Transcript of Session

How the For-Profit Schools are Profiting Using the Internet - An introduction to the best practices of Performance-Based Advertising

Association of Texas Graduate Schools
September, 2007

The final session of today--well, it's really not the final. It's the final formal session. I'm sure there will be lots of informal sessions at the reception and over dinner. It is, it's from Mark Shay. And Mark's been a friend and connected to academe for many, many years. We've known him from many, many different positions in the past. Previously of Educational Directories Unlimited, now with Halyard, where he's in charge of business development and in charge of academic relations, and he's very, very good at that. He's been a friend of ours for many years.

He's going to share with us some ideas on how we do performance-based advertising. And of course, we don't care about getting more students. Our plates are full, and there's no pressure from closing the gaps from the [inaudible] board, there's no pressure to increase the pipeline, so, Mark, you have an audience here who just doesn't care about more students, but we're going to let you convince us that we should care, and we should earn the students based upon our performance, and you're going to tell us how. Right? Mark Shay.

Mark Shay:

Thank you. Let's see, we started with diversity, and then we did rankings. Now we'll do Internet advertising, and then on to the bar, so I can see where this is going. Hopefully the program is more uplifting at the bar than on the slides, but thanks for the opportunity to be here.

Since I was last with you, our company was sold, so we, the company that was GradSchools.com and StudyAbroad.com is now part of a bigger enterprise. I was able to escape the dot-com rush, but the sort of education rush that is sweeping the investment world decided that this was, it was time to play. In essence, every one of the commercial service companies that you have been dealing with over the years has either been sold or is about to be sold or will be sold in the near term. If you think about, you know, SCT, it was part of SunGard, and SunGard was recently purchased. Aramark, the people who do a lot of your dining hall and laundries, they've now been taken private and taken off the stock market. In our direct sector, direct competitors, Petersons and Noel-Levitz were each acquired in the past year, so there's been a lot of activity, and I think it's indicative of a lot of the trends that are going on.

The Internet is driving us, the Internet is driving a lot of recruiting, and it sort of takes a new skill set, and it is an evolution of a model that I was very fortunate to be part of in the beginning. So I was the founder of StudyAbroad.com, of GradSchools.com. Today they're part of a portfolio of companies that includes admissions services and lead generation. And I thought this whole concept of lead generation and performance-based advertising is one that in the past has brought a lot of ire. People would say, "You can't do it. Title IV violations, and we can't make preferential payments," and we'll talk a little bit later about how that's not true.

What I hope to give you today is an understanding of some of the dynamics that are succeeding in the for-profit world. The University of Phoenix will come out soon with a number that will shock most of you, where they will have now 400,000 students in their ranks worldwide, but primarily in the United States. The majority now being online, but still a very large amount being on campus. The University of Phoenix is about half of the sector that is publicly traded for-profit education companies, but that also includes Capella, DeVry, Walden is part of Laureate and now to us, Walden is being taken private. From a stock analyst's point of view, private is different than a university because there's always this publicly traded private universities versus public universities, like some of you.

So this is a sense of the best practices of those organizations. And hopefully, you can take some of this and apply it to your world, or at least start to move in that direction, because it is a very dynamic marketplace today in searching for new students. And those of you who have a University of Phoenix campus in your neighborhood well know the effect that it will have on a lot of your cash cow programs, a lot of your professional programs, and your part-time programs. So in essence, we could say are you competing? And let's hope that you can start to learn from what those for-profits are doing.

Looking at some of the advertising and marketing that we're all used to, the old model was you paid for the placement, and whether that was radio or television, print, outdoors, go to a college fair, in essence all the costs for that college fair are fixed costs--your travel costs, your airfare, your exhibit fees, your literature you distributed. The original directories that came out were all sort of this fixed fee. And you pretty well had a sense. It was experience.

It was not necessarily very trackable. You know, somebody could come back to the office with a stack of postcards that people filled out from the fair, you could publish a separate phone number on the billboard on the Interstate and therefore track that. But in general, it was not something that gave you a lot of detail as to the performance of that ad. It was always viewed as--even if it was a bad buy from the performance sense--it helped your brand and visibility was good, so you could always sort of cover yourself and say, well, like, "You can't go wrong by buying an ad in the, you know, Dallas Morning News," you know. Usually you had help with agencies creating the ad or placing the ad in production of other ads.

But now the online world gives us the ability to track and measure. With the advantages of Web logs and some of the analytic tools that are out there, you can see how many people looked at an ad, how many people clicked on an ad, what they did after they went back for the first click, and really start to get a sense of what is effective and what is not. You can run multiple ads, show how those ads performed--does red perform better than blue? Do we want to call ourselves a college, university? What are the short language, the words that you use and yet how do they affect performance? To put the asterisk on the next bullet, the fast media that implemented from a technological standpoint, I'm not sure at your end they're fast and easy to implement, but at least with the vendor perspective, they're easy to get done.

You can target these, easy to segment, an example on Google, when someone types in a certain word, a certain key phrase, "graduate psychology degree," you can purchase an exact match of those three, choose two of the three, choose one of the three, do any kind of wildcards, do geotargeting. So there's many ways to isolate the audience that views that ad. On a broader sense, you get a national and international exposure with the Internet that you might not have gotten with some of your old traditional media. And in a sense, the online media is the media most associated with students today. So the content of them on the computer, instant messaging, using some of those tools, buying online ads would align you or sort of brand you in an audience or medium that you're trying to show competency in.

