Archie is the Director of Growth at Udemy, and he is an expert at data analysis, PPC, SEO, viral acquisition, email marketing, A/B testing, and copywriting.
TOPIC ARCHIE COVERS
- His role as Director of Growth at Udemy
- What is Udemy
- His expertise in data analysis, PPC, SEO, viral acquisition, email marketing, A/B testing, and copywriting
- What kind of growth he has been experiencing
- How did he use data analysis in terms of what programs he used
- What data is relevant to him
- What did he learn with that process, and what is he offering there now
- What did he learn at Udemy about SEO
- And a whole lot more
LINKS & RESOURCES
WATCH THE INTERVIEW
READ THE TRANSCRIPTION
Bronson: Welcome to another episode of Growth Hacker TV. I brought in Taylor and today I have Archie Abrams with us. Archie, thanks for coming on the program.
Archie: Happy to be.
Bronson: Here. Yeah. Think we’re going to have an exciting talk. You are the director of growth at Udemy, is that right?
Archie: That is right.
Bronson: That’s cool. Tell us real quick, what is Udemy?
Archie: So Udemy is the world’s largest marketplace for on demand, high quality online courses. So we created a platform that enables experts all across the world to create courses and whatever they’re experts in, and that allows us to have an entire library of about 7000 courses to students who want to learn about web development. Come on, Udemy. But of course, our development one of learn about growth hacking. Come on to Udemy, have a growth hacking course. Want to learn about photography? Come on to Udemy by an expert who will teach you that.
Bronson: Yeah, absolutely. I’m a fan of Udemy as an instructor and as a student, I do both. You know, I woke up this morning with new reviews of my courses, so I got to look at them and see what people are saying. You know, most people like it, so that’s a good thing.
Archie: And so as often.
Bronson: Everyone’s well, somebody doesn’t. But that’s like, you know, you can’t win them all. So how long have you been there as the director of marketing? As director of growth, I should say.
Archie: Yeah. Growth, marketing, it’s all blurs. It has been here for about at this point almost a year and a half to join last January and almost June. It’s scary how time goes by.
Bronson: It really is. So you’ve been there long enough to really, you know, learn to see how things actually were to, you know, experience some growth. What kind of growth has you to be experienced in the last year and a half since you’ve been there?
Archie: So we’ve been lucky and fortunate to have some some pretty incredible growth. You know, at this point, we were at 7000 courses. We’ve experienced pretty rapid revenue and student growth. You know, we were fortunate enough in December to raise a $12 million Series B round. That was certainly a great validation from the market that, you know, we’re growing at a very quick rate and continuing to see very strong revenue and student growth. Almost a million students, most of those paying, which is awesome.
Bronson: Yeah. You know, the intro you mentioned it was the largest kind of platform for, you know, learning. And I never heard, you know, it being called the largest before. So you guys must have grown substantially and outpacing the other ones. So that’s good stuff. So tell me, Archie, what does a typical day look like for you? You come in and you get your cup of coffee. You’re the director of growth, you know, what is it? What do you actually do as a director?
Archie: Good question. I think, you know, one of the things that defines day to day life is there really is no typical day. I think what I things you’ll find and I certainly found is you learn a lot of different hats. The most common theme is is really going to focus and really focused in on myself and the team on how do we grow every day and making sure that we are doing the right projects and prioritizing correctly and then whatever it takes to get that done, you figure it out. So some days I’m writing listicle queries. There are a lot of data analysis. Other days it’s really copywriting or doing email marketing or doing some SEO. But every day is kind of very different. But all kind of central theme of is this an opportunity that’s going to grow at 510 X if it works and making sure we’re ruthless about prioritizing correctly?
Bronson: Yeah, and that’s one of the reasons I wanted to have you on the show, because you’re actually doing the stuff. You’re not at such a high level that you just tell the minions what to go do. You’re actually in there working on the copywriting, working on the email, working on those kind of things along with your team, obviously. Yeah. And you know, we’ve had Gaga on the show before and, you know, the co-founder of Udemy and she really walked us through initially how Udemy got instructors, how they got students. He talked to us about how he kind of overcame the chicken and the egg problem and to get things off the ground initially. But we want you to do is to walk us through how you’re continuing to grow the user base, how you continually bring in new students, new instructors, those kinds of things, because it’s a different animal, you know, getting it started and then continuing that growth. There may be some similarity, but it’s also it’s a new feat. It’s a new set of challenges you have. And I want you to walk us through some of the things that you do. And I notice on your LinkedIn bio, you kind of have a list of some of the skill sets you possess around growth, right? So I want to use that as a framework for you to tell me, okay, I’m using this skill and here’s how I’m using it in the context of Udemy to grow the user. So the first one is data analysis, something we talk about a lot on this program. How are you using data analysis in terms of what programs are you using, what data is relevant to you? What are you doing with the data? Talk us through that a little bit.
