[00:00:00] Matt Russell: When you look at keywords and queries, you know, there's the traditional buckets of navigational, commercial, transactional, et cetera. And the term we coined in House was sign up intent. What I identified was a very specific subsection of queries on keywords where the user will want something enough to give their email. And so the broad plan was if you can rank well for those, the user will give you their email in order to then access the resource.
[00:00:30] Jeff Dillon: Welcome to a another episode of the Signal.
If you've ever wondered how students actually find research and choose their path into higher education, and how AI is rewriting that journey in real time, this conversation is for you. My guest today is Matt Russell, Director of SEO and Content at Inspira Education Group, an edtech company helping students navigate the road to medical school and college.
Matthew took a less traditional route into search. He earned his BA in history at Queen Mary University of London, where he also worked as a student admissions ambassador, then completed his master's in Modern British History at the University of Cambridge. He brought that researcher's discipline into digital content, and in Spira, he scaled organic reach from 20,000 to more than 400,000 books monthly active users. He's grown the email community from 1,000 to over 120,000 subscribers and earned coverage in Forbes, Entrepreneur, US News, and Fortune. Today, his focus is generative engine optimization and AI visibility, and his team has contributed op EDS on AI and education for institutions including nyu, Stanford, and Rutgers.
Matthew, welcome to the show. Great to have you today.
[00:01:54] Matt Russell: Yeah, thank you so much for having me.
[00:01:55] Jeff Dillon: And I should ask you, should I call you Matt or Matthew?
[00:01:57] Matt Russell: Matt's great.
[00:01:58] Jeff Dillon: Okay, before we get into this topic of search and AI, I I noticed on your LinkedIn your awards of recognition, rock guitar, grades one through five, right next to your Duke of Edinburgh awards. Does music or any other creative outlets inspire how you approach your job?
[00:02:16] Matt Russell: I wouldn't say that the guitar or Duke of Edinburgh is a big, like, hiking and philanthropic award in the uk, but one thing I've noticed in my time in SEO is that SEO is a game. Marketing is almost a game. You have to think about what the end user wants, what they want, and then how you can give it to them in a concise, contrite way. And I'm a big video game guy. I played a lot of strategy games, Sid Meier's Civilization, et cetera, and that game aspect is something that translates really well to my job. For example, there's a Jurassic park simulator too. You have to build your park and keep Your customers happy. So it's optimization game and, and it's the same thing with SEO. You optimize different parts of your website, your web pages in order to satisfy the user and then slowly bring them down the funnel. So I almost treat it as a game. Yeah. And I think the video game experience has helped with that.
[00:03:02] Jeff Dillon: Yeah, I love that. I think your career path is really interesting because history and SEO seem like two completely different disciplines kind of on the surface, but when you think about it, they both involve research and evaluating sources and understanding context and figuring out what information actually matters. And with AI changing how we discover information, those research skills, they might be more relevant than ever. You trained as a historian at Queen Mary and Cambridge. You researched so much medieval manuscripts and many other types of late Victorian content.
How did that path lead you to SEO and content? And what, what habits from historical research still show up in our work today?
[00:03:48] Matt Russell: Absolutely. So at the time when I first started with Inspir, I was in a full time sales job and it was really interesting. I was very engaging. I love working with people and being hands on. But I didn't think it was the right career path for me long term. I know I missed the creative side of things, content writing and just messing around with words essentially. And so I took this part time internship role within for education group, learn everything I could about medical admissions, college admissions. And that's where I really kind of cut my teeth in SEO. Because before that, and this was in 2020, 2021, I knew nothing about SEO. But very quickly I was learning about keywords and meta titles, et cetera. And I kept bugging our SEO consultant to learn more and interrogate. Why do I need to include a keyword? Why do I need to put it in a specific place? Why does it be introduction and the H1? And so I just learned more and more and more until eventually that led to a full time writing job, managing job, and now this director job. And in terms of how it helps my work today, I actually think it's an edge I have over a lot of SEOs. Not to be competitive or anything because not only do you learn how to research things thoroughly in history, but you also learn to interrogate things. Why this source was created, what it's saying, what was the intended purpose, maybe the unintended anecdotal purposes. And for example, if we think about how Google updates its documentation, it's not giving up anything to SEOs, but if you can interrogate it, understand the why, think about the broader picture of search, how Google makes its money. What its problems is currently it's spam content and AI generated slop. You can kind of dig below the surface and understand what they're doing to try and do the algorithm. And I'd say, like experts like Mark Williams, Cook, Lily Ray are really, really good at that also.