Other advantages. As we'll start to show, that performance can be on a variety of measures. It could be per impression or per time the ad is sold. It could be per click. So when someone clicks on that ad, it could be per lead as to they go to a capture form on someone else's Web site, and then that form is passed to you for a fee. There are folks out there now proposing, agencies proposing there would be per start. So as a measure of quality, the conversion rate of those leads that actually enrolled or started, and there's one company that's kind of gone off the charts with some of these distance, and they're trying to propose a cost per graduate. And their proposition is they will handle all the student relations, all the student dialogue. You as the university just teach, and then when that student graduates, you split the pot.

This concept, then, you'll see it a few times over, zero-cost branding. When you're buying on performance base, the concept still is that when people don't click your ad, you get some visibility. So while the fundamental of the buying is not based on, you know, branding, you in essence get that branding out there while your ad is hopefully being clicked.

Again, segmenting--and it's important, I think, for you to start to feel how you may divide your ad budget into what may be legacy, or some would say your annuities here in this market. You don't really want to advertise for graduate students in a 10-mile radius. You're the only game in town. But to go into Houston, to go into Dallas, to go into Thailand, it's a different game, and you may want to segment that. You may want to segment different types of degrees, whether it's a professional degree, off-campus, satellite. So again, the ability to segment your marketing to the type of degree or format that's offered. And then, ultimately, what the new forms of advertising will do is give you a chance to show a return on your investment. So "I paid an amount of money, I got some leads, it turned into enrollments," you know how advertising works.

In essence, it's very, as simple as ready, set, go. All we need to do is, you know, set some expectations. But in reality, you've got to figure out what happens after that click. And I think one of the most dangerous things, and one of the things that you're going to do in an online program, is there's a lot more to it than just buying online ads and spend it printing view books. Because what you're entering into is an interactive dialogue now, rather than the old sort of packaged dialogue. And if you think of what went in an old admissions packet, it was several stages. There were glitzy sort of view book kind of things that would get you excited about a school, but then there were detailed forms, there were a lot of rules, regulations, and so that packet was never designed to be viewed all at once. But it was delivered all at once, and you hoped it made it to the dining room table or to the desk, and then over time, they would work their way through sort of the first and second rounds of review and to finally filling out the forms.

Today in an interactive world, you're not going to be doing it that way. They get your first impression, they're going to inquire, they're going to ask questions, they're going to step forward, they're going to maybe fill out an inquiry or maybe just fill out an application. There's a very big trend in the undergrad side for the very first official contact with the school to be the application. And so it's estimated that maybe 30% of this year's incoming full applications were unknown to the admissions department, so that the application is now the inquiry. And so that, again, is a new dynamic. It flattens sort of that whole window, where your response time needs to be much quicker. On the undergrad admissions side, I'm sure you're hearing about the summer melt. That, you know, kids are now applying to 10 schools, getting accepted to four, depositing on two, probably starting at one, but you know, when you've got one's on semesters and one's on quarters, it's possible to enroll in two schools and pick one after the second week.

It is a different style of marketing, and I think when you're using the, setting goals using the Web, using interactive, you know, what is your primary goal? Is it that you're just trying to build awareness? Is it general sort of overall institutional advertising, program specific? Is it to maintain a certain level overall or within a certain segment? Are you trying to, you know, set aside a group of budget and then have it pay for itself with ROI? Or, you know, "Heck, we've just got to do something"--right?

As you move forward, thinking about your university and your ability to accept that many more applications, the issue of selectivity--certainly on the undergrad side, the more applications, in many cases, the better your institution. Can you afford to do that at the grad level? Too many applications, too many reviews, may be too much of a burden for your staff. Are there specific enrollment objectives to pursue? Are there certain programs we need to segment? In lots of my conversations with grad deans, there's that role of faculty, and is the graduate school really the spokesperson for all, because they're not making the admissions decision, or are they more just sort of a service arm? And are there those fields where giving leads to a, you know, material science faculty that has a pretty tight core of feeder schools and relationships, they want students from places they know, so unknown leads from unknown sources are not necessarily what they want. So there may need to be segmentation in what you market.

Certainly, the budget aspects of it in your model of where your revenue comes from. I know some schools where they keep the application fees. And so in a sense, a very effective, and a way to look at ROI, is on per-application, is they get to keep the 35 bucks, and they can double the amount of applications. It's a nice little windfall. The one school, when they did that, then that got taken away. So obviously, if you happen to know where there's a big pot of cash, somebody else with a little more clout may take it. Certainly there are alternatives to how you fund that, and maybe one of the considerations is an add-on fee, sort of a surcharge for some of this might be a way to pay for it.

You've got to think about your staff size, their skills, whether they'll answer the phone, who will answer the phone, whether they'll do instant messaging and things like that. Think about what you've done previously with advertising. You now try to correlate that into a much more dynamic. The one thing with online advertising is that it's very proactive. It is very involved. It's not like the old view book cycle, where you just sort of knew that the photographers came here, copywriters came here, and everybody was put into a nice annual cycle. Here this is much more dynamic, and you might need to be more active in managing it. The staff that may not be yours full year.

And I'll mention the sales cycle and the sales funnel. In the business side of things, we'll kind of look at this process of recruiting. At the top, the inputs being leads or interest, quick clicks, going into whatever your system may be of a Web site, the inquiries form, of a response back from that, of something that will engage that student into an ongoing dialogue, trying to bring them to campus, trying to meet faculty, trying to talk with faculty. Some kind of follow-up that keeps them interested, that keeps your school in the forefront with them. Generally, that's done with e mail. To some extent, some telemarketing.