Archie: Yeah, sure. So, you know, I think I write a lot of SQL queries, so doing a lot of that, I’m pretty into our database looking at what’s actually going on, what’s driving our revenue. So just to kind of step back and at this point, we’re kind of the main strategies where we’re pursuing we’re marketplace and ultimately to talk about we have all of this is supply and one of the benefits of a marketplace is your ability to scale the supply side of the marketplace faster than anybody else can. So how do. Boils down to how we go out and acquire users. That means we have a constructive community. Who’s out there? Who’s willing to promote their content to their audiences because it helps them make more money? On the organic marketplace, it gives an instructor, you know, get a lot more students in there. You’re going to start showing up in our trending sections, etc., etc.. So that’s a big growth strategy for us. And how do we how do we know we’re getting better at that? I run basically every every week, all the time running cohort analysis. So we’re looking up, hey, we sign up an instructor this month. How many new students are they going to bring in that first month? Are we getting better at that over time? So going in looking at instructors who came in five months ago after five months, how many instructor students have they brought? Let’s say it’s 100. And then going out and looking at in the first three months, the instructor who signed up three months ago. How many users are they bringing in? So really looking at that deep cohort analysis and then we say, hey, if that’s going down, what’s the action plan that we can do around it? So in a case of, hey, we’re also noticing that the skew there’s only a few guys who are bringing a lot of us. What we realized, guys, six months ago, hey, this really of our instructors were getting a lot of rep leverage from them, but it’s really just the few top guys who are bringing 90% of these students. We have all this other supply. How can we incentivize those guys to go out and do that? And we realized with all of our instructors, they actually don’t have communities, existing communities to go out and promote, too. So they’re kind of stuck. We realize it’s okay. What are some tactics we can teach instructors to really get them with that who don’t have communities to still promote to their audiences? And we started to do what we started to tie. They talked about instructor promotion, success, the number of students you get into, how successful you’re going to be in the marketplace. So we devised in our discovery a trending section. And now if you notice that our Udemy trending section, it’s actually only paid courses, but it’s just looking at the number of students in those courses, meaning you don’t get a lot of free students, you’re going to trend. So we will have our instructors do is go out to things like Reddit post a free coupon code, get a thousand students from that and you’re going to start trending for the instructor. It’s great. They’re going to make some more money on the marketplace for us. We just got that email address and now we can monetize in market to that user that’s really create that win win and really from the understanding that, hey, we really know we need instructors to promote. We’re seeing in the cohort analysis that these guys are promoting more and more than digging into it even deeper, looking at distribution of who’s promoting it, who’s not. It’s really driven by a few guys. How can we extend the tail so we’re driving, giving more leverage to the top guys. They’re going to do it. But again, seeing that tail saying, hey, let’s move that needle up and that’s going to move business.
Bronson: Sports, you guys are kind of an interesting situation where you’re not only trying to grow your platform, you’re trying to teach instructors how to be growth hackers themselves almost exact in emails like you guys are. Send me an email as an instructor and it’ll actually give me really insightful growth hacks. Like it’ll say one of them was, Hey, find someone that’s teaching a course related to yours but not in competition with your course and exactly cross-promote each other’s courses somehow and build kind of a unity there. And I’m like, Oh, that’s smart of Udemy because they’re now putting tools into my hand that I’ll actually help you guys and me as well. Thanks. You guys doing that? You mentioned something that really cool. You said not just cohort analysis but deep cohort analysis. And I think there’s a distinction there because, you know, we’ve heard a lot about cohort analysis of like, all right, everyone that joined the site in January. Here’s our revenue for that group and then February, the March, whatever. And we’re just looking at revenue usually or number of users or churn. One of those high level things you guys are actually breaking it down to, hey, let’s look at the instructors and how well they’re performing and bring in students as a cohort. And so you’re kind of drilling down in that cohort analysis, is that right?
Archie: That’s right. And then even when I think that cohort can be very deceptive at is you core, it kind of tells you where we look at cause it kind of tells you where your next analogy should be. Meaning it kind of gives you a framework. Oh, I see that we’re really driving more in this case, more students per instructor in this month. But that doesn’t really give you anything. Talk about actionable metrics, right? Either. I tell you anything you need to understand. Well, there’s two hypothesis that one is the on average, the median is the median going up. Median number of students, the instructors bring it. That would mean that all instructors there’s a rising boat. Or is it driven by a skill in just a few guys at the end? We’re kind of getting lucky. Those guys are driving again. And we found out 90, 80% of the students actually the long tail is where the opportunity for office that we can get every instructor is right now is bringing one student, two students, get all of them up to 50, 100, 200. That’s what creates the real lift to really understanding that and in saying, okay, why aren’t those guys promoting? And that’s a lot of customer discovery, right? Why are those guys in that tail promote? Well, maybe they don’t have a community. Maybe they don’t know how to do it so that we can provide actionable tips to those guys that take them to the next level.
Bronson: Yeah, and it’s an interesting strategy. You know, we had the guy that led growth at LinkedIn come on and he talked about that. He said, you know, you know, what he chose to do was double down on the ones bringing in 90% to bring let them bring in even more. But there’s still that long tail of opportunity. And you guys are choosing to go after that long tail and, you know, educate them and help them out more. So it was interesting. Two different strategies, both valid. You know, and you probably would do both end of the day.
Archie: But exactly what we found is exactly I think what we had to clarify what we found was we actually doing we had the the top guys, we had the right process for that. So we already had that built out. And there was actually the boundary on that was the number of what’s a lot of guys know besides their email list. Mm hmm. And we already provide and got a very high touch guy for guys. We had email us, very high touch. But the boundary was kind of the size of their email list, so it wasn’t that much more they could do for the 10%. We’re already getting pretty much the maximum left we could there. It was really the function was dependent on either bringing in more of those guys, which you certainly want to do, but then also saying, okay, the other opportunities to really just raise the boat because that’s where most instructors across the platform don’t really have a community. Most people don’t have an email list of 50,000 people that they get here. Mm hmm. So how do we really win any expert? How do we enable all those guys to go out and get grow that.
Bronson: When you tell them about things like it, you’re allowing them to kind of hijack an audience for a moment.
Bronson: Build a list so they don’t have kind of. Exactly. That’s great. You know, one of the other things you mentioned is kind of a skill set is paperclip. You know, obviously just making sure, you know, that the lifetime value, the customer and the cost of customer acquisition all work out and work together. What do you do with pay per click at Udemy? I mean, I don’t know how much you can reveal, but I mean, I’m sure you can’t. But if you can, I would love to know how much you’re spending on pay per click and where you’re spending it and how it’s working out. Because as an instructor, I’ll be on websites and see an ad for my course on other websites that you guys are buying on, maybe have it sometimes. And so I’m just wondering how you what you’ve learned with that process and what you’re offering there now?
