The $2 Trillion AI Panic: Is SaaS Really Dead?

Matt: [00:00:00] Two trillion dollars gone. That's how much value software stocks have lost this year because Wall Street is convinced that AI just killed SaaS. If you don't know what SaaS is, it's software as a service. It's paying for subscriptions for software, such as the Microsoft 365 suite and things of that nature.

are everywhere. One prompt and Claude now spits out a working Notion clone or they have like authentication all set up. They'll spin up some sort of financial software. The demos are everywhere. So that's it, right? to cancel the subscription, fire up all the agents, get your codexes and your Claudes and everything going.

SaaS is dead, except the companies that are supposedly dying are actually still growing. lost a quarter of its value this year, far, while signing a $5.5 billion government contract. So something certainly does not add up. So here's the thing. We wanna talk about how posting [00:01:00] these demos is not an actual product.

You can't just go to Claude and be like, "Make me everything that Notion is and ever was," and boom, there it is. It feels more like a demo. It is not an act-- Not the, the real Notion, the thing that's been QA'd and constructed and updated over years and years and years, and there's many other examples other than Notion and many other examples other than Salesforce as well.

Potentially, the scariest chart of 2026 is measuring fear and not revenue. So this is the $2 trillion panic over AI-created clones of popular SaaS products. Mike, you actually are the one that wrote up this episode, and I'm gonna pass it to you shortly. But if this sounds interest to you, interesting to you and you wanna support the show, you can go and check us out on that Patreon, leave a review or rating on your podcast app.

Join us in our Discord server or share this with your friends. And if you wanna help out not only yourself, but us as well, you can net yourself up to 20% off a Scrimba Pro plan, and you can find the link to do that in the show description and in the show notes with full [00:02:00] details on how it works in the show notes on htmlallthethings.com.

So Mike, like I said, you've wrote up this episode. You've been doing sort of the research on this, um, and we have a, quite a, a stacked show note. We're not gonna be going through everything in the show note, but it-- I will be including it in the show notes on htmlallthethings.com if you want to see some quotes and some, some, some links to some sources, other things like that.

Um, but on the screen right now, if you're watching the video version, we have the chart, uh, for year to date for Salesforce Incorporated. Uh, and, uh, it's a, a bit of a downward trajectory in general. So Mike, why don't you take it away?

Mikhail: Yeah, for sure. So this kinda came up on my radar probably a few months ago now, like three to six months even, uh, where this started to actually be talked about. And for the most part, it was a good conversation. Uh, I never really gave any to it because I thought it would recover. But it's been such a long time now, like you can see the year-to-date chart that we're seeing here, that I think this is a [00:03:00] conversation worth having at this point where what, what's going on?

Like, why are these stocks plummeting when there's fundamental reasons for believing that, like, their businesses are in their best era, essentially, in terms of like profits and revenues? And what is the public sentiment like right now in terms of, for, for SaaS companies? that's where I wanna kind of start with is like, what's the negative?

Like, what, what is the panic that's happening? And then we'll Matt and I will probably argue about it a little bit as we go through, and we'll talk about why a lot of this is going to be a little bit overblown in my opinion. But the problem is, is that our version of overblown is like a logical overblown, like why we think this, this shouldn't happen, why we think that these stocks shouldn't be plummeting. That does not control the market. And at the end of the day, what goes into the investors' pockets is probably the thing that's king for these companies and the [00:04:00] thing that they care about the most. So when their stock is going down, even though they're profiting, you know, they're recording record profits, I think that changes an industry as we see it. Like, it, it's a weird thing to say, like you're, you're making a lot of money, but you still have to pivot to appease investors so that your stock can go up so that the investors can make more money, the board can make more money, the CEOs can make more money. Such a weird place that we're in. Matt and I were talking about this right before, right before the show, and like, I can't wrap my head fully around how the markets work, and I don't think anyone can, including the s- like, you know, the mar- the, the expert of, in the markets. And because of that, it's becoming kind of a disaster for companies specifically that are trying to find their place or trying to pivot to, to the right spot the points when they're... They shouldn't be pivoting. Like, they're making so much money. Why would they pivot? But they, they're, they're... It seems like they're being forced

Matt: Well, the long and [00:05:00] short of it is that, software, like Salesforce, is a difficult thing to make, it is a difficult thing to market, it is a difficult thing to... It's a difficult thing to maintain, and it also provides a lot of value. So it has that sort of formula like, they've obviously got marketing down.

They've obviously made the product work. They have an install base, they have customers, so they have some sort of trust. personally use Salesforce, but I've heard-- I hear about it all the time and have heard about it for years. Obviously, it's a large company at this point where, you know, they do have a ticker symbol.

They do have, uh, they are a public company, right? You can purchase their, their shares in that. it's all based upon the fact that what they're doing is difficult and it, and it provides value. And I think what the panic is, is that AI has just created demos or clones or cloned demos of what some of these pieces of software [00:06:00] can do with just a few prompts or maybe a prompt or, you know, whatever.

we have an issue now where people are like, "Oh my God, it, it's now no longer difficult to do this." And so a major sort of keystone in that formula is, is Salesforce still valuable? Well, yeah. is it difficult to make? Maybe the barrier to entry is easier now. And of course, everyone's gonna start panicking.

But I mean, software, especially when it's very specific software, when it's very advanced software like Salesforce, like, uh, you and I were talking, Mike, about QuickBooks Something like, you know, Mike, Mike and I are, use our QuickBooks at a very surface level. You know, our accountant goes in there and does all the taxes and checks all the books and does all the whatever they do.

But they tell, they, they tell us, "Hey, go in here and make sure you fill in these numbers and make sure you attach receipts and make sure you have the dates right," and they give us sort of like a, you know, a long and short really of what we need to do, and then we go in and we do [00:07:00] that, and that's all we do.

