If anybody is interested, last year I wrote a detailed report [1] on Coursera's monetization journey. I describe the steps and experiments the company did to go from zero to $100+ million in revenues (and potentially over $200 million this year).
In my opinion, of all the online education providers Coursera grew the most during the pandemic. They also reacted well by offering free certificates and giving free access to their catalog to college students [2].
My company Class Central [3], a Tripadvisor for online education grew a lot during the pandemic. In the second half of March, we received 5 million visitors, almost a 20x increase. Though no longer the peak, we are still at 2.5X of what we were before.
Makes sense since B2B is an extension of their B2C catalog. But they seem to be hiring a lot of sales people, so I was not sure if they are profitable or not.
Not a huge fan of ads, but I would accept non-obtrusive ads in lieu of paying the certification fee if they wanted a traditional publisher revenue stream.
Yeah, the service is roughly $50/mo. There's no way ads could touch that. It might work in the future to monetize the older, "long-tail" content that is less relevant than it used to be. But even then, they'd probably have to cut the content, especially for cloud-related certifications, because A) the feature may have changed, B) I bet those labs get expensive.
The idea of watching ads in order to get an education seems like the start of some dystopia nightmare. Seeing kids in school buses emblazoned with McDonalds ads is bad enough.
Can you imagine wanting to check your answers to the work you just did and having to watch a commercial to do that. Probably reach a point where they'll require camera access to track your eyes to make sure you're paying attention.
It all depends on alternative. $100 is a lot of money for a kid in India, if the same kid can earn the degree while watching ads I think it does a great service to poor people because the alternative is not learning.
Alternative to a problematic suboptimal (according to you) school experience is not a perfect school experience but no-school experience.
American public schooling system has become a jobs program for adults.
Coursera has a financial aid program for this, and it is easy to avail if one is a student from developing countries or low income households. edX has a similar program as well, where the price will get reduced to around 1/10th of original cost. In Coursera, the price is completely discounted and you can do the course for free.
See: https://learner.coursera.help/hc/en-us/articles/209819033-Ap...
Spotify do regional pricing, the cost in India is about a tenth of what it costs in the United States. I don't know if Coursera do the same but I don't see why they couldn't.
Hello, I wanted to let you know that your HN account is probably banned as most of your comments are dead and the only reason I can reply to this one is that a mod undeleted it.
But if you’re going to argue that it does, that you’d send your kids to CourseraU, you are conflating the aesthetic experience of learning with actual learning.
Much like many video games, quite successful ones actually, reenact the aesthetics of work but are in no way shape or form actual work. This may be important for why they are an appealing product but don’t go and argue (as hardly anyone does) that League of Legends is work. That’s why it would also sound absurd to say Coursera is learning.
The University of Phoenix people are total shitbags. Charging for some shit certificate is a complete shitbag move. Just because it’s online doesn’t make it special.
In that sense, yes, expect that ads will happen. I think you’re onto something.
I had a really good experience with Coursera... In fact it's where I learned to code.
I don't think I understand the comparison to video games. Yes, sometimes games will simulate work, but that's not at all like online education simulating education.
It has its issues, but motivated students can and do learn online.
I like Coursera and I'm quite experienced. Their courses don't enforce the rigor of a college course; practically anyone could pass them without reviewing the content through sheer brute force. So I feel like the certificates themselves are worthless, but the actual content is good enough to learn from if you put forth the effort.
Huh...? Coursera provides video courses, readings, and quizzes. Is your assertion that it is impossible to learn from written material or videotaped lectures?
All the edx courses I've taken are basically text books in video form. Coursera has a very strong formula of short videos with frequent questions, often a question every two to five minutes to make sure you understand the material
According to a Google search of class-central about 4000 Coursera courses match the "may be unavailable" distinction.
These are generally any class that was niche and traditionally taught. I.e. a philosophy class that covers prolog.
Classes that are alive since ~2015 or earlier are usually remade every year or two to be more like a video introduction and are in a theme of either universal interest or that is mandatory for a particular degree (the first being good for bringing in new users and the second being good for certificate income.)
Does anyone know of a company that accepts any of their certificates as valuable? I am curious to see if they are actually making a difference in the employment markets.