Archie: Yeah, sure. So for us, it’s interesting. So we have a we’re we’re a marketplace. And one of the things that the marketplace talked about is you have an incredible supply. But one of the challenges of supply is, as you know, as an instructor, you have the beneficial the margin is more to you than to us. And so when you’re doing pay per click, your gross margin is really important. That’s ultimately what determines LTV. So for us, when we think about paid acquisition, we think about it is more how can we leverage paid acquisition in a way that feeds our competitive edge, which is scale? And what that means is we do things like retargeting, that stuff. We also do a lot of stuff that’s going on to AdWords and because they’re a huge scale, we can create campaigns and very low KPIs that advertise every single one of our courses and actually take up a lot of traffic that each individual product maybe isn’t getting a lot of traffic, but in aggregate it’s getting a lot in a place where our competitive advantage, which is we have an incredible scale play. So let’s go out and let’s find how we can leverage that scale. And that means extending very deep, not spending a lot for a course because again, the margin can be an issue with spending wide as opposed to deep, and that’s been very effective on Facebook and stuff like that. We’ve got great lift from Facebook, partially because we do some really cool stuff with the with some of the Facebook Connect data we get. So about 50% of Udemy users, they come on, they Facebook Connect and they actually when you get that data, you have a lot of data. And so we’re running what we call affinity analysis, which is basically look at, let’s say, anybody you enroll into at your growth and of course on Udemy. Look at all of their common likes and say, okay, these guys all like Facebook platform, they all like TechCrunch and they all like a random tech blog. Andrew Chen’s What you’re not surrounded by, but they all like. Andrew BLOCK. So when you go create a target group that is advertising Bronson’s course, just to those people who have similar likes to the people who are already paying for blogs. And of course, yeah, in that way we’re able to get really targeted ads and we do it pretty effectively at scale for the whole scheme in a very systematic and efficient way. So again, for us, it’s all about that scale play. And I think when you think about PPC, the key thing is what is your competitive advantage? What are your really good at and can you afford to pay, you know, have an hour. If you have an hour TV of 500 bucks, you can go out and you just need a few key words that you really optimize for and you’re all good. If you have a lower LTV or lower margins, you’ve got to think, how can I scale or how can I pick up cheap traffic across the board? And that’s for us. That’s what we’ve been able to do.
Bronson: Yeah, that’s great. You know, when you talked about leveraging the Facebook data that you guys were. Yeah, you know, light bulbs are going off our mind, as you say that, because, you know, I always think, you know, should I do Twitter, log in, Facebook, log in or whatever, but or just, you know, roll your own kind of stuff. But it’s like when it comes down to it, Facebook is the one that gives you a lot of actionable data that then can be rolled back into an ad network. You know, Twitter, not so much. Google Plus, not so much, but Facebook. I mean, you really are getting solid data. So it’s making me think hard. On growth narrative. You should have a Facebook log in because they all I can do is I can look at all the people that, you know, become a paying member, like what do they have in common on Facebook? And then I can target people like them. So that’s a really cool paper clip growth strategy there. I like that a lot. Overall, right now, have you guys hit a stride with pay per click or now it’s just dumping money in and knowing that it kind of works out for you? Or are you still in the middle of, you know, testing before you really, you know, pour a lot more into it?
Archie: Yeah, I think for for us, for pay per click is always going to be again the marketplace. You have an incredible supply, so you’re going to want to leverage that supply to bring your users in. Typically that’s in structure promotion that’s going to echo a little bit later on. But so paperclip for us, we view it as it’s it’s not our not a primary growth channel because again, our margins are such that we can really have that scale play. But it’s it’s tough to really just dump tons and tons in there. Those are effective strategy, but it’s definitely are going to second or third focus as a business.
Bronson: Yeah, no, that’s great. I ask that because, you know, people don’t realize that when you’re building a company, there can be five things that work, but one or two that really work a that kind of work. And then the fifth one that’s like, well, it’s a net positive, but I’m not going to spend a lot of energy on it. You know, there’s actually a mix that, you know, all these things kind of come together. You mentioned SEO. Let’s talk about SEO a little bit. So, you know, you know, on your LinkedIn page, you talk about on page CEO and off page CEO, what are you doing and what have you learned at Udemy about SEO?
Archie: So it’s interesting. So you know, the biggest thing for SEO when you look at the guys who really dominate SEO, the the trick I always use to see if somebody is getting a lot of SEO traffic is just do something Google you type in site semicolon the URL you, Tamika and you look and you see how many pages indexed do those guys have is that will give you a sense of hey, if you have a lot of pages, you’ll get all the guys who dominate SEO. Think of Yelp, think of even Odesk. Another marketplace example. Think of Amazon, eBay. Those guys have millions upon millions of pages indexed. What that gives you is gives you an incredible long tail traffic that you’re paying nothing for because your your supply is creating all of those pages for you. And that is really how you dominate SEO and the short tail stuff exists. But it’s really hard to actually to really in most verticals to just live off short tail SEO guys again who make SEO work. It’s all in that long tail, particularly when you’re in a consumer facing business and your LTVs aren’t at a B2B sass level where those guys get they can get some short tail to work, but you need a lot of users. It’s going to come from that long tail. So in that case for us it’s all about how can we get our instructors and how can we get our students to create a high text content for us? That’s our that. So again, we think about SEO. It’s how can we create a ton of text content that will allow us to scale the rod number of pages we have indexed because we know each of those pages will get picked up X amount of times. Yeah. So what we do is we, we have a little challenge for us because we’re a video site. Right. And video isn’t great for SEO. Exactly. And so it’s and you guys say it’s great. That could be video. Google can’t know what’s going on. They’re not going to index. So what we’ve done is we’ve done a few things. One, we’ve taken that about that short tail. So when we get a course in, we SEO optimize that. We have a VA process where they go out and they make sure the title tag, the description has the right keywords and we see great lettering that we take every instructor. We just make sure it’s perfect. SCA On some of the the off page stuff, we make sure that instructors, every instructors linking to that course landing page, we get a lot of press. We’re making sure we’re getting that off air stuff. And for us, that kind of takes care of itself because we have, again, a supply as a vested interest in in linking to their content. So that helps a lot of the page SCA the big thing and on page we’re doing that with our our VA process going forward. But we need to scale the amount of text content we have on the site. And so if you want a few different things, one is, is transcriptions. So right now we’re about that.