But in the background, there's reports being generated, there's things that are being, uh, you know, you know, created, uh, there's, uh, tax rates, there's laws that are being adhered to, um, or the software's preparing to do, to do that. I mean, even crypto tax software, obviously, you know, you, you... if you're in Canada, we don't have the IRS, we have the CRA.

You look for, you know, sort of CRA certified or, you know, s- Canada-friendly income tax when it comes to crypto software because it's difficult to sort of have those software mechanisms that adhere to the law. You have to have someone that understands the, the legal law stuff. You then have s- have to have someone that translates that into technical ability, which is the software having the technical ability.

And I mean, prompting that into existence when there's so much abstracted away from us, like in, in Mike and my situation, the QuickBooks, like QuickBooks itself is, is highly advanced. It has... It's highly advanced because it's doing a bunch of math and things, but it's also highly advanced [00:08:00] because it's allowing an accountant to log in for us.

It's, it itself has a bunch of guardrails and stuff that's set up by the, you know, the jurisdiction that you're in and where you are. And in Canada, you have provincial and federal taxes, so depending where you are in the country, uh, the taxes are different, which I assume is the same for the States and the various states that you're in.

And, so like the, the amount of prompting, if you just think about it, the amount of prompting to get to that point would be, I mean, a lot. And then at the end of the day, you wouldn't all receive necessarily a standardized output, right? Where, where like maybe I give the... I prompt into existence, you know, a free QuickBooks as an example.

I prompt that into existence, and then I give that to our accountant. W- what if that's not the software they're used to? What if... Like, what are they gonna do? Each of their account, e- each of their clients, they're gonna learn how to use e- these little individual SaaS not SaaS things that people have [00:09:00] created?

What about next year when the laws change?

Mikhail: I, I think, yes. I, I, like we talked about this before, I agree with you that there's some software that's really difficult to replicate, and an internal team won't just, like, go out and replicate QuickBooks most likely 'cause of everything that you just said. I do, what I do see happening is with AI, with agents, as they get better and better The moat becomes smaller and smaller.

So even if you have very complicated integrated software with very complex like taxing and, you know, taxing systems and reporting systems, the reality is, is that the better the agents get, the better they're gonna get at like figuring that stuff out, the integrations, including the relationships potentially even that re- are required, so that it'll be able to reach out to the right agent. Maybe all of these like banks are gonna have their own agent platforms that agents can talk to, right? Like who knows? And the investors that are investing [00:10:00] in companies are seeing the future of like, hey, maybe that is the case. Maybe like Quick- a QuickBooks competitor like three years ago would take years to be developed and then years of relationships to form, and so the moat was really, really heavy and really, really big. And now the moat's getting so small that, you know, a QuickBook- competitor can be built with a small team in like a year or six months, then just a bunch of agents, right? That could actually come in and, and compete with Qu- QuickBooks on s- on like, you know, functionality and be more UI/UX friendly because it's like, it's a greenfield project, et cetera. So like th- there is some argument there that I could see being made as they get better, and that's something that I wanna talk about is like I've changed my screen right now if you're looking at the... if you're watching the show. And what, what I'm showing here is the demos that people talk about, quote unquote "demos", um, which is a Notion clone.

So, uh, Matter, uh, who works at Cognition right now, who-- the, the, the creators of Devin AI, and they do a bunch of other stuff right now. I [00:11:00] think Windsurf was the-- uh, they bought Windsurf. But regardless, um, he's testing out the new model that came out yesterday, which is, uh, Fable, the Mythos-level model that Anthropic released. it-- he asked it to just build a, uh, a Notion clone, So if, if you see the screen right now, it's sh- he's showing a, a bunch of different functionality of Notion, and it's looking pretty good. Like we don't have the-- again, this is a demo. That's why I say demo, because we don't have the cape- like we're not looking at this ourselves.

We're not trying it ourselves. But from like high-level perspective, this looks like Notion. it did it in 30 minutes, what is Notion's moat, realistically? They're-- the No- the moat does exist, let me be clear, and I'll-- I'm gonna talk about that. Like the No- the moat does exist, but it's becoming smaller. And if, if a company like, let's say Cognition, which is a pretty big AI company right now, is like, "Hey, we built this. We have a few engineers idle because they're doing other AI [00:12:00] stuff. They're waiting for AI stuff. Why don't we just put this into like an internal product and ca- cancel our Notion subscription?" That's the story that's being told, right? Like there's this story of like- If you look at a company's bill, their bill, like the, the amount of money that they're paying subscriptions is really high.

Matt: Of course

Mikhail: if you look at all their AI coding tools, like that's a subscription model. If you look at all of the other like accounting tools and productivity tools that they're paying for, it's very high.

Like the, these, these companies charge a lot of money to use their products, and they charge per, on a per seat basis as well. And so there c- there will come a point when the agent good enough that maybe it makes sense to start looking at that and start comparing it to an engineer's cost and time plus an agent's cost and time and be like, maybe it does make sense to spend some engineering effort with an, with, with a- agents build alternatives to these like lower hanging fruits. Not to everything obviously, but to these lower hanging fruits. Like Notion is, [00:13:00] is a prime example. So like that, that's the steel man argument. Like that's one of the biggest steel man arguments is as the capabilities jump, the n- the requirements and the, the, the ability to create these lower hanging fruits SaaSes starts to become easier and easier and cheaper and cheaper, and companies might start going in that direction. And so investors see this, they s- get scared, and uh, yeah, they're-- Like if you, if you've ever watched Shark Tank, a lot of questions are a- on, on Shark Tank are asked like what's your moat? stopping, you know, Joe down the street from doing what you're doing? And a lot of times it's like expertise, whatever, like super intricate expertise.