[00:05:29] Jeff Dillon: Yeah, I think one of the things we, we sometimes lose sight of in higher ed marketing is that behind all the analytics and conversion rates and the search traffic are there's, there's real students trying to make some pretty significant life decisions.
Having spent years working on university websites, I've seen how easy it is to focus on the technology and forget about the person on the other side. You got to experience the process face to face before moving into digital marketing. Before you're working at EdTech, you were an admissions counselor, I think at QueensMary, sitting on the student side during the open days and clearing.
How has that early, very human view of the applicant experience shaped the way you think about the digital student journeys now?
[00:06:18] Matt Russell: I think ultimately you just understand more about the students, the emotions, what they're feeling. My job was working on the clearing lines, helping students gain admission to Queen Mary. I'd also do campus tours, answer any questions, take them to their interviews with academics, and during that downtime, when they're waiting to go into the room with the consultant who will make or break their application, they ask us questions, what's it like? But they also talk about what they're stressed about because they're applying to five other schools in the uk. And I think what you really get from that human interaction is you just understand that these kids are stressed. They need to perform. Let's take medical admissions, for example. They need to perform and do more than anyone else has done 10 years before. They need a higher MCAT score, a higher gpa, more shattering hours, et cetera, and even more refined application to compete for the same number of spots, but with 10,000 more people applying, they need to be almost a more perfect version of the last applicant, the last cycle. And that's stressful. And I think if you are in this bubble, this marketing bubble, where all you're thinking about is sales, qualified leads, marketing, qualified leads, et cetera, you really lose that perspective of what students are feeling and how you can help them. And I'm grateful for, you know, over my nine years in higher education, those first three years, worse actually speaking to students, understanding their pain points.
And I will say that's led to our success in inspiration because we're really focused on the students what they're worried about and how we can help that directly affect our products, how we train our consultants, and our just overall approach and messaging.
[00:07:50] Jeff Dillon: Yeah, when I think about students applying to medical or law school, the stakes are really high. They're investing years of their lives and potentially hundreds of thousands of dollars in their education.
And there's an enormous amount of information to sort through. And they're trying to figure out not just where they can get accepted, but where they can have the best chance of succeeding. It seems like it's an area where better information and guidance could make a really big difference. Inspira works with students heading into medical school and law school, two of the highest stakes admissions processes out there. What are applicants struggling with most right now? And where is the technology making a real. A real difference for them?
[00:08:32] Matt Russell: It's interesting. I guess I'll answer in two parts. One is maybe a more timely answer then the other is just persistent answer. But it speaks to the heightened competitiveness we mentioned before.
Right now we have a lot of concern in students talking about how they'll afford these things.
We were just chatting the other day in our team and a private medical school now the average cost is around about $400,000 and a public medical school is about $300,000. And with the new federal loan changes, the grad plus loans now are capped at $200,000. That leaves you with at least $100,000 of debt, if not $200,000 if you go to a private medical school, which is an astronomical amount of money, especially if then you don't earn any scholarships or have money from family members that can help. So they're really stressed about how they can afford it. One of our key selling points is that we do scholarship support. And I think we average about 35,000 as an average, what we earn for students. Some students get full rides, others don't. Depends on their profiles. But there's a lot of concern and anxiety about how to pay for it. And then students question whether they should apply or not, which I think is the saddest thing I've heard in a long time, because you could be the most amazing person, have the best stats, and be a generally good person to be a doctor. You could be empathetic, energetic, and just compassionate. But if you can't afford it, that's prohibitive, and then they won't apply. And that's the saddest thing for me because what I personally want and why I've been in this education, this industry for so long is I want to help students access more opportunities, whether that's law school, other grad programs, med school, college, et cetera. We want to share those opportunities with people, but if it's that expensive, we can't do it. So there's a lot of concern about scholarships. The only other thing then is it's less timely. But there is always a persistent concern around GPAs, MCAT scores, and then just how competitive applications are. As I mentioned, the medical school admissions, to pinpoint that is getting harder and harder. I was looking at stats the other day. 55% of applicants on average are rejected from a medical school every year, at least one. That's a crazy statistic. And then when you talk about the top medical Schools, Perlman, Harvard, etc. You're looking at substance rates less than 10% and tens of thousands of people applying. As I mentioned, you almost have to be this bionic, perfect human to get into these programs now, because what a lot of students don't realize, and they ask is, is my 3.2 GPA and my 500 MCAT good enough to get in?