Archie: Exactly. So the challenge is transcriptions. Once you start paying for something, it all comes down to the unit economics. Right. And just like quick math cost. X amount of dollars per per transcribe page and we get x number of visitors per page. How many visitors sign up? And then from those guys, how much money they spend? And so it’s very simple. Does that justify the cost on a profit, three month, let’s say payback period. So right now what we’re doing is we transcribed a quite a few of our lectures in our testing out some of those hypotheses. We have an estimation they can get 5 to 10 visits per page, sign up rate of 20%, etc., etc., go into the revenue and see if that works. Then we really can scale out our our transcription process. And if we can do that, that’s one way we can really build out our text content. The other way is leveraging our student community more and getting those guys creating better questions in the courses where that’s projects that we’re releasing a few things the next few months that will hopefully allow students to more easily create text content that lives on Udemy that again, we can optimize for here.
Bronson: Yeah. Let me ask you about this too. Like the transcription, will that be behind a paywall or in front of it? And because if it’s in front of it, then they can read the course, you know, and then the course is what you’re paying for. And so some people won’t want to read no matter what, but some would be like, All right, I can read this. And then you’re losing on that issue as well. So do you have to wrestle through that as well?
Archie: That’s a great question. So what we’re attempting to do right now is we’re basically we’re linking from the course landing page to actually go into it. It’s when you click on. So if you look at a Udemy course landing page, it has all of the lectures listed out. So if you click on that, you get a little description of the lecture. If you actually go into the code, actually notice there’s an extra two way kind of hidden lecture style landing page, which just as the transcription and then 2 minutes of that video lecture. So if you’re at a paid course, only 2 minutes of the lecture and it’s not accessible from somebody on that easily accessible on Udemy, those pages are only open to ACA or to Google. So you have traffic coming in and the call to action is actually the full course. Click here, then they go to the landing page and then from the user, a little tricky. They don’t know where the rest of the the transcribed lectures are and their perspective is just this one lecture transcription. So it’s all a little it’s a little tricky, but it’s a, it’s a way of protecting both the paid content they’re paying for it while also getting the SEO benefits from.
Bronson: I gotcha. So that’s great. I’m glad to ask. So you’re using it as a landing page? Almost a page. It’s an extra feature. Come read our courses. It would actually be very difficult to come and read the courses. Exactly. Chose to.
Archie: Exactly get. The way we think about it is almost like, you know, you can pay for blog posts or you can pay for for a transcription. If the transcription work is a lot cheaper than having bloggers kind of go out, freelancers write a lot of blog posts, it becomes very, very expensive. You may, as you turn all these video lectures, pretty cheap in the text content and make it in a way that it’s pretty much, as you said, it’s a squeeze page. It’s basically just designed for SEO traffic. Yeah. Then you have a pretty big win.
Bronson: That’s great. I love that idea. You know, one of the other things that you you talk about is buyer acquisition. And I’m really interested in this one because I don’t know how well it works with you. To me, it doesn’t seem like the kind of thing they can really, you know, put the viral coefficient in its favor, you know, like something like a free kind of, you know, fun app sort of thing. So I’m assuming you’re trying a lot of stuff, but it’s not the main thing. That was my whole acquisition for you guys.
Archie: Yeah. And quite frankly, you’re absolutely right. It doesn’t it doesn’t move the needle. And I think when you think about growth, I think a lot of people think about mostly they hear growth, they think viral acquisition. That’s all growth is. That’s really not what growth is. That works for a certain type of product with a certain type of competitive advantage. Typically that’s communication tools, social tools, etc., etc.. For Udemy, the core experience isn’t social. The product, yeah, there’s some social learning stuff, but quite frankly, most people come on to you if they want content, they want to learn and yeah, they might tell their friends about it, but they’re not going to do it. So we’ve tested some things, particularly e-commerce, which e-commerce marketplace to allow it, incentivize referrals. So refer a friend, get $10. We’ve tried some of those things. You know, they’ll have viral coefficients in the, you know, .2.3 range. Nothing, nothing special. The big thing, though, is when we’re thinking about viral videos for an e-commerce site, you’re really talking about an amplification model of viral. So stepping back to Facebook, you can have a viral coefficient well over one because that’s the nature of the product. But if you’re an e-commerce company, how can you you can get your buck efficient up to a .3.4 and then you’re paying for traffic. You can reduce your your kayak. I doubt it now. And also you can go out and you can spend a lot more money on paid acquisition is often you just lowered your kayak and those are the guys they compete against. They may not have lowered their kayak and you can get a lot more volume. So we’ve done this things like giveaways, which is actually sound simple but very effective. You come on. Enter your email address for a chance to win this. And in order to enter it, you have to like us on Facebook. You have to share it out on Facebook. You do that. You do the viral math on that. You’re going to get a battle coefficient of 0.3.48, a direct decrease in your CPA or your cap, whatever you want to call it. You can go ahead. You spend a lot more money on Facebook or AdWords. So you’re thinking about kind of viral acquisition for an e-commerce marketplace. We think about how does this change my my acquisition cost as opposed to how is this going to be the acquisition strategy that’s going to drive the business forward. But quite frankly, it won’t. Yeah. But it can lower your cap, which becomes really, really important.
Bronson: It’s great insight because if someone else is just buying AdWords based purely off their lifetime value, the customer, but you come in with that and some sort of viral coefficient, even if it’s low, you have an upper hand. Even that value may be identical because you need a little extra. So I love that. I think it’s great insight. You know, it’s funny you mentioned a contest. I remember one month you guys were on a contest. It was, you know, for the top instructors, if you jellybeans, you’ll get a hoodie or something, a sweatshirt. And so I was like, all right, let me send out a few emails. So it’s a enough emails. And then I got a hoodie in the mail. I’m like, awesome. Yeah. So, so, like, you to be hoodie my closet, you know?
Bronson: Yeah. So it works exactly well and sold a bunch of courses just because you guys, you know, prodded me with some cheap price. It works. You know?
Archie: It works. It’s okay. But but, yeah, it’s it’s it’s really interesting when you’re talking about how you incentivize people, in fact, this guy going into a little later but unbelievable just small things how that can really move the needle for people for students that hey, we’re in a contest right now that’s you ever watch it? The most content this month wins $1,000 worth of courses and you’re seeing like this is really designed to engage our top users and really show them some love and it’s amazing guys emailing in a what’s my with my score right now like my week laugh like we got to go in these small small things can make actually a world of difference on your engagement and also your acquisition.