Our experience with the University of Phoenix is that when a student fills out a University of Phoenix inquiry form, they push Submit, and within 90 seconds, your phone is supposed to ring. And in essence, that phone won't stop ringing for 10 days, until you either pick it up and tell them to go away or make some resolution as to whether you're going to go to that school or not. They view their lead cycle as 10 days, and they push it hard through call centers. [Inaudible], you guys are set for that, but there is, too, expected when you file an inquiry form on the Web, that there will be more action other than, "Thanks. Glad you're considering us."

You may want to increase your communication levels by doing maybe not just targeted e-mails but something like newsletters, some form of periodic touch to try to keep that student engaged, and then hopefully in the enrolled student at the bottom.

You know, ideally, in the sales funnel, those red arrows are ones that say no, and they're you saying no to them because they don't qualify, or them saying no to you because your program doesn't meet their needs. And if you can do it in an automated fashion, then you're going to save yourself some time and effort.

And so, you know, as we market and as you try to represent your institution as a whole, but then are stuck with things like departments and faculty and how do you speak for those who may choose not to speak? It's important to try to determine how easily do they get followed up with, any action that you take, then what's your response going to be? How internally are you going to be sharing this information, performance of this? Are you going to be able to go cross-department, or is this something that just gets put into one person's to-do list? The follow-ups, and again, if you do things online, it's going to be much more active follow-up. You let them speak. They're going to ask questions, and you should be there.

And the important thing about, you know, I say "sell," and sell's usually a bad term. But it's, somebody's got to advocate. Somebody's got to be in front of the students promoting, and increasingly, it's going to be detail-oriented. It's, many of you have gone to these fairs. You stand at the fair, somebody comes up, and instantly they've got very detailed questions about some machine or some track of research that your response is, "Go to the school Web site," or, "I'll take your information and we'll have somebody call you back." To do this effectively, to get into the business where there's performance base, you need to kind of incorporate that closing part of the sale into your process.

Some of the ways that we can advertise, some of the ways that the Web can be used to recruit, banners and graphical ads on Web sites. We're all sort of familiar with those. Generally, landscape now. The click-through rates of those ads--and that's the number of people who click the ad versus those who see it--continues to plummet. It is not uncommon now for one-tenth of one percent of the viewers of an ad to click it. And back in the '90s, it was common to have one percent. And we will continue to see this drop. It's just sort of the nature of the game. There are so many Web impressions out there that your ad--no matter how compelling it will be--just won't fit the right audience, and therefore you have to hit the audience over the head many times. In advertising, there's a "theory of three" that says a person will act on the third exposure to an offer. And so whether that's a billboard or whether that's an online ad or whether that's a radio ad, you have to also understand that some of these support another.

E-mail is still very effective, still very powerful, and I know very noisy. We all use e-mail in our day-to-day. It's believed that the next generation won't. But somehow we're still moving along. The social networking, the ideas of having personal pages, the ideas of sort of posting a personal response and a personal question, using a social network, maybe that will overtake. I have a 16-year-old, and until last year, I couldn't send her an e-mail. I had to IM her, I had to send her a text message or call her on her cell phone, but e-mail was just too slow. So she has sort of gotten an e mail account for adults, and so that's kind of her way to communicate with the adult world, but amongst her friends, she's still not doing that. There's, you know, 10,000 text messages on her cell phone every month. You know, now cell phones have got keyboards.

A lot of the things that were sort of nefarious advertising still are out there on the Web, the little pop-ups and the spyware. You have to kind of beware of that. It still is very prevalent. Traditional ads, there is an evolution from paper-based guidebooks and directories to online guidebooks and directories. GradSchool.com, Peterson's being some of them. Those are still very popular, still have a lot of utility. Those have paid inclusions. One of the things to pay attention to as you explore those is whether there's a listing fee or whether it's comprehensive like a Yellow Pages. There's, the term is "paid inclusion" in the Internet, where only those who pay get in the directory, and you see this a lot with some of the online directories, that only the programs that pay to be listed or pay for a lead are the ones that will be listed in those directories.

With search engines--Yahoo, Google, and the like--there is what they would call "organic" or "free" search, which is the main attraction, but there's wrapped around that an advertising model, and so those are per-click ads. There's all, they're something to pay attention to, that they are also part of the networks. So the Google AdWords, if you buy 60% of that inventory runs on Google, but 40% of that inventory runs on other sites. And so it could be The Washington Post, it could be Cars.com, it could be any number of Web sites that will use the Google Ad network in place of its own network, ad network. So you have something to pay attention to.

If, in a sense, there was a presearch, though, there's a goal to try to get to the top of that, because that's very productive. If you can be number one on a search term, then you will get a lot of Web traffic, and you'll in essence avoid a lot of Web advertising need. So they call that "search engine optimization," and that is trying to build your Web pages and your Web sites to be very content-focused and focused on certain key words so that your page will be found first. And now, as part of even of our company, there's a whole generation of companies that's out there just collecting names and then selling those names to you with a varying degree of qualifications, and so they'll be the lead generation vendors.

So buying online ads, the flat-fee model, sort of the original model, it's cost per listing, cost per ad. Some relevance, perhaps, to the traffic or the visibility. CPM is another one. CPM is cost per impression--cost per thousand impressions--so you're buying views, the amount of times that your ad is viewed. Again, this number, back when Yahoo was just starting up, was $45.00. Now it's a couple of bucks per thousand ads.