Unfortunately, the answer is no, because the med schools use these test scores and GPAs as a filter almost. If you don't make the cut, it's highly unlikely your application will even be read. Depends on the school, of course. Some, all of them use holistic admissions, but there's a persistent concern around that. And then even if you have a perfect GPA 4.0, and a perfect MCAT score of 528, around about 15% of those are rejected from med schools, too. So even if you're perfect on paper, you still might not get in. And then that's where we come in, is the intangibles. Our consultants know what med schools look for and the kind of the nuances that maybe students wouldn't know otherwise. It's all about consistent narratives. In your application, we use authentic experiences. That's a buzzword, of course, but really showing the admissions committee why you have a unique need to become a physician. And that's so hard to translate. You know, you can ask AI to help you, but AI doesn't know what the admissions consultants know. So it's. There's a lot of concern around test scores and GPAs and just how to differentiate yourself from thousands of other omnia perfect applicants.
[00:12:01] Jeff Dillon: Yeah, to your point about how hard it is to get in just on paper these days, I can testify to that, because when I went to school, this is decades ago. I mean, we're talking. I went to college in the early 90s I remember I graduated high school with like a 3.6 I think and I had a guarantee if you get above 3.5 you can go anywhere in the state and you're in there's just pick your state school. It was just so much easier. And I so now I have two children in graduate school and they both got into their colleges, but they were close to 4.0 students with extracurricular activities. They applied to a bunch of schools. They did not get into a lot of school. Like they got their undergrad. They did awesome. They're in grad school now. But I could not believe, believe these two older all star kids I have like were being declined by any school. It was just incredible. But you helped grow Inspira's organic reach from about 20,000 to more than 400,000 monthly active users. And having managed university websites myself, I know that kind of growth doesn't really happen by accident. It interests me because we often spend so much time chasing the latest SEO tactics when sometimes it's a handful of strategic decisions that really drive these results. And I'm sure there were some experiments along the way that maybe didn't deliver as you expected. But looking back, what decisions moved the needle most and what did you try that turned out to matter less than
[00:13:27] Matt Russell: you expected in terms of the decisions that mattered most? We had a very clear vision about we want to help as many students as possible, give them access to a lot of resources that are as helpful as possible.
And to do that we literally asked students who were signing up for packages with us, what kind of resources do you wish you had? And we looked on a lot of Reddit forums to Quora forums about what people are saying, what they wish they had. So essentially just really understanding the pain points of our mark demographic, which is students, by speaking to people. You know, of course you can use Claude Chatgpt to do that research now, but we're really just we wanted to learn from humans and, and produce those kind of resources. I think that was really the key to success. For example, one of our, I would say most valuable assets is our full length MCAT practice test. It's a few hundred pages I believe with answer explanations. It's really, really in depth. I worked with several of our consultants to produce that all handmade and students love it. You know, we have I think 100 plus hours of MCAT topic focused content too and webinars on our website from our consultants talking about the admissions process.
All of that, those helpful free online resources fed into inspiration Becoming known as this brand that can give people free educational resources. So that was the really key because the key of that was understanding what our users wanted and then giving it to them. And of course there's very specific things too about time to value is very important. Making sure that if you have a URL about MCAT study schedules, you put the schedule at the top, right, make it easy for the user. That also helps SEO rankings, UX metrics, etc. But yeah, really understanding what the users wanted and then giving it to them straight away is the key.