Patents are a big one that people talk about on Shark Tank. You know, like, uh, and then, and then it's like, oh, the engineering effort required to build this. Well, one of those moats is almost gone. Like engineering requirement is almost gone. And it's never been a very good moat, let me be clear about that.

Like engineer, like if you, if you, if you would've showed up on Shark Tank [00:14:00] and been like, "You know, I, I, you know, I'm a, I'm a developer. There's not many developers out there, and I'm the only one that can build this," they're gonna laugh at you. It's like it's not just Shark Tank, it's investors in general.

They'll be like, "Well, you know, get an Apple engineer in here. I'll, I'll pay him $200,000 or $300,000 and I'll have him build this thing in, you know, six months to a year." Right? Like this is before AI. it's, it's kind of the same argument, but like now it's like, "Okay, I'll get him in here and I'll, he'll build it in a week." That's, that's the kind of conversations that the VCs would have, like the, that the VCs would come back with.

Matt: it was a, it was a moat. The engineering moat was a good moat against the general public, though. We're, we're starting to have these, these AI tools be used by complete amateurs. People who are not in tech, they are not coding, they've never touched code. They don't know what code is really. And so ultimately one of the value adds a company has, is that they're doing something difficult that the c- that, uh, the, the public, especially if [00:15:00] it's a publicly available product, that the public would have difficulty doing.

Obviously, your competitors, like talking about competing in the market, the, with B2B, you know, business to business, you're kind of having those arguments, those fights, and those, you know, the competition, I mean, straight up between businesses. a different story 'cause yeah, one, that's not a great moat against your competitors, but it is a moat for demand, potentially

Mikhail: It, it is a moat, but like, it, it's difficult. I think there is some pricing in of what you're saying, Matt, of in like the general public being able to build these like little micro apps using, you know, Google has presented it as like a, a new search interface where like they type in something and it'll build them a little micro UI for that specific thing that they might have paid for a SaaS before and they won't now. I-- that's like, in my opinion, the least concerning thing. I think the biggest concerning thing is giving this tool to engineers makes it so that they can build their competitors' [00:16:00] products 10 times faster now,

And therefore, the moat of like time to build has gone down quite a bit, ability to build and the cost to build has gone down quite a bit to a certain degree.

Matt: But we, we saw this, we saw the same thing though with, with website builders, where if we, if we zoom out of the developer space for a second and we look at who's consuming the, the, the websites, it's people who have businesses, small businesses, large businesses, and everything in between. It's people who have businesses.

And websites used to be, you know, the, the, the biggest businesses would have the fanciest websites, and then website builders came in and really attached themselves to design trends and parallax and sliders and animations and fancy text and fading things in and out and transitions and all these things, and it democratized the skill not only to engineers, where, like, engineers had to do it, and it, it, it made it so that the everyman, the every person could just [00:17:00] go in, sign up for one of these services, and create a really nice-looking And, like, we had a conversation recently with how, uh, on a, an episode about how long websites last and There certainly is still a difference between, you know, the t- a top-of-the-line, you know, kinda cutting-edge marketing website versus someone who, you know, is just trying to make sure that people can book tee times at the local golf course.

There's certainly a disparity there. But the disparity is not huge. It's not like one person's using plain text and the other person's using fancy images and videos. It's like they all have fancy text, they all have transitions, they all have videos. It's just one may be, you know, updated every two years, whereas the other guy's gonna keep his around for five to 10, 'cause it doesn't-- it's a different scale of business.

But my point is, is that it's, like, been democratized, like, the skill's been democratized. However, web developers remained. On the back end of all that, you still have web developers. I mean, we still ship small business websites. [00:18:00] We still ship, we still ship small business websites all the time. I help people with small, small t- I-- my, my primary thing is small to medium businesses.

It's why I'm not super into all of the hyper-modern coding stuff. It's because they don't use that stuff. They use the WordPresses, they use the Webflows of the world, then I enhance it with code 'cause I, I... But I'm not just brought in for that part. I enhance those platforms with a bit of custom code to ma- to push those platforms further than they can out of the box, right?

However, it's not like the client is coming to me with a complete, an almost completed thing, and they go, "Matt, just do the little, you know, the little bits and bobs that you add." come to me and say, "Do the whole thing. I don't wanna learn how to do this. I don't wanna learn how to add a, add a slider.

The heck with it. I don't wanna, I don't wanna touch it." So it i- it, it is like a, it is still, like, a bit of a moat, and I think it, it, and think it adds credence to the fact that there-- this is an immediate freak-out because th-the things are shifting. Website builders shifted things. [00:19:00] Website builders absolutely shifted things.

But it wasn't completely destructive. It was not an apocalypse on web development and web design. It was not an apocalypse. Templates were not an apocalypse. being readily available for free or for, for relatively affordable, it was not an apocalypse. I don't think that this is an apocalypse either.

It could be. There are things that become apocalyptic to industries. Like the horse industry is g- was, cars destroyed them. I mean, like, how many people are ripping around on horses compared to what they were back in the Old West? I mean

Mikhail: So my question to you, Matt, and I think this relates exactly to what investors are doing or like investment bots or whatever and wh- why we're seeing the sell-off, right? My question to you, let's say you had a friend who has a little brother and he comes to you and he's like, "Hey," you know, he's right before university, he's like, "Hey, should I get into web development?

Should I do what you do?" Right? Like, should I-- is that [00:20:00] something that I should invest my time into? Should I go into school for like software engineering or web dev- whatever, like should I go to college for s- web development, whatever. What is your answer to them?

Matt: Well, my immediate instinct is that the, the industry is going through a major change. Like, h- h- it almost sounds hypocritical from what I just was saying, where I was trying to, like, calm the alarm bells.

Mikhail: Mm-hmm.