Cost per click is that ad mechanism on Google and Yahoo Search. Depending on the popularity of the term, the competitiveness of it, those are generally ordered by bid, so the more someone bids per click, the higher on the page they'll be. So that's why you'll see University of Phoenix tends to be more towards the top of those, because they're very sophisticated in what they're buying. It is possible that they would pay $20.00 per click. And if you think about that, it's one click to look at a Web page. That's got to be a pretty good lead, and they're pretty smart folks. So when they're paying $20.00 per click, they know a lot about what that word was and what the conversion rate of that word is.

Cost per acquisition or cost per lead, that would be more on the lead model. Again, using Phoenix, Phoenix will pay its lead providers based on the conversion rate of those leads. And so if you're a high-quality site that doesn't repurpose the lead or sell it to five or six other people, then it's possible that Phoenix would pay you $125.00 for that name. But if you resell it and, you know, you got the name by offering somebody a sweepstakes to win an iPod, then maybe that lead's going to be worth a lot less.

Viewing this model, viewing this evolution, that's one of the successes, if you think about it, with Phoenix, what they have done is their advertising is all variable. They are paying per click, per lead, per acquisition, but everything being variable because in essence, they're not saying no to anybody once you qualify. So if there's a qualified candidate, Phoenix has got room for you. Phoenix can scale. So every 20 students, it's another adjunct professor, every 200 students, it's another server, every 1,000 students, it's another campus center. They're in the publicly traded stock market, there's no real limit to the capital, there's no real limit of faculty. All these things become business problems, so all they need is leads.

And so all these ads that you see out there are all lead-based. So if it's a billboard on the highway, if you, you know, carefully take note of the phone number, but you'll notice that that billboard might be a Clear Channel--who owns a lot of billboard--that might be a Clear Channel phone number. And Clear Channel will be running the Phoenix ad as what they would call a "remnant ad" or "house ads" on that spot. And Clear Channel will collect the lead by using a call center, pass it off to Phoenix, get paid for that.

You might see a television commercial, and it might be Comcast Spotlight, their local division of Comcast. And instead of running commercials for an upcoming program or for their new service, they may be running a Phoenix ad, their own phone number, their own Web click, selling lead to Phoenix. And then certainly, on the Internet, there are a ton of sites that somebody is paying for a click, sending it to a site that looks like a university site, capturing that lead and then selling it to Phoenix.

Phoenix has sort of a quality process where they'll eyeball their lead vendors, go through some scrutiny about the requirements to build their site, but they'll be what they call "microsites," which is five to eight pages, not a whole lot of detail, just enough to kind of get you teased, and then so that you can click into the ad. Some of the lower quality lead vendors, what they'll do is that they'll have that click go right to an inquiry form. And I'm sure you've seen some of these, where they'll say, you know, "Online degrees now. Click here," and right away you don't get any information other than the inquiry form. "Fill out this inquiry form, and we'll tell you more." And the quality of those will determine what those folks get paid.

It's something to think about, and if I can impart one vocabulary word to you today, it would be "cost of acquisition." Try to start thinking about cost of acquisition. What does it cost to have a student start? The for-profits today seem to be about $2,500 across all media. Internet leads for them generally $2,000, a little bit lower. Television ads and those types, more like $3,500 or $4,000. But just, as you can start to look at your investment in advertising and try to correlate it to a cost per acquisition, you'll go a long way to helping you understand and the value of investing in advertising.

And believe me, the folks at Phoenix, they have analysts who are just combing through this stuff, looking through all this data, performance data, trying to constantly improve. The lead vendors are looking at it and constantly improving. When you buy that little block of text on Google, you get 35 characters across. Well, what 35 characters, what words are you going to translate those into? You know, sort of the "Buy Now" button. Right? How big should it be, what color should it be? And those really impact. They are doing all kinds of studies on the Web site as to how big the Inquire button should be, and where should the Inquire button be, and should it be colored, and should it twinkle, and you know, should it move around? I don't think you're there yet. I don't think any of us need to be there yet, but it's, you know, some of those practices we should be hopefully incorporating.

Going back to the flat fee model, the flat fee model is what we're all most familiar with in the traditional schools. The risk is yours, the buyer of the ad. The risk does not come from the publisher. They're going to sell it to you. They make their money, and if it works for you, great. If it doesn't work for you, you may not renew. There are typically, there's no guarantees on that, little maintenance for the advertiser. You place the ad, it works, it doesn't work. Again, you can always fall back to branding. Worst case, we got our name out there. It is for the unsophisticated buyer. You've got a budget. The quickest way to dispense it. We run into this with ad agencies, the old, traditional media ad agencies. Give them a $1 million budget, "Let's see how fast we can spend it, because we get 15% commission. The quicker we can spend it, then the more money we'll make." And then examples of how you might be doing that.

GradSchools.com is a type of ad like that. The buttons that are sold on the site are an annual cost, fixed-fee basis, take what disciplines. I think you have to be careful with the, where you place your ad. Most of the time it's done either good salesmanship from the publisher or, you know, you having a gut reaction to where that might produce. More and more, though, you should be asking for performance. Newspapers have run into all kinds of questions lately abut auditing and their circulation. But if you're buying online, you do want to start to see the performance, how that is measured, how many times that ad was seen, how many times that site was seen. And you know, you're not being billed for performance. Asking for it, getting used to seeing it, will keep everybody honest and will keep everybody sort of started into this game.