I've taken a few courses on Coursera, including the deep learning specialization, and I thought the material was great. However, I noticed that in courses with ~5 hours of video, quizzes, and a project, the average completion times were measured in minutes. Based on this, I assume most people who received certificates cheated and didn't take the course, though I'm very much open to alternate hypotheses.
I heard that there is a 'trick' to getting a certificate for free ($) on Coursera: First you just review the course which is free of charge, and then after you're done with it all you start the free trial. Now you have 7 days to finish up the courses which shouldn't be a problem due to the reviewing.
So that could in part explain the short completion times, although I'm not sure how many people use this 'trick'. Coincidentally though, I've learnt about this trick in relation to the exact course you mentioned (deep learning.)
"If you complete a course during the free trial period, Coursera reserves the right to require you to pay for a one-month subscription in order to receive a Course and/or Specialization Certificate."
I took the DL specialization and it took me a while because I worked really diligently through every single thing and made sure I understood before moving on. It was purely for self knowledge and the certificate was super tangential to my aims. I noticed later that basically all notebooks were copied and out into GitHub and it would have been trivial to simply copy and paste all answers and get a certificate.
That made me pretty convinced a cert wouldnt be worth much, which is unfortunate because I really put in the work to learn the material. OTOH, I largely consider certificates to be a bogus signal, and even often negative, so that not really surprising.
This is actually a rather interesting question because of second- and third-order effects.
Most people will tell you that a Coursera certificate is useless for getting a job, yet the folks at interviewing.io found it was very strongly correlated with interview performance [1]:
"For people who attended top schools, completing Udacity or Coursera courses didn’t appear to matter. However, for people who did not, the effect was huge, so huge, in fact, that it dominated the board. Moreover, interviewees who attended top schools performed significantly worse than interviewees who had not attended top schools but HAD taken a Udacity or Coursera course"
Why are recruiters ignoring such a strong signal?
My guess is that most people recognize how easy it is to cheat a Coursera course, so it is generally not considered valuable by recruiters. Therefore most job-seekers do not place any value on a certificate, so they will not be bothered to try to cheat to get one. So the only people remaining who complete the course are those that are doing it because they enjoy the subject matter, which is clearly associated with passing the interview. There is therefore a perverse feedback loop where any recognition of the high value of the course leads to cheating which causes it to become worthless (this reminds me a little of Heisenberg's Uncertainty Principle).
My experience somewhat mirrors this. I've completed several Coursera courses on a variety of topics, but I don't advertise them on my LinkedIn or resume.
The reason I take the courses is because they are useful. If you put an effort into doing the labs, then you'll get something out of it. My approach is to try to write the solution to the problem based on the description, then compare with the solution provided to the lab. This ensure that I really learn the material from the labs. Every lab I've taken has provided enough time to get through it using this approach.
However, the labs offer the solution of the problems right in the description, and you get an infinite number of retries for the lab questions. So you can easily complete the entire course without doing any of the work.
So, the course material is designed to actually teach people who put in the effort, but the certificates themselves are a worthless signal of the amount of effort someone put into the class.
I've never asked those I'm hiring, nobody ever brought it up. Ages ago I asked recruiters if they thought it mattered, they only care if the company does. No investor I've spoken to thinks the certs matter either. I'm like 0/60 with my population.
It seems like it could be useful at the very end to show you a re committed to learning and staying upto date. Or diversing your skill set. In all honesty is probably value those certs ina candidate more than where they went to school a decade ago.
Lots of jobs have a four-year degree or equivalent as a hard requirement. You might not even have been asked to interview if you didn't have that on your CV.
Right, when I care, the requirements is just some 4yr, accredited uni. But, mostly I don't care about that either. Experience is my key metric. And for entry level (which I do a lot of) it's about a code profile and showing you can hack stuff together.
It's a bit chicken and an egg situation: if one has a degree, nobody ever asks about it except the very first job after graduation. So any course in addition to your normal degree will matter more because of what was learnt rather than the certificate at the end. Now for those without degrees taking coursera courses hoping they'll land a job,it may take more than just 2-3 courses, especially for the first job.
Also,it depends a lot on the market: certain countries are obsessed with certifications and will be very happy if candidates can provide them,so cultural context can't be dismissed completely.
At some level nobody cares anymore where or how you studied. They are interested in what you can do. As in "What you can do for us in $field" And they mostly look at experience of what you have already done.
