to get caught up in the numbers because they are not 100% accurate. JavaScript, analytic tracking say they’re about 95-97 depending on who you ask, because that’s about the population that has the job descript enabled. Look at the trends, the monthly trends, and see what’s going on in your site when you’re looking at big buckets of data. The more you can segment and filter, the more you can look at individual data. But just looking at starting out, before you’ve got all this stuff set up, trends give you more of an idea of what’s going on and what’s being pushed.
3. Set up a recording schedule and track key metrics. I do this once month. At the first of the month, I spend one day, digging through all of our reports, finding out meaning, and putting together a report that’s meaningful that I can give to the president and the staff. Yeah, they probably don’t read it and they probably throw it away, but at least I’m trying. And it’s important for me to kind of know what’s going on, so that we can make our decisions for the next month looking at the previous month’s data. You know, I mentioned videos and interactive map. If you notice you’re getting a lot of views on the interactive map and not that much on video, then you’ve got a couple of options there. Looking at the trend data, maybe it’s 20% more, you can either spend a little bit more effort trying to market your videos or you can spend more time trying to make your map better because that’s what more people care about. These are all decisions you have to make, but it’s much easier to make them when you’re listening to your audience than when you say, "I think we need to do this," or when your boss says, "I think we should do that." Then you can say, "Well, the data’s telling us something different."
4. Goals. You have to set up goals. You must define what is important to you. Set up goals, track them on a regular basis, and report them. This is so important. If you do not set up goals, you’ll still be lost. You won’t know what you want to do or what you want to accomplish.
Just to kind of throw out a couple of goals that are worth taking a look at: X campus visits scheduled a month. I don’t know, but you probably have an idea about. You get about a hundred campus visits a month. I don’t know, but you probably have an idea about if you get 30 campus visits or 100 campus visits scheduled a month, but what if you could push that up 5% or 20% next month? And then do things online to kind of push that and measure it at the end of the month to see if you met your goal. That would be pretty valuable.
Number of viewbooks downloaded a month. You set this tracking and you’re looking at viewbooks, and you’ve got ten viewbooks mattering. If this is important, maybe you market it a little bit better, using e-mail marketing campaigns, using social media, using your website with graphics to try and download that more. Measure this and see what’s being successful. If you try and you do all this marketing effort and it’s not working, well then next time around, you know not to put your eggs in that bucket.
Lower bounce rate on specific landing pages percent over next month. So you’ve got a landing page, pushing people to go apply to your school. Cool little landing page, you keep it very clean, and there are not a lot of option but to schedule a visit. You try to push it from other channels; you see if maybe you can increase the conversion rate in being able to measure it. That didn’t work. The successes are just as important as the failures, guys. Every time you fail that’s something else you know not to try next time. You try a little bit of this and that, because you have goals and you know what you want to accomplish, so your time is spent on valuable things on your site; you are trying to accomplish specific things that are important to your institution instead of, oh so-and-so wants me to toss this up, so I need to do that now. That’s not bringing students to our school. You have much more leverage when you have these things.
Getting donor e-mail campaign to generate x-amount of dollars over next month. You’ve got a big push, and you want to see if the e-mail campaign can bring in $50,000 next month. You can set up these kinds of things and measure that and track that, and know, well, we reached our goal or we didn’t reach our goal, or what did we do that did reach our goal? This e-mail worked really well and this one did not. You can go back and tweak the one that worked well and try it again.
Setting up these goals. Increase our video exposure x-percent over next month. There’s a lot of ways you can increase video exposure. Do you know which videos are working and which aren’t? You can make more of them work. Maybe you can market them a little bit better or a little bit different. You can try imbedding them in pages and you notice a 20% increase in viewership. Hey, I did something that worked; you know that. Maybe you imbed them in a page and you find 10% less. Okay, well get that back out and get it into a player. Goals. Promote and increase presence on a social network by x%. You want to get everybody on LinkedIn. You know that’s your goal this month, so you use your marketing effort on various pages to get people. Once you hit a critical mass, whatever you decide it is, you kind of let it go from there. Making these decisions at the beginning of the month, of what do you want to accomplish this month?
Any questions? I made it through. That was like 65 slides, everyone. Has anybody done any training to help your marketing people better understand the data in reports? Just to clear that up. In Google Analytics, you can schedule it to send you a report on any basis you want. If you are looking at a specific reporter or dashboard, you can schedule that to send it to you the first of every month, so you get these sub-profiles you have set up. You can schedule them up to send to your Director of Admissions, your alumni directors. To go back to that, sit down in a meeting with them, and go through what you want your goals to be. Once you’ve got clearly defined goals, then you can report to them on specific things that do have meaning to them because now you know what you are looking for.
Our web server is on an external thing and I have access to the server. But if you’ve got a template, you would drop this in your footer tag right there above your closing body tag or at the top. If you have them all templated, you just dump it in there. It’s on all your pages that are templated. That’s what the on-click is for. The links start tracking that….No, it’s just one code, one code. It’s just one snippet of code and if you add that to your domain…
If you’re trying to run a store or you are tracking additional variables, you need that code, you need that inside somewhere. You call the code you need to make sure you know what it is trying to read.
Anything else?.... Thanks Guys.
[Applause]
Announcer: For more presentations from the 2008 HighEdWeb Conference visit HighEdWeb.org/2008 or sign up for our podcast and feed at HighEdWeb.org/podcast.xml
[End of Music]