Bronson: Yeah, I know. Those are great ideas. Now, another thing that you do, there is a lot of AB testing, multivariate testing. I can imagine you’re doing a lot of this because you’re playing with probably the copywriting and playing with the, you know, the course and just how you, you know, structure it and stuff. What kind of A-B tests are you guys running and how is that working for you? Is that a big piece that you do systematically or how does that fit into everything?
Archie: Yeah, absolutely. So it’s interesting, I think so. We have kind of a one challenge that we have and this happens, I think to a lot of marketplaces is because the instructors actually, you know, you kind of own that course landing page. There’s actually a kind of it’s a little careful about how much AB testing you do because I think that people writing in saying, wait a minute, I created it because I thought the landing page looked like this and now you’re telling me it looks like this? And okay, that’s not that’s not what we want because that instructor community is so important to us in terms of bringing users and engagement. So where we spend a lot of time with our A-B test is more an email and B that’s a medium we control and then a general discovery. So, you know, right now we’re testing a few different search algorithms, so different ways of sorting courses when people search for them, for instance, you can think of that search pretty, you know, top problem to use you saw by conversion rate. That’s one way, just the simplest way of doing it. I ask operating forces, boom. Well then how do you deal with new courses? How much where do you get to those guys and has that hurt your conversion rate over time? And then also, do you maximize for day one conversion? So just getting people into a course, but then not some courses are better than others, which leads to different to repurchase rates. So do you actually optimize ultimately trying to optimize for LTV? So do you sort of more by engagement then of course we know that drives purchase rates. So we’re testing a few different models, how we rank things in search and really try to understand what’s the best approach for us. And there was an email doing all sorts of tests, particularly mostly around different types of merchandizing of content. So do you give a lot of do you put a lot of merchandise and our case courses right into the email or do you do a lot more? That happens just a big button. Go discover courses and drive them to a specific landing page. Yeah, timing and frequency, particularly to different user segments. So do a lot of optimization. Our welcome flow for email is again, if you think about any ecommerce or basically any company email address, delete is really hot, right when they sign up. And what we see is you don’t get somebody to convert in their first 15 days, they’re not going to convert. Yeah. So you got to be really smart about how you how you welcome them to, to the site. And so we do a lot of testing some stuff that’s really hard, 50% off, 24 hours, get it now versus a lot softer. Things like this is why we start. To me, this is why people are learning on Udemy, etc. etc. and then wait a few days and then give them the harder and see what a good offer. And we notice is actually the first one. Just straight up really hard urgency with a good offer that almost always converts better than anything else you ever did.
Bronson: I was going to that was my next question, because I’ve experienced the exact same thing. There’s nothing like a time limit and a discount. Something about it is magical, right?
Archie: It’s it’s scary how effective that is. That’s why, like, maybe the biggest thing I’ve learned about Udemy is, hey, you could take you doing all the things you want. And what people really want is they want the right offer at the right time. And you got to force them to act on it and make that impulse purchase for that purpose. Right, then, yeah. If you know that.
Bronson: Yeah, you almost have to. It’s like a false scarcity. Like, look, there’s a time limit. Yeah, we created the time limit arbitrarily. It’s based on nothing. And the deal will still be there in a couple of months if you just wait. But for right now, this is the time limit and you have to act. And Jack does something to people and they just, you know, it happens to me. I go and buy stuff for the same reasons. So, you know, it’s like we’re susceptible to it as much as we do it. You know, it’s a part of our products. Like, it just works.
Archie: It works.
Bronson: That’s great. What have you found out in terms of, you know, creating the content in the email as opposed to pushing them back to the site and really digging in there? Have you found that the short emails work or the long ones?
Archie: Great question. So what we found is that for new users, it’s really important to merchandise and book content in emails. Those guys are kind of unfamiliar with what Udemy is, what they can learn on Udemy. So you’ve got to merchandise a little bit. New users just signs up on average in the first session, a user look at two course landing pages. So they really haven’t seen our entire catalog. So we want to do that first email, first you emails of merchandise, show them, hey, we had the web doing the class. We have a good acting class, we have a charity class. We’ll do a lot of segmentation, but we got that merchandise in front of that. But then once the user is familiar with the product, has looked around, has been on the site, has logged in a few times, it’s all about just getting them back to the site and getting to next. Boy, they’re averaging the power of our merchandizing on site to do a lot of the work for us. So it’s really I think that’s usually how you’ll find it. Guys were on promote the product. You got to do a lot more in email. Once they’re familiar with the product, you just want to be getting them back to the site and what the site do the work.
Bronson: Yeah. And do you use the Facebook login data to inform the emails? So, you know, these are interest. And so then you put those kind of merchandise in the email. Do you guys do that?
Archie: So actually so it’s interesting. So what we do for the our interest based or how we segment people based on interest is a lot more on on the actions they take on the site. So we have a little score that we use. Let’s say you come to the site, you look at a course that in the technology category, you get one point from being the technology category. Let’s say you enroll into a free course in the technology category, you get five points. You actually pay for a course. In technology category, you get nine points and then you look at it so that they also did that for the business courses, three design courses. Every night we run a just a drone job, ranks everybody’s preferences and then updates their their segmentation. So we get that data immediately. You come in, you look at only one technology course landing page think your first it we know you’re at least somewhat interested in technology and we use that to inform all of the emails and all the merchandizing we’re doing.
Bronson: Gotcha. So that’s how you guys know what to put in my emails on the consumer side of things. I was wondering because I look through it and I’m like, I didn’t check any preferences and log in with Facebook. I actually use an email address. So I was like, How are they doing that? And then I thought, I don’t think they’re tracking at that degree, but they might be tracking what categories I’m looking at and buying it. And you are. So that’s cool. That’s awesome. What have you learned about timing on emails in terms of, you know, when you send them out, have you found that certain times and certain days work or is it just for certain segments you’ve learned certain things?