If you're buying in cost per thousand or CPM ads, it's sort of the lowest performance. You're going to be guaranteed that your ad gets seen. Generally, this is done through a third-party ad server. It's kind of a trusted mechanism that your ad's been seen by this many people. Because the cost of it is so low, you know, fraction of a penny per view, you're not going to get a whole lot of scam and a whole lot fraud going on here, so that, it is what it is. A thousand ads, a thousand times it's been seen.

The broader-based sites will be offering those. You know, the MySpaces of the world, places that just have tons and tons of ad inventory. And the rate will depend on competitiveness. If you have ever gone to MySpace, you'll notice that there's an awful lot of kind of junk advertising, dating services and types of things, University of Phoenix ads. Because they're just serving up so many billions of impressions, that it's hard to sell. And in essence, a lot of those media, the ad is landscaped. The ad is part of the background. People aren't going into it for the ads, so your ad has to really be compelling to draw people in.

And the little, lower risk for the publisher, they're going to get what they deliver. They have to deliver traffic, but most likely they're not having troubles delivering that because that's what brought you to them. You still get the branding opportunities. About.com is an information site recently purchased by The New York Times, so they sell on a CPM basis.

Here, the conversion rate you would be looking at is click-through rating. The publisher is selling you 1,000 impressions. If they're blank, if they've got crappy creative, then you're going to have crappy click-through. So you're working on click-through rates. So it does, is important to create ad images that are captivating. It is important to test those images. One of the tricks that you could do is a rotation so that you run two or three different versions of an ad so that it has some uniqueness to the visitor if you've seen it more than once. But you also then get a sense of which creative produces better. We did a test just on GradSchools.com ads, and I hate to say it, but red ads produced better than blue ads, you know. So that old adage that red sells--well, in our little test it did.

You have to make sure that what counts as an impression. Make sure that you are actually buying an ad on a page that's viewed and not on something that's constantly refreshed. There are some sites that offer streaming, where the content keeps coming. And in those sites, those ads get refreshed every 15 seconds or every 30 seconds. So if somebody's on a music site and they subscribe and it's a little window that's on their computer all day, streaming music that's in the background, and you're running an ad on that, no one's going to see it. So it's important to try to understand where the ad placement is and whether it's relevant to actually people reading versus just coming onto a Web site.

We recommend with any online advertising that you drive them to what we would call a landing page. A landing page being something unique that you can track and that from that you have an inquiry form. And that in the sense of, think about Amazon.com, every page that you ever go to on Amazon.com, your shopping cart is on the top right. So you always know, whatever you're doing, if it's time to check out or if it's time to finish, you can just push that button and off you go. And maybe you want to think about on your pages and on your sites, that's the Inquire button or the Contact Now or the Admissions Info button, and so no matter what pages they're on, they've still got that consistency of having that mechanism or that closing mechanism on the site. So you're looking at click-through rates, you're monitoring these click-through rates with the CPM type ads.

Cost per click, or pay per click, this is the Google-Yahoo search engine model. There is shared risk that if no one clicks your ad, then that advertiser's not going to make money. The difference between Yahoo and Google, Yahoo is strictly a bid, so that the one who bids the highest will get to the top per click, and Google's a little more sophisticated, that it's the most revenue per click. So they look at the amount of times that your ad gets, it's the most revenue per impression. So they look on it per impression, so it's the most times that your ad gets clicked factored toward the bid, or should I say. So there's a performance measure there, so you could pay more and still not show up well because you don't have compelling ad copy. And on Google, if you have really good ad copy and a really good message, you could be higher up on the ad network because your ad's getting clicked a lot, and therefore Google's making a lot of money. I think Google is now making $1 billion a quarter on ads. A billion dollars a quarter. And there seems to be no, almost no end to what people will bid for some of these words.

You have to be a little bit careful of, on the CPC stuff, again where it gets networked. Some of these syndicate their content, and so the ads will show up on other people's sites. So you want to pay a little bit of attention as to how it gets segmented. If you've never experienced this, I would suggest you do, it's called Google AdWords. It's fairly easy to manage. It can be done with a credit card. It's per-click limits, per-day limits, per-month limits. It all can be set up with a credit card. It's a really great place to kind of play and get used to Internet advertising. So that if you can set aside, can get an intern on it or you can get a grad student on it. But start to fool around with that, because the education you'll get by interacting with it will really help you understand the dynamics behind this. And by putting $20 a day into an ad campaign or a couple of hundred bucks a month into an ad campaign, you can really start to get your hands on this, as get some experience, get some data, and then hopefully get some results.

Again, here you see the Google page. The top three that are shaded are advertisements as are the, sort of one, two, three, four, five, six, down the side. So online degrees, you can see that these are aggregators, these are sort of these foreign spaced advertisers. They're going to pay per click, and then they're going to sell Phoenix or Capella their lead once they generate it. And they're making their money on that margin of paying per click and selling it per lead.

And then where you really want to be is here, because that's the free stuff. And so they call that search engine optimization.

So as you're pondering this, you've got to make sure, you know, where your ads are appearing. There's affiliates, there's all kinds of places that the ads may go, so you want to make sure that you're signing up with somebody reputable, and that when you define your campaign, you're putting some restrictions on it. With the Google campaign there's a little check box is that you want to use affiliates or a network. There's also what they call "content matching." So there's key-word matching, which is the search engines, but there's also content matching. So if The Washington Post or New York Times, as an example, there will be an article, and along the bottom, there will usually be ads by Google, and those are Google's algorithm determining what the article's about, pretending those are the key words and then pulling ads from its match. And the context key words, it's not somebody actively searching for that word, but it's somebody interacting with a Web site that has a subject that matches. And then they do have different performance, and sometimes they have different bidding mechanisms as well.