Matt: are alarm bells going off in my head, of course. There's major shifts that are happening. There's major changes.

There's lots of promises of, "Oh my God, we're gonna be... You know, we're gonna-- Our code, our, our next coding model's gonna replace everybody," and, you know, all this stuff, there's a lot of panic there. However, if I try to look at it more objectively, is it absurd to say that there's never gonna be a tech worker in 30 years?

And the reason why I say tech worker and not developer is I don't know if developer is gonna be around, that role, but people that are in tech can bounce around. People that are in tech are able to learn adjacent skills and are able to [00:21:00] maneuver around tech quite a bit. I mean, we were never taught IT security in college, and one of our college mates is, uh, a, an IT s- an IT security tech.

And so obviously, you know, he was able to learn new skills somehow and then, know, move on to that, and as far as I know, and did not go through any more sort of post-secondary, you know, college or, or university training. And so I, uh, and so I, I, I would honestly just say that. I would say that if it... Like, if you're really into just the coding part, maybe not.

But if you're willing to learn the coding today, and you're willing to expand your own skill set through adjacent jobs or your own research or your own side projects or whatever, and those don't have to be coding side projects, by the way, but if you're willing to expand your technical skill set, do I think tech workers are gonna be around in 30 years?

Yeah

Mikhail: I mean, I, that's a good [00:22:00] answer, I would say. Like, that's a good, that's a good indicator. I think it, it's, it's really difficult right now to, like, recommend a path because of exactly what you're saying. Like, it's just so up in the air. I don't know what it's gonna look like.

Matt: Right

Mikhail: gonna be a requirement for people, like, I think. Um, I, I say that with a caveat, obviously. But like, if someone came up to me today and they're like, "Should I get into web development?" Or, "I should get, I, should I get into coding?" Uh, I'd have a very tough time saying yes. Um, but I would also have a very tough time recommending something else. I think that's where I'm stuck, is like, what else is there? Okay, go down, like what, what if they don't wanna do, like, physical labor? Right? Like, is there really any other job in the spa- in the non-physical labor space that's, quote-unquote, not going through a revolution? It's

w-

Matt: when something becomes this displacing though, you know, I'm not one to say, "Oh my God, the government's gonna come and save us." However, the government's gonna take notice of this [00:23:00] one because they do- they, they are gonna realize, "Oh, if I help the people that are potentially gonna be displaced, I'm gonna get votes."

Whatever side that is, left, right, up, down, center, I really don't care. The point, the point of the matter is, is that they're gonna take notice of, "We better do something," and/or they're gonna say, "This is a great marketing opportunity for us. We're gonna be the party," whatever side it is, "We're gonna be the party that helps the people that are worried about being displaced, and then we're gonna get all their votes."

And so, uh, like I think that there would actually, like if something is this big, big as they keep saying, there would be like some sort of intervention which would change the trajectory of the market if it were just run by the private sector, if that makes sense. Like for example,

Mikhail: Ugh.

Matt: if, if, if there, if literally every industry, like you're saying, 'cause even heavy industry, like people keep saying this labor stuff.

I think that's all nonsense, personally, and the [00:24:00] reason why I think that's nonsense is 'cause we have robots. We've had robotics and advanced robotics for years. The missing piece to the robotics is the intelligence. Now we're starting to get the intelligence. Are you telling me that, "Okay, Matt, like you learned to make these, you know, complex IT systems and everything else and, you know, all these things.

Cool. You don't need to do that anymore. Go lift that box and move it from point A to point B, and that's all you do all day, realistically. That's all you do all day, hours a day." You, you don't think that they're gonna try to get a robot in there to replace that, to replace me picking up a box from point A to point B?

I think they would

Mikhail: Yeah, w-without intervention, like you're saying, I think that the only, the only reason I say that l- like the, the, the labor is the maybe safer path, quote unquote, is I think it's just a little bit further out. It's a little bit harder to repla- It's a little bit harder to make a general purpose robot. Let me be clear about that. It's not hard to make, or not, not, not hard, but like we already have [00:25:00] very specific robotics, like the robotic arms that can do certain things. But a general purpose robot that can, like you said, learn and, and be intelligent and do the regular labor that is required of a human We're doing it right now.

Uh, there's m- like probably hundreds of companies now in that space.

Matt: Right

Mikhail: I still, like from my, the demos that I'm seeing like CES and, uh, Computex and all that, it still seems like a pretty far leap. Like a lot of companies, for example, the, the way that they're doing it right now is like a lot of companies will hire humans to control the robots to gather the training data required to then replace them.

So like we're still in the data gathering phase, and think we're still a ways away from the robots being like agile and strong enough and general purpose enough and intelligent enough to really start replacing the general labor as a whole. So it's not a, not to say that this is a full-time solution, but like, you know, if you're [00:26:00] choosing a more safe place, maybe like for the next, you know, five to six years or five to 10 years, it's safer in the general labor. Um,

Matt: But I will say this, the factories today are designed with machines working alongside humans.

Mikhail: Yes,

Matt: we may completely radicalize how the factories go. And obviously, usually the answer to things is not one side or the other. It's usually somewhere in between. So, I mean, even the gigafactories of Tesla, they were supposed to be virtually all unmanned.

And then they found out, well, it's a real pain to like get, you know, a robot to line up a bolt or to do a slight adjustment because humans can think on the fly and robots are like, check all the parameters. It looks like it's okay. Why isn't it working? You know, and then it doesn't really have a way out of that thought process.

again, like the gigafactory is kind of a gen one product in a way it's a gen one of a, hey, let's try to make this entire factory automated. [00:27:00] And it's not like they were like, well, all right, back to the drawing board and do it again. It's like, no, it's too much money. They want to make Teslas. They have to make Teslas, you know, the cars.