[00:15:10] Jeff Dillon: I remember when I was director of web at my school, we could drive traffic to the website, but getting the anonymous visitor to actually submit a form to start this prospective student, you know, journey was the challenge. The RFI form wasn't always this compelling offer.
You managed to grow an email list from 1,000 to more than 120,000, which tells me you figured out something about the value of exchange. What convinces a prospective student to hand over their email and what should university marketers learn from that?
[00:15:46] Matt Russell: When you look at keywords and queries, you know, there's the traditional buckets of navigational, commercial, transactional, etc. And the term we coined in house was sign up intent. What I identified was a very specific subsection of queries and keywords where the user will want something enough to give their email. And so the broad plan was if you can rank well for those, the user will give you their email in order to then access the resource. So for example, AAMC MCAT score calculator. Because MCAT scores are sometimes different for practice tests, we built a two that will just give them an overarching score. Which med school is best for you? That's another classic one. Which Ivy League school is best for you? We would break down the admission statistics for each Ivy League, match it to someone's profile, and these are very labor intensive quizzes and tools we made. But MCAT practice test too, that is a huge value add for a student that is so much helpful, so many helpful resources in there. And they know that. And I think they know that enough to say, okay, I'm willing to part with my email in order to access this resource. So that was really the key is like identifying those keywords and then just building the content around those. Because once it ranks on the serp, you're golden. You know, if you're in the top position, you're looking around about 30% CTR might be slightly lower now due to AI overviews. So, you know, if 1,000 people are searching it, roughly 300 people will click on URL. Then our conversion rates are pretty solid. So I'd say at least 250of those would then give us their email. And that's what's really scaled that email subscriber list also. And in terms of what university marketers should borrow from that playbook, I think it comes down to identifying again what your user base really is looking for. We know our students want study schedules, practice tests, they want question of the days, they want question banks and they'll give you their email for those like it's that signup intent type of keyword I mentioned. So build those, build them. Which takes a long time. And of course you can use AI, but we worked with our experts to make sure they are top notch, the best you can get, which is a manual process. But that means when you rank the page and the users actually use the resource, they go to Reddit and they recommend it. And then more people see that Reddit post and then you get a nice branded search for this piece like Inspirate Advantage, MCAT Practice test and then that means you rank higher or at least maintain your rankings. And then it's this brilliant cycle where eventually you'll just have this email signups almost automated. The key is just understanding what your users really want.
[00:18:15] Jeff Dillon: You know, right now we are going through one of, I think the biggest shift I've seen in my digital marketing career with AI can answer the question without the student ever even visiting the site. And for universities that have invested heavily in SEO and content, that really changes how we think about visibility and success. Your title now includes geo and AI visibility. For listeners who grew up on traditional SEO, how do you explain the generative Engine optimization and what changes when the answer comes from ChatGPT or Perplexity or Google AI overviews instead of just the 10 blue links?
[00:18:56] Matt Russell: Absolutely. Long story short, compared to the 10 blue links, it just gives you the answer straight away.
You know, the rate at which you can access helpful and relevant information is dramatically increased, which is really helpful for the user. In terms of how to explain Generative Engine optimization at a high level, it's practically SEO in my opinion, but with extra steps. You have to make sure your content is formatted into like nice chunks, which is just short, easily scannable sections, usually with a topical theme. You have to make sure the code in the background, the schema markup is consistent, especially for things like your product Pages to ensure ChatGPT Claude LLMs understand what your products are and we'll get to the off page piece in a minute I think. But that's a big part of it too is being recommended for something is a completely different game for generative optimization than SEO.
[00:19:47] Jeff Dillon: Now I want to add on to that and provide a comment my take because this is getting to my wheelhouse where I talk about this all the time. I spent the last few years of my career on search for higher ed and get your thoughts. No real question here but with these summaries now reducing this, the click through rates, what I'm seeing is traffic for universities is going down and they're freaking out about that a little bit. But what's happening is that traffic that's getting to their site is more qualified because they've got more information. These students are trying to confirm it. They're trusting these AI answers more and more every month and every year. But they still need to verify. So when they get to the website it's more important than ever for these schools to to convert. I talked to one of my professional colleagues here and he recently posted Andy Crestedina from Orbit Media about really tracking the highway on your website. You can look in Google Analytics and find where people are going. What is the highway? Not just all these little roads where people you think people are going, but really find out where people are going once they land on your site. Make sure they can find what they need.