Archie: Yeah, great question. So I think the for us that we’ve noticed is for guys who are really engaged with the site sending most engagement happens during during the week. You know it’s a big follow up both in terms of log ins and in purchases that are happening on the weekend. That’s pretty typical, I think for a lot of from both sides. People are aware they’re not buying. So the key thing is, is starting it typically early in the week does better said Tuesday and typically around lunchtime. You know as most of our purchases peak around midday and you get the use case for for you to me what I’d say again sort of later about kind of educational products. Educational products are fundamentally aspirational products. Most people say if I only knew more about web development, I could get a better job. I could launch my start up. I only knew about yoga. I could lose weight or whatever it is.
Bronson: They’re on break when they’re daydreaming.
Archie: Exactly, exactly. They’re bored at work and they get an offer that hits their inbox around 11 noon, 1:00, they come over. You know what? I’m really bored of my and my job. Great if I did this and eBay purchase. So, you know, this is really this is kind of times how when you timing and also how you position the emails it’s all about tying into what does the user all do. Why are they buying these courses? They’re really making aspirational purchase. And so the timing needs to reflect that the messaging is reflect that and the impulsivity needs reflect. And that’s why price impulse works well because when you make an aspirational purchase, you’re not really doing it rationally. You’re doing it because you feel feel good about making that purchase. And so you give that urgency, good offer, right time, right messaging, you convert really well.
Bronson: That’s such good advice because you know, there is no one answer to any of these questions. Now, it depends on the psychology of the people using your product or using your site. And so you have to really learn your user. Where are you guys? It’s aspirational, it’s lunch time. It’s them daydreaming and making a quick purchase with a deal that may be completely irrelevant for some of the product, but it is what you guys need to do exactly. The goal is to go back to people watching this and go out and figure out what you need to do based on what you’re learning. But use this as a framework on how to think about it.
Archie: Exactly. Absolutely. In different types of you know, for us that more engagement style email. So engagement happens at night, not during work. So we’re sending out more engagement reminders about, hey, take your course. Those are going out at night. And because that’s when most of the consumption is happening. So we get to your point, really understand even within your segments, your user base where they perform in certain actions and really dial in on that. Prosper does happen around noon. Engagement happens at night when people come home from work and are now starting to work on this.
Bronson: Yeah. Talk to us about copywriting a little bit. You know, we talk through copywriting a little bit with Abby stuff and with email. Is copywriting a big part of what you do is finding the right words for things?
Archie: Yeah, you know, I wish it was something that I actually, as I kind of learned maybe the hard way, how important copywriting really is. When I first, you know, even a year and a half ago really thought, hey, copywriting is kind of this not that important. Copywriting really matters. And my advice is that it does. And my advice to anybody out there, go learn copywriting and go learn how to write good copy because it will make the world of difference. Particularly, I think I will try to think about always is make sure you’re describing the benefits and what is in it for the customer. Many people in their copy we did this so lot described the features, oh, you can watch the course on demand, but what does that mean? Which means you can watch it when you are coming from work. It means you can watch it when you’re on the train. I mean, you can watch it wherever you go so you can learn whenever you want. And that and describing those benefits really makes all the difference. And the other thing about copyright that we found is simple direct sentences and cost actions work better, stuff like that. People hate it. Click here, you tell people what to do and people do it at a much higher rates than if you’re trying to be all fancy and secretive. Now, people are pretty. They’re not taking much time. You hit the benefits yet. Why they should do this and you give them a very clear, direct call to action that tells people what to do. And that’s how you get the most left. And, you know, something for I think kind of people think about is when you’re writing copy, you often try to write it. Well, what I really like getting this email, do I feel good about this? And it’s actually not the right way to do it because you’re so inside the product, you understand it, you get it, get to really make it simple and you got to do stuff that maybe at first doesn’t quite feel right, but then you notice the results are very, very different in very counterintuitive. The more direct you are, the even the harder you are. In describing and selling the benefits, the better you’ll be.
Bronson: Yeah, that’s great. And for anybody watching, you know, to go back and check out Joanna Webb’s interview. Yeah, because her whole thing was on copywriting and I learned so much is doing that interview because I was asking questions and I wanted to know the answer to and she answered on. I was like, This is great. I’m going to using all this stuff right now. So I went through a number of things here. We talked about data analysis, pay per click, SEO, buyer acquisition, A-B test, email marketing, copywriting, kind of the whole matrix of things that go into to growth hacking in a sense, out of everything that Udemy is currently doing, what’s performed best for you guys? If you had to pick one thing on that list that you would not let go of and you could only pick one, what do you think it might be about?
Archie: Question would be email marketing. Really? Yes, email marketing for us. You know, ultimately growth is about driving LTV. In our case, an e-commerce marketplace and in other sites is retention in pages. But if you don’t have a an ability to reach people who are dressed in your product and have expressed interest, you will not be successful. And email drives a huge percentage of our revenue every month. Yeah astronomical and email ultimately it’s kind of people only talk about email it seems kind of passé, you know works and you try to build your. Follow me on Twitter. You tried to build it on Facebook. I think Facebook clamps down on how they are on their edge and on how their their news algorithm. Hey, you’re going to lose that. Twitter, same thing. Twitter actually kind of sucks for a lot of e-commerce, GM or VR type stuff. Email works. And if you take an email list, I’d much rather have an email list of 10,000 than just have 100,000 users come to the site, sign up. Then you never email marketing to them to 10,000 emails that you’re really marketing to will always perform better than 100,000 that you have to your site and you don’t email it at all.
Bronson: Yeah, I’m so glad you’re saying that because, you know, this is one of those things that until you’ve actually done it, people are confused by what these numbers actually mean, right? So for instance, on growth, RTV, when they’re watching the free video, you know that little pop up that stops them. At first I was like, All right, let me have a tweet, something out to continue the video. And then I thought, I know too much about Twitter. I’m not going to kid myself. Let me get their email. That’s the game that has long term value. And so to give an example, if somebody really important with, let’s say, 50,000 followers, targeted followers on Twitter, they want to blast out a tweet about growth after TV. It will move the needle just a tiny bit. I’ll get a little bit more traffic and maybe a sign up or two. You know, I send an email to just a few people. I can do way more. And so people don’t realize that a massive Twitter following you just it doesn’t move the needle like people thinks it will. Same thing for Facebook, but email list. I agree with you hundred percent. And that’s why I’m captured emails and not something else.