And so away they go, right? They do their thing. I, I think like when, when I talk about government intervention, what I mean is, is that if it was, if it's as gonna be as apocalyptic in terms of a job market as... If it's gonna be as, as, as apocalyptic as the people on the very cutting edge think. When, when people th- when people say things like every industry's gonna be completely revolutionized and we're not gonna need people, and it's gonna be automated.

Yeah, but that doesn't really work though, because people are the ones that are the, the ones that need, want, and purchase products. People are also the ones that own houses and pay mortgages and pay for roads to be created and all the rest of it. That, th- like the instant that a large portion, like let's say 50%, if, if 50% of the population within a five-year period becomes unemployed, I don't see the government just sort of [00:28:00] idly standing by being like, "Well, I guess we're gonna get crushed."

They're like something, something's gonna give. Whether there's gonna be some sort of tax on if you use this much technology you have to pay back into a fund, and then that goes into a mincome. And I'm not saying I agree or disagree with any of this. All I'm saying is, is that if you, if you make it so that like a, like a government, especially like a democratically elected government bad, they're scared of getting voted out.

Mikhail: Mm-hmm.

Matt: so then they're going to back against the thing that potentially is gonna get them voted out, uh, like every time. We're seeing it in Canada with like AI data centers being kiboshed where, you know, oh, this is a po- potential AI data center. Protest opens up and then boom, you know, we don't have it anymore.

Mikhail: Mm-hmm.

Matt: But I'm not, but here's the thing. I think that if it completely guts tech, I don't think you'll see much government intervention 'cause it's just tech. It's not tech and agriculture and, and, and, and the [00:29:00] trades and everything you're saying. then, and then you're, you're saying, you know, maybe it's safer to go into like labor.

I don't necessarily think that's, I don't necessarily disagree with that. If it's going to go the way that they keep saying it is, and that's a big, big, big if, where like AI is gonna in- inject itself everywhere and add, add automation everywhere. If it does go that way, if you hop around jobs, you're just delaying the inevitable.

And the inevitable, you know, may not be 50 years away where you're like, "I'm gonna, I'm gonna very quickly, you know, work for 30 to 50 years and then try to retire." It's like, it's probably coming for you faster than that

Mikhail: Yeah, it's c- it, it's an extraordinarily complicated topic, um, that we have tackled a few times on the podcast at this point, like we try or at least tried to tackle. Like I-- It's to the point where, like, we can't worry about it on a day-to-day basis. If we do, then we're just gonna be analysis paralysis, and we're not gonna get anywhere, anything done, right?

Like the real [00:30:00] advice here is you gotta learn the tools that are available to you today be as efficient as possible with them, and that's the, been the advice, let me be, like before AI. You know, when IDEs became smarter, there were people on the other side of the IDE argument being like, "I'm using Notepad and I'm never switching from Notepad." And then there were people who were like, "Well, I'm just gonna use VS Code because it's faster," and m- a lot of times that benefited them, um, because then in the, in the interview that they had coming up next time, they were asked a bunch of VS Code questions, and they could answer them, and they might have, you know, seemed better.

So able to use the tools that are available to you is really the only we can do. I think the, the hardest part for me right now is like, you know, giving advice to the younger generation 'cause I have nothing. Like, I have nothing I can tell them that, uh, makes any sense. Like other than like, "Hey, learn the tools as well, and be as efficient as you possibly can with those AI tools without, you know, making yourself dumb." [00:31:00] Um, but it's, a difficult-- It's a really difficult conversation for the newest generation. For us, we just gotta, you know, day by day, and it should be okay to a certain degree. I, I do think that before the government really interjects, there's gonna be some pain. Like, there has to be some pain that causes the public to start voting in different ways, obviously.

So that, that's what I'm most worried about, and that's what I'm trying to mitigate against the most internally is like how do I, how do I, how am I not part of that pain? know what I mean? Like how... Yeah.

Matt: Well, of, of course. I mean, that's what we all, that's what we all try to do. I mean, going into tech, tech was always like this miraculous land of, oh my God, we're always gonna need tech, and we're always gonna have tech, and we're always gonna have developers, and we're always gonna have and we're always gonna have...

And like, I mean, that, that's still the case. Like, that still could be the case. Predicting, predicting the future is kind of a fool's errand if you're, like, actually trying to lock down certain things. It is a bit of a fool's errand 'cause you don't know what's gonna happen[00:32:00]

Mikhail: Yeah. And that's why I wouldn't make like a, a radical change out, out of tech. Like, I wouldn't just all of a sudden be like, "You know what? I'm gonna be a car mechanic," uh, and just like go be a car mechanic today, because it just doesn't... Like for us specifically, it doesn't make sense. Like we've, we have a lot of expertise in this, and I'm willing to adapt to it, so that's where I'm, that's where I'm gonna invest my time and effort and hope that that pays off.

Matt: Although that path isn't that crazy, 'cause there are, there are people that are getting fed up with tech. Like, I mean, even myself, like I, I notice the cracks where I'm just trying to roll my eyes at a lot of tech articles and things and, and it's less, um, optimistic, and it's less like, "Oh my God, this is gonna be amazing in the future."

It's more, "Oh, okay. Well, you're gonna take my job, and then this is gonna go downhill, and then that's gonna be a problem." And whether, now whether that's the millennial pessimism, you know, coursing through my veins, I don't know. But do find myself rolling my eyes more than, you know, kind of jumping for joy and cheering when, when something gets [00:33:00] announced.

And, and I mean, like back to the episode with like the whole like SaaS, like the whole like SaaSpocalypse thing, it's like how many, you know, thousands and thousands of not only developers, but also marketing personnel and office administrators, and I mean, even the executives really, you know. I, I know people don't, don't like executives, but even the executives, I- Like how many, you know, hundreds of thousands of, of, of workers across all these different SaaS companies are employed because of different SaaS opportunities?