Your key word selection is hugely important, and if you start to buy words like this, so you want to pay attention and not do generic things. If your advantage is having multiple degrees, then you want to start having multiple key words. And so "graduate degree" may not be as effective as the qualifiers like "chemistry" and regions and "full-time" and start to put some of those key words that really will affect yield rate.

I did a consulting project on, for a wine merchant, actually, and when they used the word "buy," B-U-Y, in all of their key words, their yield rate or their closing factor improved by a factor of four. So people would go to the search engine, and they would type in "red wine" or "burgundy," and it wasn't anything to do with buying wine, but when they put the word "buy" in, it was such an action word, that it had just an incredible impact on the power of their campaign. That, added to the fact that not a lot of their competitors were doing that, allowed them to get those words for even less. So one of the things that you will do is when you have more specific key words, is you'll have less competition. University of Phoenix is not going to compete with your mechanical engineering departments. So if you're going to be advertising mechanical engineering, your competitive space is going to be much less. So you want to be specific in those words so that you can lower your costs, because it will be less competitors.

In its more extreme measures, buying leads, cost per lead, cost per acquisition. This is the, you know, for the advertiser, this is very risky because you're paying somebody a big chunk of money for a name, and one would hope you'd do something with it. But there was recently some story about Career College, the company who had spent tons of money on leads and was not following them up, which led to, you know, some crisis in their management, where they spent several million dollars on lead generation, sent them an invitation e-mail and did nothing else with it. So don't buy leads if you're not going to do anything with them.

There are vendors out there that will offer you leads, will offer you measures of qualities of those leads. We all know that's from some ways, GRE search is a lead gen. The SAT, you know, list is a lead gen. So be aware of the quality of the provider, because certainly they're not all equal.

So one of our other companies is e-learners, and the e-learners model, it's an online directory of online programs. What they will do is have all their advertising drive you to an inquiry form, and that inquiry form is a capture mechanism on the e-learners site. When you fill that out--here in this case Penn State World Campus--is going to pay e-learners a predetermined fee for what came from an ad that looked very much like an ad might have looked on GradSchools.com.

So if you're getting in the lead business, again, know your vendors. Education is very hot as far as Internet advertising goes. It may be one of the highest returning values, and with the crash of mortgages and the funk in the auto market, they're looking to education, and more and more folks are out there capturing leads. One of the big online guides is called ClassesUSA. It is owned by Experian, the credit people. So Experian's got a site called LowerMyBills.com, where you enter in some information, and my guess they are thinking--it's merged, actually, into the same division that handles ClassesUSA. So maybe they'll do a credit check with their lead gen. I don't know. It just shows you that, you know, there is a lot of interest in the education sector. If they can't buy universities, then they want to, investors want to buy companies that service universities.

Asking these questions of your lead vendors, "Are you the only in that e-learners example?" E-learners is very proud of the fact that the average visitor inquires to 1.1 programs. And so that it's not repurposed five and six times. Someone who takes the SAT is going to check, I think, on average, 60 schools. So if you get an SAT list, you're out there with 60 schools. And each one of those is sending an average of three pieces of mail, so an SAT taker generates 180 pieces of mail.

There's all kinds of incentivized offers to be aware of that leads come from--iPod sweepstakes and all kinds of other things. It could be a check box if you filled out a subscription form for something. So you want to make sure that the leads are coming from a quality source.

That one big issue on collecting leads, whether it's from a vendor or it's yourself, is how much data do you actually gather? You want to capture name, address, phone number for follow-up, but is it important to capture citizenship? Is it important to capture country of residence or age or gender or ethnicity? You know, every one of those questions that you ask is going to turn a small percentage of people off. We see the experience that we've had with inquiry forms. If it's a multiple-page inquiry form or an application form, you will lose about a third of your visitors, each page that you give them. So if it's a four-page form and you can drill it down to three pages, you're going to improve by 33%, and it multiplies to try to fit it all in one. You get more than 20 fields of data on an inquiries form, people are getting turned off.

But, on the other hand, if you don't ask enough, people are just going to say, "It's really easy to inquire." So it's one of those debates that I'm sure you all run through with the business reply postcard on a postcard. Do you put postage on it, or do you make the person hustle and put, you know, a 24-cent stamp on it? Is there a clear quality difference to that person who puts the stamp versus the convenience, and you do get jokers with the business reply. The experience that I've always had was that if the postage is enough of a deterrent to add to the quality, that you shouldn't do business reply.

Is CPL the best? It's certainly, if you have an online program with a national footprint, CPL is ideal. There is some debate as to if you have an online program, can you get a national footprint, or are you strictly going to be still in sort of that local footprint that you draw your traditional students from? So that's something to determine whether you're going to get into this lead business. Because mind you, while there are a lot of vendors, and if you choose a vendor and you don't produce, they're not going to get paid either. So odds are that relationship won't be long-lived.

Quality, again, for the performance. Say, "So why isn't everybody buying CPL?" Well, the drilling it down, it is a large business. It's a large-scale business. It's, I can't do a telemarketing operation." You know, leads need response. The way to get value out of a lead is to work the lead. Like I said before, it is sort of that secret weapon. It's why we see Phoenix everywhere, because they can make it this variable cost, that the burden is put on the publisher. Can you go to that extent? Probably not, but it's something to, you know, start to think about as you ponder this.