That's what I've seen so many schools struggle with lately. They also don't know tell me if you see this too, how much of their traffic is bot traffic now?
[00:21:00] Matt Russell: Absolutely.
[00:21:00] Jeff Dillon: It's really important to know know that as well. So just one of my hot topics.
[00:21:05] Matt Russell: Well, it's something you see across the industry. If you look at an AHREFS graph of almost any website, organic clicks are decreasing. And I completely agree with that. Take you know, I can see for example with our website, a lot of traffic we've lost is informational intent traffic. It's Harvard Medical School tuition fee acceptance rate, et cetera. Of course we do still get clicks and I do find that the clicks are more qualified. But I think the onus is really on being very clear about how you'll take a user from an informational query and an informational page like how to get to Harvard Medical School and eventually get them to purchasing a consulting package. What does that look like? Is that a CTA in the content? Is that a pop up trying to get their email? Put them in a nurture campaign to slowly push them down the funnel. And I think that's where a lot of teams struggle is having that kind of team wide, coherent cohesion around. This is how we make money and work backwards. This is how we make money. This is a student who is anxious about getting to Harvard, anxious for doing well on the mcat. And then just make the conversion path easy. If there's no CTAs in your content, you'll struggle unless you have very clear CTAs otherwise. But you make it easy for the user, right, to convert. Ultimately make sure they know you're not just this informational entity out there, you're a business. You sell something that will help them. And again, I've said this a few times now, but if you're really creating content around the pain points in your industry. So for us it's anxiety about getting in. It's the stress of medical schools. You know, if you apply to 20 medical schools, which is very common, all of those will send you secondary applications. So you're writing probably around 100 essays within a few weeks. That is mentally stressful. And if then you write an article about that and then conveniently you say, well, work with one of our experts who will not only make your essays better, but make the process less stressful. That's the perfect place to then pitch yourselves. I wouldn't do it aggressively, but we really need to think about how you help the user. Now. I think almost conversion, the conversion pathways needs to be a little more subtle now, I would argue.
[00:23:07] Jeff Dillon: Yeah, I agree. And students aren't just searching for college lists of colleges anymore. They're having these conversations with AI about where they should apply, what programs fit their interests, and even their chances of getting accepted. And it's really fascinating to me that AI is moving from helping students find information to really influencing some of the biggest decisions they'll make. What are you seeing and how those AI answers shape student decisions? And I guess, what should admissions enrollment offices be doing about it? You kind of touched on this a little bit already, I guess.
[00:23:40] Matt Russell: I think this is actually one of the few use cases in around higher education admissions that AI is actually extremely helpful because a lot of the questions we all see is, I have XYZ gpa, xyz, MCAT score and X amount of service hours, shadowing hours, et cetera. Where will I be a competitive applicant before they'd have to trawl through the MSAR purchase subscription, see, roughly, okay, are they hitting the medians of certain med schools? Because that's ideally what you want to aim for so you're not just automatically disqualified because your GPA and MCAT score are lower, but AI can extract where there might be a good fit in terms of their profile in terms of stats. Of course, the soft skills and things like the personal statement is not something you can account for, but at the very least it can build students a relatively good safety list, target list, and then they can go from there and really assess whether it's the right fit for them to the location, program, culture, curriculum, et cetera. In terms of what admissions and enrollment officers should be doing about it. Be clear with the stats. I see a lot of medical schools, for example, they won't always post and especially in the law school space and the MBA space. Also they won't publish class profiles all the time. That container all the information needed like median scores, median GPAs, et cetera, acceptance rates, out of state, acceptance rate in state acceptance rates. I would say if they're much more clear about that, it will help students assess whether they are a viable candidate or not. Of course that might be then against what they want, which is people just to apply and pay.