I mean, we mentioned QuickBooks, but like Intuit has a bunch of different offerings. Intuit's the sort of main company, I suppose, um, of QuickBooks. It's, so QuickBooks is one of their pieces of software. And, and many companies are like that. I mean, there's full, there's full ecosystems within ecosystems.

WordPress is a thing, and so Automattic is a thing, which is tied to the creator of WordPress. But then you also have, uh, the company, um, like which immediately,

uh, w- left my head, the one that makes ACF and all that. Uh, WP something. Uh, WordPress engine, I think. [00:34:00] WP Engine, I believe. they, like they have a bunch of offerings, different plugins, ACF being one of them, and, and like there's ecosystems in ecosystems, and they're SaaS-based. But I will say this, is that you could, you could say, and I hear this all the time, people are sick of subscriptions.

So you could say, sure, that AI is gonna be this big thing that, that, you know, kills SaaS and this and that, but you could also argue that the writing was already on the wall for SaaS. the reason why that is is not because the entire industry is gonna be torn out and spit out and thrown in the garbage.

could say it's because people can't afford, this is another $10, that's another five, that's another 50, there's another 30, that's 100, that's 120, this one's 10. f- go- being someone who works with WordPress and being a person that goes okay, like I'm gonna, you know, use some popular plugins on my customer's websites.

Oh, I hit a small limitation. Go look. That's an extra 20 a month. Oh, I hit another small limitation. That's an [00:35:00] extra 20 a month. Hit another small limitation. That's an extra 30 a month plus another 10 'cause I have to get a plugin of a to do this. And then people could say, "Well, you could just code it up yourself."

You're right, and sometimes I do do that. I absolutely do. However, still a thing where you could see the nickel and diming really apparent in something like WordPress, and that's not WordPress's fault. That's the nature of the industry. But just let's say zoom out and go to like a regular person that's not in tech, like just a, a person that's sort of living life.

How many SaaS are they a part of? Netflix They're a part of, like, you know, Paramount Plus. There's, there's subscriptions for f- for cars to have, like, auto-start and stuff in your car and stuff like that, especially if it's a luxury vehicle. You have, have a bunch of stuff like that. They might have accounting software that they use, so that's a SaaS.

They'll... They might do work at home, so they might need the Microsoft 365 suite. That's a SaaS. The list just keeps going [00:36:00] on. Every single time they need a piece of software, it's another, it's another subscription, it's another subscription, another subscription. I mean, e- even, even things like, "Oh, I'm gonna u- you know, use Amazon."

Oh, Amazon Prime, there's another subscription. I don't want Prime Video with ads." Well, that's another subscription. Like, the list... You know what I mean? Like, the list just keeps, keeps going on and on, and you can see how things can easily and quickly spiral out of control when it comes to SaaS, and you can see the fatigue of SaaS in the market.

Like, instead... Like, I- I'd honestly be enticed if it was like, "Oh, I'm gonna, you know, alleviate my, uh, you know, one of the limitations that I'm having, one of the issues I'm having with my WordPress site. a little plugin that allows it, allows there to be a toggle switch. It's $10 once." Okay. And I might even just eat that cost and, and write it off and not pass it on to the client.

But every single time I'm gonna do that, do that, it's like, oh, that's $10 a month, that's $20 a month, $50, $30. And being Canadian, it's even worse, 'cause oftentimes those prices are in American. So now you [00:37:00] have, like, this whole other thing, you know? And, and I understand that WordPress is open source, and I know that there's alternatives out there, and I'm thankful for the open source, to the open source community.

the very fact that I could even have this sort of, like, little rant, m- you could argue that SaaS already had a bit of the writing on the wall due to fatigue, and that this is just sort of the guy that came in with the hammer and said, "All right," and the wall was already cracked, and then boom, there, there it goes.

Mikhail: And, and that's absolutely the case, by the way. Like, uh, apparently as of, uh, 2021, uh, growth has started to decline in SaaS, and that's, again, 2021's before AI even came into the market.

Matt: Right

Mikhail: So SaaS fatigue was already a thing. the problem is, is that what you're saying, where like these SaaS companies have hundreds of, thou- or hundreds of thousands of employees, uh, to support this, this model, that those two things can't be true at the same time.

So we're see- seeing SaaS start to fall and people [00:38:00] are tired of paying monthly subscriptions and, but they're still willing to pay like, you know, a year- not a, maybe a yearly, but like a, a one-time subscription fee, you would have to remove a lot of employees for that to start making sense.

Matt: Correct.

Mikhail: most of these businesses have infrastructure costs.

The reason that SaaS even came about was because like, hey, like we have these servers, $10 one time, giving you access to those servers is impossible from a like cost perspective. So we need a constant maintenance thing. We c- we need constant maintenance costs to be able to support our thousands upon thousands of people to come in and start using this.

So like would have to rethink the software model or like revert the software model back to where, what it was in the like early Windows days, where it was more local software that you got, you purchased and it wasn't updated. It was updated once a year and you had to purchase a license once a year if you wanted those updates.

But it didn't, it also didn't rely on a massive server infrastructure, massive amount of like [00:39:00] s- like, you know, Amazon AWS instances and stuff like that, that needed constant upkeep. So you have to like almost go back to first principles with software Go back to first principles with company size as well.