Bronson: Is there anything that you’ve tried at Udemy that you just not for sure would work? Awesome. And then you do it and just like, well, I didn’t do anything.
Archie: Yeah. Good question. So I think that, you know, for us, the thing that was kind of surprising was actually, as you talked about, a lot of the the Twitter and the Facebook example, we used to go after instruction of big Twitter followers. And the logic was, hey, these guys have 50,000 Twitter followers. We get them. We get them to promote their course. It’s going to do magic. Yeah. Fact it do tend to crap into anything. And if you think about it, you’re always breaking down the kind of just the impressions you get 50,000 Twitter followers that goes out in a stream. How many people are actually going to see that? Exactly. And you look at a few thousand of those who are going to click, you can view a most can have a ETR of 1% at most. And so that it really doesn’t the math never works out. And so when you break it down like that, you realize Twitter and the social stuff doesn’t matter. When you have a paper tweet like yeah, you used to have and the viral on that was your point, you know 5.1. It’s not like this is this is a joke relative to actually doing what’s quarter, which is for us capturing email addresses and then for our instructors finding guys who actually have email list. That’s what’s going to that’s what’s going to move the needle ultimately in social social again, I’ll come back to like Facebook works. Well, if you’re paying to promote your posts and you have a really good offer for people who are willing to buy from you, Facebook can be really effective and we had a lot of leverage from our Facebook page not in the typical like do a lot of like like us on Facebook and we used to have posts registered to back up. We we did when we started going to the yoga vertical, we built a huge, pretty solid Facebook page and we were getting posts with that were getting shared. Basically some of these posts, over 1500 shares. We had a post that had 5000 likes and only 25,000 likes, so pretty ridiculous engagement. The guys actually came over and signed up from Udemy. Was basically next to nothing. Yeah, because it was all this engagement stuff which people talk about the brand and the actually it doesn’t naturally move the needle though. What’s going to move the needle is going out giving somebody an offer for a product they want with the right urgency. And that’s actually what’s going to drive actual dollars and users to your company. So these devices don’t waste your time trying to do a lot of soft stuff that just kind of like you can. You should do a little that you need to you need to have a nice brand, but ultimately it’s going to really move. The needle is good offer right price from people who actually want to buy your product. Yeah. And that’s pretty.
Bronson: Simple. Yeah. I hope people are, like, aware of how good the advice you just gave is. I really hope they slow down and think about it because email marketing, right price, right time, urgency, I mean, it is absolute goal. So I just hope people get how important what you just said is. I know it firsthand. That’s why I agree so wholeheartedly. I won’t do it another way anymore.
Archie: So exactly.
Bronson: It works. Just let me ask is what’s on the horizon? What do you want to try that may or may not work? But then you’re just like, you know, on a rainy day when I got enough resources, I’m going to try this and see if it moves the needle. Anything you got on the on the list there?
Archie: Yeah. I mean, so the biggest thing I think is we talked about that SEO strategy. So we’re really kind of saying a marketplace model. We have all these content creators, we have a lot of content, but it’s video content. And if we can turn that into an SEO really scale with your strategy, that’s a game changer for us. We’re still in the early, early, early stages of testing that out, but we can get that to work. That’s something that we’re really excited about doing and going out. The other thing that I think is really important for us is we’re focused on as a company is really driving our sound growth, but it’s really focused on retention and really making sure that we have users who are coming in or taking their courses. We’re very good at getting somebody to take their first course, but how do you get people to take their second, third, fourth, fifth course and really drive that LTV so you don’t have that high churn? So we love going out, we love getting users. I think we have some good things in place to make sure we do that sustainably. But really it’s they’ve got the CEO and also they got the retention piece or getting people to buy three, four or five, six courses as opposed to making two impulse purchases and not really engage with the content and therefore going on to be lifelong customers.
Bronson: Yeah, you know, it’s funny, so many people come on here and they’re like, you know, we focus on retention, which isn’t very growth oriented, but then everyone says it. So I think it is growth oriented. Everyone brings it up. So retention matters as well. It’s huge.
Archie: Exactly. And just to that point, when you think about ultimately in your business, e-commerce metric or marketplace models, all of that LTV, ANCAK and ultimately LTV isn’t just about paid acquisition. LTV is also about SEO. It’s also about what you can do with content marketing because you can’t put out a piece of content that only gets seen by 500 people. If your LTV is is very, very low, you’re not going to justify the time or the expense of creating that content or pursuing an anchor strategy that that really is pretty small in volume. So when you get your LTV up, you can do a lot more on paid. We can also do a lot more content marketing, you can do a lot more SEO, you can do a lot more interesting things because you have a lot more DAI from your customers. LTV is really, it’s obvious, essential to growing the business.
Bronson: And retention is what brings that LTV up exact. And when that LTV is up, it allows you to try so many other things with so many more resources, but without retention, the entire house of cards falls down.
Archie: Exactly. You’re done. So is it generally.
Bronson: As you know, before coming to Udemy, you actually founded Amplify Teaching? Yeah. Tell us real quick what was amplified teaching.
Archie: Yeah, sure. So our goal is to turn every every classroom teacher into a rock star. And what you when you look at a traditional K-through-12 teacher, we’re focused in that space. It’s really, really hard. It’s teamed up with a former educator, and we created a a tool that is an app that allows teachers to really effectively manage the classroom experience. So you’re a teacher, you’re really trying to say, hey, Brian, to understand this topic, it can actually read like bank notes on that are really overwhelming. They have 100 kids and you go back to your computer and you’re like, Oh, I can’t keep track of all this. So you could take all what they were doing on paper, put that onto a to an app in an app form and allow them to walk around the classroom, enter that data, that data entry really seamlessly, and then allow them to also pull up data. Hey, Bronson really gets really good at math, like give them snacks, extra homework that that would be a game changer for teacher because they could actually both add teaching as opposed to keeping track of information. So that was what the product was. We made some pretty big, big errors, really a learning experience. We raised a little bit of money and then unfortunately, we spent the classic mistake of spending way too much time in product development at that point out to a lot more product stuff. And we didn’t really focus nearly enough on actually getting people to to use the product and more importantly, to pay for the product. Why free usage? But I go to dynamics to K-through-12 and the sales cycle there, but nobody paying for it. So when we came time to hey, we’re actually we’re running out of time wasn’t enough there. And I think in hindsight, we spent a lot of time really trying to find you can add tuning the product as opposed to going out and and selling and hustling and making money.