Did you learn it from books? Coursera? MIT? Less interesting than you did it here and this other place with this kind of success given the projects respective constraints. And that can often be a personal project.
Anyone getting too excited about the signalling value of the name on your school is looking at the wrong thing and might be a bit garbage to work for as well.
I’ve done it as a new grad because I didn’t have much to put on there. I think it definitely adds value. It shows them you’re intrinsically motivated to improve your skills.
I, too, am curious about this. Not just Coursera but also LinkedIn Learning certificates, etc.
This isn't that I would avoid doing Coursera but it might mean I'd opt for other trainings offered. For example you can get AWS certified multiple ways, but the best way is prob through Amazon's AWS training.
A lot of companies value some sort of deliberate continuing education. I doubt that MOOC certificates are worth all that much--maybe with the exception of micro-master type of things from name institutions.
I think they will become even less valuable than they are now. If the cost to entry and completion is so low and people are speedrunning the material, the certification becomes worthless over time.
There's more to it than that. If you're applying for a junior position with no experience and a 4-year CS degree, you will very likely get an interview. If you apply for the same position with an online course or two, you're far less likely to get an interview.
Had a really terrible experience with Coursera lately. Signed up for the free trial, started the class and it recommended signing up for another class as a prerequisite. I signed up for that one as well but it required a bunch of materials I didn’t have on hand. Then I got busy and lost track of time. I never got an e-mail from them about starting the trial, nor did I realize the billing was $50/month for each class. Anyway I was checking my bank later and they were just silent charging me- $300 dollars so far.
They don’t offer refunds or even have customer support.
I remember taking Andrew Ng’s ML course several years ago and being so grateful and excited for the new possibilities, and now I just feel like Coursera is a scam, using dark patterns to make money instead of trying to make a great experience for students.
I’d do way better if I could take these courses with a group with people who could motivate me and hold me accountable. Especially if it was in person.
Is anyone working on something like that? When I do a course by myself it’s too easy to let other parts of life take priority.
I thought that was one of the features of Coursera. At least for some courses you join a cohort and all progress at the same pace: if you have questions you'll be asking them on the forums and getting answers from the same fellow students every week. Sometimes there's also peer grading where you review assignments for others.
Of course, they also try to make it so you can take the course at your own pace and for less popular courses there may be no one working through it at the same time as you. There's never going to be quite the same level of peer accountability/motivation as in an in-person class.
On one hand it's enormous, clearly suboptimal, and clearly in need of change.
OTOH, anything outside of institutional norms is relegated to extremely small niches... with self learning being the most vibrant. Very hard to innovate within the box. Very hard to access the market, in money terms, outside of it.
Your online education product can be 10X better than a college course. It can costs 10X less. That's still not going to mean you compete with the college course, even if the college course is mostly online.
previous EdTech founder here. can only agree. It is a weird business. I think it boils down to this: most students don't care about learning. They educate themselves for other reasons.
So if you make a good learning tool, it only will be liked by the small niche of high performance self learners / students.
The self learning market is completely disconnected from the formal parts of the market. There's very little substitution between them, in terms of money
My experience is that even community college classes are 100x more rigorous than online classes through services like edx. And that's saying something, given how easy community college is.
I graduated from one of lowest ranked CS programs in the country, and even the easiest CS course I've ever taken was orders of magnitude more difficult than any of the dozen or so online certification courses I've gone through.
Until places like Coursera step up their game and start offering something like "honors" courses, where you're likely to fail if you don't do all of the course work, then I don't think they will ever meaningfully replace college degrees.
> Your online education product can be 10X better than a college course. It can costs 10X less.
Or 1,000x less in cost is probably more accurate.
I don't know what the current rates are (I never went to college) but Google says it's about $50,000 to get a CS degree if you live in the US.
But an online course might be $50 or $100.
I've had lots of folks who took my build a SAAS App with Flask course say they learned more about web development in my 10 hour course than their 4 year degree. Of course you learn different things during a CS degree, but it's interesting at how big of a gap there is in time vs money. 4 years and $50k vs a few days or weeks and $50.
You're comparing apples and oranges there, in fact you say it straight out: "you learn different things during a CS degree". It's not a web development course, which is why there's such a big gap in time.
I could believe finding a well-produced collection of courses equivalent in material to a CS course for say $1000, but not $50. Not by a country mile.