CPL vendors in general will resist segmentation. They'll resist specialization. They would like to give you all these leads and let you filter them out. It is something that, if you're paying $15.00 and $20.00 per lead and you're getting 50 or 100 of these a day, you can see how it could mushroom real quick, and if somebody quits in the admissions office, and all of a sudden these things start stacking up, it's kind of like Mickey and the broomsticks. And the next thing you know, you've got all these leads and no follow-up, and the phones start ringing off the wall, and complaints start, and it just--it can be risky. But it also can be something that you can use to prove an ad model.

As you think about how sophisticated you'll get with your online advertising, sort of this risk factor grid--the lower the risk, the more, the less you need to manage it. The higher risk, the higher reward, but the more management it needs.

You know, to the issue that we're all facing as to what to do with our money and what our money does, accountability, Internet advertising does give you that ability to track as never before. The things that you're doing today with fixed cost, and I think you need to include salaries. I've had this conversation with a lot of folks. "I haven't had an ad budget. I have a $50,000 ad budget," but yet you have 10 people in your grad office. And you're sending them out, you're putting them on planes, you're getting traveling, and those are fixed costs. Can some of those costs be offset by lead generation or online inquiries? I think that's the transformation that you'll all be weighing and have to weigh to really make effective decisions to see if this can work.

If you think about e-commerce and think about Amazon in what you do, then you can get a sense of how your Web site and how your process work.

Pay-for-performance is in essence commission. The Feds have been clear that commissions paid to employees are not allowed. Performance-based incentives to employees are not allowed. But performance incentives to vendors as an ad vehicle are fully endorsed. So you're not going to risk any of your Title IV funding or any, you know, claims of impropriety if you start buying variable-based or performance-based advertising.

And I think, you know, that some of you might have done this for years with foreign agents where, you know, in a sense for enrollment management, you wanted to get some foreign students. "We had an agent, that agent charged us a couple of grand a kid." You know, interesting, the $2,500 fee on the cost of acquisition may be what some of you have paid to get some students from Pakistan or Asia.

To succeed, what do you do with all these leads, right? And if you get in the lead business, you'll get lots of leads. So just sort of as a general management practice, when you get leads, you'll get them electronically. They'll be prime for a database. You want to put these things in a database. The only way you're going to be able to track these and do any kind of sort of post-game analysis is to put these into a database. When you get a lead, it's imperative that you respond within 24 hours. And whether that's just a, "Thank you, we have your inquiry. Someone will be back in touch with you in two weeks." Whatever the realistic time frame for a follow-up is, you need to give it to them, and you need to let them know right away. So set the tone. So an autoreply to an inquiry form, sort of giving them what's coming next, is very appropriate.

You do want to make multiple contacts. These folks are being, are used to being hounded. If they filled out an inquiry form with Phoenix, that phone's ringing off the wall. So your sense, and probably the sense of university folks as not being too pushy, I think you've got to break that when it comes to the Internet, and you've got to be a little more pushy. The idea that it's going to take multiple contacts. In the business world now, to make an average sale, they estimate it's eight contacts to call a client to make the sale. So those contacts are all on not the telephone. They're a variety of a mix of e-mail, of postcards, maybe, of some kinds of reminders, of phone calls, of leaving effective voice mail. And if you think about it, you're going to do some calling, and the majority of time, people aren't going to be there. So an effective script to get somebody to leave a 35, or 30- to 45-second, powerful voice mail that gives some data, that gives some reminders, is something that should be part of this.

Everything that you do, all those communications, should have some call to action, that "Buy It Now" button. So if it's an e-mail, it's not just to inform. The majority of it may be to inform, but have that follow-up. Lead them to an inquiry, lead them to a Web site, to that landing page, that inquiry form. And then, as you go through this, put all this data in the database. So if you're making calls, if you're doing e-mails, if your database can be empowered to do some of these things automatically--and these CRM-type databases do so--then let the CRM database do the work.

What you want to do, though, is not all these are equal, and you want to pay attention to that. And as you have these in the database, start to find ways to segment the leads. And then, as you get used to doing this, you'll start to find and be able to determine what filters might be to find real highly valued leads and the less valuable leads. And you may find, as a Texas school, that you know, kids from Illinois may be a really good lead because of the weather advantage. But you may find that kids from Arizona or Louisiana may not have that same advantage. So you may be able to develop geographic filters. Or some other metrics that can help you sort of prioritize those leads.

And then as this database becomes more sophisticated--and this is that whole concept of CRM, that customer relations management that you'll see in some of the vendor systems--is that they become intelligent enough to understand what some of these variables are and can prioritize the leads and help stage these leads. So some people may get view book mailings and some people may get postcards and some people may get e-mail. And some people may get three follow-ups and some people may get five. And over time you build this into a knowledge base that makes the recruiting more automated, more effective.

And then consider which of this can be outsourced. Advertising as a whole has generally been your burden. Maybe using some agencies, but as you can see from the for-profits, this whole concept of lead gen really puts the burden on the agency. And so the for-profits are hiring firms to represent them and give you leads. So a question will be, and we're seeing it from the business side of the admissions picture, of, "Are there companies coming in that are going to start outsourcing admissions-type services?" You know, what in the admissions dialogue is sacred to you versus the ability to be given to a vendor? Never suggest that a vendor would make admissions decisions, but could they collect enough applications for you to that whole function outsourced?