Bronson: No, I’m so glad you said that, because the kind of lead to my next question, when you found a company in you are the CEO, you are the guy that has to do everything, answer everything, figured out everything. You get a real good taste of why growth matters, why selling matters, how data analysis and pay per click and all this stuff comes into play like is right there, right next to the metal learning all these lessons the hard way. Do you feel like that your experience there, being the founder, has really informed you as the director of growth at Udemy? Has it been a valuable kind of a thing to learn from for you?
Archie: Absolutely. Is by the best experience in terms of going on to be and to be at Udemy because you realize kind of all or pieces of the business and how they all tie together. They also used to wearing multiple hats, doing bit of product, doing a bit of data, doing a bit of marketing and sales. And so Udemy kind of can see that and understand that and also understand, you know, ultimately what the business that is about growing it. You can think in a good just a cleaner way about what actually matters to driving the business forward, which is essential.
Bronson: No, absolutely. And some of the advice that I give people when they want to learn how to you know, how to grow something, how to growth, hack whatever you want to call it. I say just go start something. You start and then watch it fail and you’ll learn something because it’s good that you’re going to mess up. It’s going to be a disaster in some way and you’re going to learn things that you just you can read in a book, but it won’t stick with you the same way when you go out there and just watch your money disappear. It’s like, Ah, I get it.
Archie: Exactly. It’s. Yeah, and I think that’s absolutely, absolutely right. I think a lot of times on paper, especially if you just, you know, read a lot, you watch a lot of it sounds really easy. And then when you actually go to as actually it gets pretty challenging when you actually get in the weeds, you actually understand that it’s competitive out there to acquire any customers. There’s a thousand companies who want the same exact customer that you want and to better beat that that other person. You got to be innovative. And if you’re just copying or just doing everything, somebody tells you a book or something not going to be successful until you really get in the weeds. You take those lessons and you apply it to your own business and you figure out, okay, what’s make my business unique and how can I leverage that to go get more what people are to pay their money for my product?
Bronson: Yeah, absolutely. Actually, this has been an awesome interview to have. Just a few last questions for you here. Yeah, we’ve talked about Udemy. We talked about Amplify teaching. You’ve given us as much tactical advice as anybody possibly could, I think, which has been awesome. Tell me, what’s the single best growth hack that you’ve ever implemented at any point in your life? Just that one thing, maybe it wasn’t repeatable. Maybe it’s not an evergreen strategy that you could build a business, although but what something you did and you’re like, Wow, that really worked.
Archie: It’s a great question. So I’m actually going to give a little on that. I think the best growth attack is actually staying relentlessly focused on what actually moves the needle. And so it’s actually really been about finding out things like, hey, what moves the needle is email marketing and a good offer at the right time and continually finding ways of getting more of those offers into people’s hands in a contextual way. Meaning it’s you can’t just send the same damn offer every time, but different ways of getting feedback. And so, for instance, we did this kind of a weird promotion in November. We did the 12 days of Black Friday. And so it’s very simple. And let me tell you, it was for the guys out there, it start at 70% off and then every day until Black Friday, 3% drop and just a ton of emails about it titled Hello Buyers. And it was very simple. Nothing special. Everybody does that. But the amount of money in the trust was pretty astronomical. And so I feel great when you’re doing that, but it actually really drives and moves revenue. So that kind of growth, I would say is nothing in particular, but just figuring out what really moves the needle for your business and just doing it and doing it right and figure out ways to do it and more and staying focused on that thing. And it’s pretty simple, but that’s actually what gets the job done.
Bronson: That’s awesome. And just I want to make sure people understand what you just talked about with the Black Friday because it’s such a good idea. So there’s a discount of, you know, it starts at 80% off or something, right? Exactly. And then it goes down by 3% every day. So if I wait a week to buy the course, then I get a much less of a discount. But if I buy it right now, I get the full 80% and then tomorrow I’m at 77 and go on from there. And so every day that passes, there’s this little voice inside of you saying, you should probably do it now. You should do it now. And, you know, I watch my sales of my courses gobbled in the time. So that’s why I knew it was working, just because I could see my own data and I’m like, well, that’s, that’s great. I got to remember this growth, if it’s working for them and, you know, vicariously for me as well. Now also, what are some resources you can recommend to us? You know, you obviously have learned a lot. You’ve figured out a lot. You know a lot. Is it from blogs? Is it from reading? Is it just from trial and error? What resources do you really go to to learn some of these things?
Archie: Absolutely. So can I give a shameless pitch but some good, really awesome Udemy courses, your course being one of them on Growth Hacking Growth Hacker Conference, The Copywriting Course on Udemy. There’s a lot of great stuff on Udemy, but then also a lot of great blogs out there. You read Chance Blog, some other, you know, really excellent blogs there. But I think most of what I’ve learned is really by trial and error. And you can spend all the time you want reading, watching stuff. Quite frankly, at a certain point, you’re not going to learn much more anyway. The way you’re actually going to learn it is by actually trying it and just doing it. And so I think what we try to do at Udemy is create a culture that our team is really just go and try stuff and just move as fast as you can to get the results. If it doesn’t work, doesn’t work and don’t be afraid to make a mistake. Hey, if there’s a few typos in the email, that’s okay. You got the results that is. So go out, try stuff and really learn by trial and error. Really the best way to do it. Read some stuff, get some ideas, but then you always have to apply it to your own business, your own unique situation. And that’s really the challenge in anything you’re doing. So go out, learn stuff, but then ultimately go back and try it. Yeah.
Bronson: Well, Archie, this has been an incredible interview. Thank you again so much for coming on Growth Hacker TV and for being so transparent. You didn’t hold anything back. You really told us what you’re doing and how it’s working and what you’re not doing. And it was just awesome advice. So thank you.
Archie: Appreciate it. It’s good to be here.