To me Coursera has become worthless. It started by giving me a chance to learn about machine learning and some data science. I did 3 courses and was happy with the result. Then they had to screw up their login. It is now impossible to even log in for me, as it depends completely on Google Shittylytics. I simply refuse to use it. There is utter disregard for users built into the login, the very first thing you need.
What's more is, that without logging in, you apparently cannot deactivate the newsletter which keeps spamming me with their incessant "learning how to learn", which I really do not need. Not even sure it is legal to require login for unsubscribing to a newsletter instead of providing me a magic link.
> Not even sure it is legal to require login for unsubscribing to a newsletter instead of providing me a magic link.
I had this problem with italki and solved it by asking for the email of their General Counsel and informing them that they were breaking the CAN SPAM act, which depending on interpretation requires either one click or two click unsubscribe but definitely doesn’t allow them to force you to log in.
2.5 billion? This is just part of the investor bubble of these startups. I agree coursera is a high value company, but 2.5 billion seems a bit inflated.
Its revenue went from $60M(2016) to $170M(2019) [1]. In 2019, it reported over 45 million learners on its platform. It is adding enterprise customers at a steady pace [2].
A conservative estimate of 250 million revenue in 2020 puts it at 10x revenue. This is pretty normal for a SAAS company that is growing 50% y/y.
I'm extremely bullish on online learning. I owe my livelihood to things I learned online.
That said, after trying countless structured courses for a variety of subjects (ranging from design to music production), I've found that my best results are from coursed created by individuals, not large companies.
Individuals who deeply understand their core topic and are financially motivated to make the best possible course, I think, are the future of learning. Not "institutions" like Coursera.
It may sound nice in theory but never works in practice. I've been doing Salesforce administration/development for about 5 years. They almost invented badge system followed by certification. On their trailhead platform,there are so many of those that it could take a couple of years to complete. They also have about 25 different certificates, varying from junior admin all the way to architect level. There are so many people out there who have 10-15 certs without even seeing the platform properly because they meemorised mock tests and etc. That's why interviewers rarely rely on certification and they are usually no more than a box ticking exercise.
Edit:
Even though I'm sceptical about certificate value in recruitment process,I strongly recommend spending some time poking around Salesforce Trailhead website. They are pushing it very heavily, most collateral is very slick, a lot of PR explaining that people can go out and learn how to use it. Recently Microsoft started offering very similar platform by copying most of it. It's an excellent example how company attracts people to learn, as opposed to something like SAP,where you can't learn shit without having it at work.
I don't care if you followed a two month course by Joel Grus on Fizzbuzz and graduated magna cum laude. I care if you can actually code Fizzbuzz.
Coursera courses on your resume, is like listing your Microsoft Office skills. It would be noteworthy, if for some reason, you can't find your way around MS Office, or can't pick it up in a weekend.
I had a much better experience with online classes with EdX with no mandatory peer discussion because it allowed me to maintain illusions about my fellow students. The courses where I had to do some peer grading were demoralizing because of the obvious plagiarism and low quality, very low quality, of the original work.
Some courses have peer feedback, but I have found it lacking. In the course I'm doing, peers are expected to assess 2 other peers as a minimum, and they do things like this:
- just enter a "." or "-" to get pass the restriction of having to submit _something_
- a peer wouldn't give me full marks for an assignment, but then not leave feedback as to why, usually just a "good job!"
This frustrates me to bits. Coursera could invest in some cleverer technology here, e.g. if it sees a peer deducts points, then if sentiment analysis on the related feedback is only positive, fail validation on the form.
I've ranted a tiny bit about this on one of my course notes blogs:
This kind of thing drove me away from MOOCs for certain subjects. There was the Human Computer Interface course where your UI was graded by three peers. On multiple assignments I would get a strong positive with comments, a strong negative with comments contradicting the very things the other peer graded positive, and then a "good job". It was utterly useless feedback on things where feedback really was important for learning.
You can skip first step. Those certs are HR toilet paper. I used mine to placate my mum, she in turn rubbed Stanford/UM diplomas into our overly inquisitive aunts faces.
The "course" may be a facade, but intrinsic structure is helpful. If I'm simply learning by myself, I find myself skipping all over the place and then not completing the work due to being overwhelmed by the amount of material to cover.