So maybe like instead of having, you know, 10,000-person company, go back to having like 100-person company, having more and more of these micro companies that are self-sustainable but not gener- not created for infinite growth. So that might be something that happens, and like we've talked about this on the show too, where like, you know, the democratization of coding might develop more comp- competition in spaces that lacked it, and therefore it'll create smaller companies that can compete with larger companies, and therefore lower costs across the board, and maybe potentially even create jobs in that space. Um, that's all gonna take time, obviously. Like, we're not gonna see that happen overnight. But you're right in the sense of like, hey, SaaS was already becoming exhausting. I'm already exhausted having to pay for subscriptions. I can't... Like, I can't justify [00:40:00] paying for any more subscriptions anymore. Like, I'm, I'm removing subscriptions. From a business perspective as well, the other thing that's really, a real big pain in the ass is like seat cost. Like, if you're looking at something like a HubSpot or Vercel or all these other companies that are like, "Hey, this is our core," our core product is like hosting services, okay? That's, that's Vercel's core product. And the way they make money is obviously on the hosting part, but they also make money on the fact that every developer that you put into Vercel, you have to pay an extra $20, $30 a month for their seat. you're like double charging them, and the, the seat doesn't make any sense. You already have the hosting, and so like we're getting exhausted from having to get double charged on everything. And we're seeing actually happened, like already, is that some of these, uh, companies are coming out, Netlify being one of them, and being like, "We're dropping seat prices to try to compete," and be like, "Hey, we understand that your frustration, we're gonna only charge for hosting." Now they're charging more for hosting to compensate for the seat prices [00:41:00] being dropped. But like, there's already micro effects in the market due to the pressure, and it, it's not just AI pressure like you were saying. This was happening in 2021. Like the downfall of SaaS, this has just accelerated it a little bit because investors are starting to out a little fr- from, from these companies, so they have to react in some way, shape, or form.

Matt: Yeah

Mikhail: Um, to say that SaaS is dead as a whole is almost impossible. Because again, a lot of these services require massive amounts of servers that can't be subsidized by a one-time cost. Or if it is subsidized by a one-time cost, it would have to be like Plex, where they just increase their cost to like, what, $750 for the Plex Pass or something like

Matt: Something like that. It's, it's quite expensive for the lifetime license

Mikhail: X'd it. Yeah, they five X'd their lifetime license, but that's because they have to maintain a bunch of server infrastructure for those, for those people that buy the lifetime license, and they have to make like a, an indication of like how long [00:42:00] person gonna use it and how much money is, are they gonna cost, and then what's the profit margin on that? And then they start doing that, and then they're like-- And then not only that, they're like, "Well, if they need investments, they need to show profit and growth, so we need to price that in as well." So like how do we grow with this kind of subscription? And like it becomes a very complicated balance sheet issue due to our, like capitalistic nature and how markets work that as soon as we figured out SaaS, it's almost impossible to go back if you wanna maintain growth, which every public company or even private company that wants investments needs to maintain to continue on the same trajectory that they are. It-- So companies are either gonna have to die, they're gonna have to adapt, or, um, a different model is gonna have to come, come out. Like maybe again, like an agentic mo- I don't know, like s-something that I don't have a, a, a good clue in on what, what that could be

Matt: I, like you and I were talking in, in, uh, in this episode, I [00:43:00] believe, but also a- at before as well for sure, talking about how, you know, all these agents are... You know, agents are talking to agents, and agents are d- are buying things now, and agents are gonna run the world, and agents, agents, agents, and that's all we ever hear about.

I think we... Well, we definitely talked about it before the show. I don't remember if we mentioned it in the, in the course of our conversation sort of live here. But, um, that's all we hear about, right? Oh, agentic browsing. Give your agent a bit of a, a retainer or, like, a wallet, you know, and, and allow it to spend $100 for you, and it'll decide things for you, and it'll do this, and it'll do that, and agents, agents, agents.

And I, and I, I mentioned, I mentioned... Man, people are just constantly calling me too, by the way. Um, I keep, like, silencing a device, and then something else goes off. Then I silence the next one, and something else goes off. Then now my watch is going off, and something's ringing

Mikhail: Yeah. It's probably the agents. Their

Matt: Has never, has never happened in the, in the history of, of this show, I don't think.

And now I'm just getting... It's just like, "Matt, Matt, Matt, Like, I wanted to change my ringtone to that actually on, like, really busy days, where it's just, [00:44:00] "Matt, Matt," constantly. But no, that's a bad idea.

Mikhail: Mm-hmm

Matt: but the point of the matter is, is, like, you know, we're talking agents this, agents that, agents this.

agents can maybe spend some money and understand how to spend money, and they could do some budgeting things, and they could do this and that. The buck still stops with us, the humans. We are the ones that have the d- the, the motivations, the desires, the wants, the needs, and we are the ones that power the economy.

At the end of the day, the agents are not gonna be the ones that are like, "You know what?" Like, "I think I'm gonna start a new currency." you know, unless they suddenly gain sentience or f- somehow are programmed to have motivation, we... You know, the buck stops with us. And so if all of our, if all of our economic systems collapse, we're...

A- the agentic browsers still gonna be given money by us to then go and buy subscriptions and all these things, right? Like, no. Um, and so when you're saying, like, different business models, I think that, like, that makes sense. Having different license [00:45:00] tiers. I mean, uh, just recently, and I'm just paraphrasing, but the new CEO of Xbox, Asha Sharma, she was having a conversation about how people are gonna have that sticker shock at a $1,000 console.

If you're into video games, consoles, I don't think have ever been $1,000. Um, I know adjusted for inflation and things, that's why I said I don't think they ever have, but in terms of the actual sticker price, I've never seen a $1,000 console short of... Well, m- was the Xbox One X maybe? I don't know. Doesn't matter.

That was a niche console. But i- regardless of which, now we're talking about, I don't know, 11 to 1,500 American or, uh, Canadian, I'd say probably, 1,100 to 1,500 for the next round of consoles, PlayStation and that, due to RAM shortages, due to the nature of the business, due to COVID, due to, due to profiteering, due to whatever, right?