So one thing I didn't, I guess, tell you about was my sort of shifting career. Today I don't run GradSchools.com or the EDU division. The company that I founded, it has been handed over to what was my controller, so I have moved up into a corporate job at Halyard and doing, as Mitch had said, some of the corporate relations, academic relations, so I'm kind of the business development person for our four divisions. GoalQuest is a pretty neat series of communication tools that is designed to help improve yield rates from inquiries and student retention. I'm primarily used on the undergrad side, and the retention tools are primarily using the computer and engagement tools, survey mechanisms, quizzes, things like that.

But we're real curious and really eager to try to get them into the grad market, because I know retention's a real big issue. And if you've got a lot of off-campus researching, maybe they're going to be logging into your campus portal every day, but more of them checking into faculty every day. So hopefully you'll be hearing more about GoalQuest. If you've got online degree programs, we suggest you consider e-learners. Hopefully, you're all buying ads on GradSchools.com, and World Class Strategy is a big ad agency, primarily focusing on the for-profits. But I can help you understand what Halyard Education Partners is, if you'd like, and hopefully we'll see you at further venues doing the same kind of thing in the future.

Thanks so much for having me. Any questions?

Unidentified Speaker:

This may not be a fair question, but then [inaudible]. You said something very early on at night that just tears by me is that so many of our contacts are from the graduate adviser or faculty members, who sometimes forget that when faculty and classes are not in session, that they then have a responsibility to have some contact with potential students. So you don't get that quick turnaround. Do you discern or expect the differential in prospective students who are seeking online programs versus face-to-face? Or give an academic discipline that offers both online and face-to-face, do we eventually need to have multiple advisers, one who has the immediate reactions to people who are used to living on the Web and online, and other people might say, "Okay, I'm coming face-to-face. I want to talk to you about graduate assistantships. I may not give you immediate turnaround." Is that an issue that we have to face, or is that too much segmentation?

Mark Shay:

No, I think it's a fairer one to consider if you think about even that 10-day lead cycle that Phoenix says, "You know, if you've inquired about an online degree, after 10 days, if you've not made your decision, we give up on you." So it's a much tighter, and I think because, especially with the bigger programs, depending on when you start. And if you've got cohorts starting every 48 hours, then that's very different than, you know, a once-a-year cycle. So some of it might be when in the year they're contacting you may be a factor in that as well.

With the online degree, I don't know enough to say how important faculty is. I would think, from my experience and what I've heard from you, it would be on-campus, that one-to-one that you get with the faculty is what brings them in. And so they're very selective, and they want to meet the faculty, and it's a very important thing to have that faculty interaction. You know, you want to package that in the message to the student, even if it's not the faculty response. It might be enough to sort of package that, "This is who our faculty is. This is what they do. If you have more questions about the faculty, contact us at the grad office, and we'll arrange," so that your faculty is not overwhelmed with hundreds and hundreds of inquiries. Because the worst thing you could do is forward hundreds of these on, thinking they're replying, they don't do anything. You've paid all that money, or you've assumed that there's that big pipeline. So in some ways, I think you need to know they're a key part of the deal, and maybe they're the closer. If you look at it in that perspective. But you don't have your closer do all the setup work. And then certainly, it's a longer cycle.

Unidentified Speaker:

We ignored you when you said Phoenix would be a competitor with 40,000 students, we ignore them as competition now with 400,000 students. In another decade there will be 4 million. Maybe then we'll consider them a real competitor

[Laughter]

What are we going to do about them ?

Unidentified Speaker:

My question is, what are the factors that went in, where did that $2,500 recruitment number come from? What are, how did that?

Mark Shay:

I think that is their sort of industry average of total marketing spend divided by new starts.

Unidentified Speaker:

And is that for all colleges?

Mark Shay:

No, that is the publicly traded for-profits who are reporting their numbers. And this is kind of the secret--you can follow your competitors, once they become official competitors, because they're having to post their results. And in fact, the for-profit sector--the university sector--has recently gotten a series of investor warnings because of the cost of acquisition keeps growing. And at Phoenix and most of the schools now, it's greater than 20% of their entire revenue is spent in recruiting. And that number seems to be escalating, partially because they're so aggressive and the number of students now has kind of peaked. And I think, you know, Phoenix got a really good start by finding, discovering an underserved sector and serving them well. The real challenge for them is how do they grow beyond that? And they're really running into some struggles, and so that's where they're really having to bid much more than may be fiscally viable.

Unidentified Speaker: [Inaudible - microphone inaccessible.]

Mark Shay:

But now you know their weakness, too. And it's one of their weaknesses is just this insatiable need to grow. And all this investor pressure is forcing that, and then they can't raise tuition, they don't believe. In fact, there was a little bit of a price war that broke out last year, and Phoenix shifted 20% of its enrollments into something called Axia College, which was what they're calling their competitor at the junior colleges. But in reality, it's 20% cheaper, and they did that in response to Career Colleges, American Intercontinental Unit, which dropped prices 20%. So that sent shivers across a lot of people's, you know, thinking, "Oh, god, there's a price war." And I think in everybody's sense it's a price war, because so far, the for-profits haven't come in and discounted. But yet 20, over 20% of their total revenue is used in student recruiting. I mean, that's, if you think about what you guys have as a budget, what you spend in recruiting, and I wish you'd spend that, kind of a little.

Moderator:

Let us continue this over the reception, if we may, because of the timing. Mark, it was delightful.

Mark Shay:

Thank you.