Myself and many people I know bootstrapped technical PhD (earning potential in postdoc = $45k) into data science/ML jobs (100k+) using Coursera, textbooks etc.
Can you elaborate on what "bootstrapped technical PhD into data science/ML jobs" means? Because to me it sounds like you got a higher paying job because of your PhD, not because of the Coursera stuff.
I have actually learned useful stuff on Coursera that I use nearly daily in my current work. That's more than can be said of a lot of the university courses I've taken.
Over the course of about two years I worked on and off on the Andrew Ng ML course. I loved the site at first but was less enthused about all the incremental changes made to the site over time as they tried to realize a business model. Things like suggesting I pay for the free material in exchange for a useless cert, or verifying my identity via typing analysis so verify I was indeed myself etc
Can only comment on the first two.
Coursera: most courses come from universities.They are structured,aims are clear. Depending on the level and the teacher,some can be dull or/and going 1000 miles an hour,so picking the right level is important.
Udemy: less formal,less academic. Some courses are created by those who have no idea what they're talking about,while others are well prepared, broad,and deliver high value.
I wish I could comprehend what Coursera even does, other than handing out worthless certificates and paywalling content you can find online, for free.
I guess these investors are hoping Coursera becomes the Harvard of online education if they throw enough money at it. It may work, but good lord, what a strange way to decide to spend your limited life, hoping to make money (investors already have money...) from status symbols. Why not, you know, actually do something useful with your life?
This reminds me of people who bought bitcoin because, you know, other people will buy bitcoin... Why not, you know, actually do something useful with your life?
[Note: I'm enrolled in the MCS program at UIUC which makes extensive use of Coursera. Not a big fan of Coursera and most of the content for the courses is available through other avenues.]
They have two paths they can follow:
1) "Worthless" certificates
2) Platform to purvey certification for universities or other trusted credential organizations.
The certificates? They are what one makes of it. The problem of creating a usable, online education platform however is a hard problem and one worthy of investment. Academia is a tailspin trying to find an effective way to offer courses online.
I'll qualify my critique in that it depends on how the class leverages Coursera. I've taken five classes on Coursera two of which were through the UIUC program (procrastinating a type inferencer assignment as I write this).
There are a handful of things Coursera gets right:
Videos.Playback works. I generally download them all immediately regardless.
Progress tracking. Easy confirmation of what I've done already.
There are several problems it fails to solve:
Course Communication. There's a discussion form. It's not used. These are supplemented with other means of communication depending on the course that all need to be checked (email, Piazza, Slack, dedicated forum, etc.)
Live Events. Typically these are links to a Zoom instance with all the attendant difficulties.
Integration. Instructors often want to use outside tools for automated graders and the like. Coursera has tools to integrate authentication to make this first class, but it often means learning the idiosyncrasies for a new tool for every class. Generally, the integration is slightly broken in some way requiring some irritating workaround.
Quiz/Testing platform. Difficult for instructors to get right. This results in things like never seeing the correct answers to a final because view rules aren't well understood by the instructor or free-form answers being too picky.
Proctoring. Coursera doesn't have a solve for this. Remote proctoring is generally garbage and easily the most frustrating experience for me. UIUC uses ProctorU which I despise, but that's another rabbit hole.
Content Navigation. It's based on a week-by-week view. This can give a false impression of progress. This can make navigating between thematically similar content in week 4 and week 5 difficult.
In most instances, I would prefer an index page that mimics the the syllabus with links to the content and discussion forum.
As for the MCS it's been a good investment thus far. I've taken two classes, Data Mining and Programming Languages & Compilers. Like most things, you get out of it what you put into it. It would be possible for me to learn most of this on my own but the long term motivation wouldn't be there. This is especially helpful for formalism I struggle to learn on my own. Most of the classes have a project component which allows a significant amount of depth.
In my opinion, of all the online education providers Coursera grew the most during the pandemic. They also reacted well by offering free certificates and giving free access to their catalog to college students [2].
My company Class Central [3], a Tripadvisor for online education grew a lot during the pandemic. In the second half of March, we received 5 million visitors, almost a 20x increase. Though no longer the peak, we are still at 2.5X of what we were before.
1. https://www.classcentral.com/report/coursera-monetization-re...
2. https://www.classcentral.com/report/mooc-providers-response-...
3. https://www.classcentral.com/