So it depends on the business use case there. But purely from the student perspective, I think that would be extremely helpful because then AI can scrape that and retrieve an answer. And they say you'll be a competitive applicant. Of these 10 med schools, you'd have a decent chance at these other 10. And then for these reach ones you do have a chance. Here's what you need to do. Of course the here's what you need to do. I wouldn't necessarily trust that's where an expert will come in. But in terms of purely looking at stats and comparing your profile to the class profiles and of matriculants of these med schools, I think AI can do a wonderful job.
[00:25:44] Jeff Dillon: Yeah. INSPIRA has contributed op EDS on AI and education for institutions like NYU and Stan and Rutgers. What's the core argument about where AI belongs in a student's education and where it doesn't?
[00:25:59] Matt Russell: Absolutely. So as I touched on, it's definitely a helpful use case to build a school list, especially against your personal profile, which is different from every other student.
We spoke with our team recently and we surveyed our team and we published an article about it which I can share after the call.
And it's really interesting and unsurprising where our consultants do not recommend using AI for almost any part of your application, whether that's secondary essays, personal statements, et cetera. And I want to hone in on the personal statement for medical school as an example. A big part about that is the why Medicine. Why are you pursuing medicine specifically? Why are you a brilliant applicant? And why would you be a great doctor, essentially.
And a lot of that is down to your personal experiences. For example, if you volunteered at a hospital or clinic, emergency clinic in your local town and you saw someone's leg get amputated, God forbid, I'm sure that's a really important part of your life and that really galvanizes like this is really what I want to do, I want to help these people, et cetera. Maybe it puts you off, but let's say it's the former. That is an authentic experience which you can then talk about in your. And reflect on.
And I think AI can't do that. Of course it can completely fabricate these things, but it can't give you those personal moments that really tell yourself, tell your brain, tell your heart why medicine is for you. It can help you if you describe it, whether that's in a voice note or you type out the whole perspective. It can help you maybe with thoughts around. Here's the messages I can extract from it. Here's how I can describe it. Maybe slightly more nuanced way, but in terms of the core message, AI can't give that to you. It really can't. Where it can be helpful is like secondary essays. You want to make sure those are as school specific as possible.
Some schools are really looking for people who are interested in rural, servicing rural and underserved areas. Brilliant. AI can help you extract that. Like this is really what the med school is looking for. Base your essay around that. Or if it's a private med school, it might be slightly different, but again it can help you with the broad. This is the direction you should maybe go. We do not recommend using it for any kind of writing, especially as admissions departments are starting to use AI detectors in some forms is a bit hush hush about how they're doing. So. But if you get. If I was an admissions officer and I saw 100% AI written essay, I'm not sure I would give that the time of day.
[00:28:20] Jeff Dillon: Right, yeah. Great advice to close us out here, Matt. Look ahead three to five years. What does a student search journey look like? And if a university marketing leader could do just one thing to prepare, what would you tell them?
[00:28:34] Matt Russell: I would say if I worked for a university, they ultimately are looking for a specific type of student for their med school that matches their curriculum, etc.
I would be very clear about the type of student I want, which they often aren't. And there's a lot of hush hush around. Exactly the perfect formula. But if I was looking for students who would move on to be doctors in rural areas or help urban underserved areas, I would say that explicitly because then applicants who aren't quite right for that would filter themselves out naturally. At the end of the day, you know they're looking for a specific type of candidate. And if you say that clearly and explicitly, that's what I would say.
[00:29:12] Jeff Dillon: Yeah. Great advice. Well, Matthew, it was great having you on the show.
I will put links to your Inspira website as well as links to your LinkedIn in the show notes and love your perspective. Thanks for being on.
[00:29:26] Matt Russell: Thanks very much Jeff. Appreciate it.
[00:29:28] Jeff Dillon: That's a wrap of this episode of the Signal.
If today's conversation sparked a new idea or challenged your thinking, that's exactly the point. This show is about cutting through the noise and helping you see what's actually shaping higher ed right now. Please subscribe so you never miss an episode. And if you found this valuable, leave us a quick review. It helps more higher ed leaders find the signal for deeper edtech insights, news and trends delivered monthly Subscribers subscribe to the Signal monthly
[email protected] thanks for listening. We'll see you next time.