Like, list, list goes on and on and on. And she said people are not going to react well to that sticker price, and so you're, there, you're gonna have interesting business models come up. immediately my thought was, "Well, I'm gonna have the console for f- for [00:46:00] eight years or whatever the, you know, what- however long it's relevant.

Charge me a s- make it a SaaS. Charge me a monthly fee. I'm in your, I'm in your ecosystem. You're making money every time I make a game, like a royalty or whatever. I'm using your Game Pass. I'm using your Xbox Live or your Xbox Network or whatever they call it now. I'm doing that. Charge that for me. I'll, I'd sign up for that, 0%.

Sounds great." And I would do that. And I'm not saying that... That's not financial advice. Someone might be like, "Matt, you're an idiot," but, I mean, I really don't wanna pay, you know, 1,500 Canadian for a console, but I wanna play my games. And so if you tell me 300 bucks up front, then you're paying 30 bucks a month, okay.

Sounds all right.

'Cause I'm gonna-

Mikhail: right now. The, the, the problem with that model is, like, you're already doing that, so why would they, why would they go there looking for more money? Like, that's not... It would have to be way more than that, because you're already paying for Xbox Live or whatever, whatever it's called

Matt: It's just Xbox Game Pass Ultimate now, yeah, for me

Mikhail: Yeah, you're [00:47:00] already paying the 30 bucks a month, plus you're paying for six, $600 for a console. So it would have to be something like, you know, $500 for the console and then financed, a financed subscription fee guaran- like a two-year financed subscription fee of, like, another 60 or s- or $80 a month

Matt: Something like that. Or,

Mikhail: justif- to justify it

Matt: or what I, what I said to my one buddy was have it be a contract of sorts where

Mikhail: No

Matt: I have to buy Game Pass. Like, I'll pay, still pay monthly, but I have to buy Game Pass for three years or five years, and then have it, have there be, like, a big exit fee,

Mikhail: Mm-hmm.

Matt: something like that. And I, and, like, this isn't what...

You know, this part of the conversation is not what Ash has said. I'm just sort of, like, talking out loud to my, to my friends. But, like, at the same time, it's like, makes sense. Like, you know, the... So, like, different business models are gonna come out of this, and, and we're already seeing this in gaming.

you're into games, like, you'll, you'll start noticing that single player games are generally now [00:48:00] cheaper than these sort of big triple A, big multiplayer behemoths that are not free to play.

games are, you know, nine- some of them are $89.99, talking Canadian prices. Some of them are $99.99 and things, some of these, some of these games are bucks Canadian, which is considered cheap here. 50 bucks, 40 bucks, 30 bucks. You're starting to see the rise of the double A game again, and it d- doesn't all have to be triple A game.

You're starting to see that. Indie games already kind of fill that niche a little bit now, or a little bit, but double A games are just, you know, just a bit more budget, just a bit of a bigger, a bigger team. Um, a- and, you're seeing the market react, and so this, all this, you know, panic and everything, the AI stuff is so fatiguing, it's like, AI's gonna gut everything.

Is it? if it guts everything, then there's nothing there,

Mikhail: Mm-hmm.

Matt: and if that happens, someone's gonna come in and, and, and, like, try to stop it or stop it. Like, they're not just gonna let everything just go absolutely [00:49:00] down, down the crapper.

Mikhail: But most of these companies are rushing to be the ones that survive and capture the biggest market share before the correction needs to happen. That's the reality of what's going on right now. It's like the AI companies know that this can't take over the entire industry, every, every part of every industry, but they also know that going to be a, a, a, a period where they can profit, they can profiteer quite a bit before the governments catch up and, you know, they start to be pushed back and, or they start to get like some other... Their, in their ideal case, it would just be like a mincome where people still pay to use their products, but they don't have to work, but they still have money somehow. I don't know. It's, it's all kind of crazy to me. But they're trying to rush to be the ones that profit the most in the shortest amount of time before this correction needs to happen. So they're saying whatever they can to get people to believe it, to

Matt: Of course

Mikhail: to it, and to get people to use them a- as much as [00:50:00] possible. So there's a lot of marketing lingo going on, and the other part of this is there's a lot of AI psychosis happening as well, and I, I'm part of that psychosis sometimes too because I use AI a lot. And what, what I mean by the A- AI psychosis is like they believe this stuff. Like if you talk to Dario Amodei, who is the CEO and, oh, like founder of Anthropic, like if you see any interviews with him and he, like the stuff that he says like, "Hey, we can't release Mythos 'cause it's too dangerous. We can't do this 'cause it's, you know, like all of a sudden like, you know, jobs are gonna be gone.

In six months, no more developers." my opinion, he's not saying that as a marketing gimmick. I know some people think that he's saying that as a marketing gimmick.

Matt: A- and it's possible, like we're sp- like you're speculating as well, like

Mikhail: I am 100% speculating, 100% possible, but I don't think he's saying that as a marketing gimmick. I think he's under AI psychosis beli- believing that it's real because he's very like, he's very isolated, [00:51:00] He's only ta- thinking and looking at this stuff. But the reality, and this is what I wanna talk about right now, is like the industry doesn't just shift on a dime. Like you, you can't just ask all corporations to sudd- all of a sudden trust AI with everything and have it, like have all engineers be replaced in a year, have every, have every pipeline be changed, how, uh, have every like tool that's being used be like integrated with AI.

It's just not possible. It's happening slowly and probably faster than other industries have, have happened before, but it's not going to be this like switch that happens. And it's the same thing with SaaS. We're seeing it right now like- You can't just clone up a Notion replacement. A lot of times what's gonna happen is you're gonna clone up and it's gonna be cool for a day or two, and maybe even you're gonna give your-- get your company to start using it, and then some Joe, like s- some, like, you know, Bob down the, down the hallway is gonna be like, "Hey, my, um, [00:52:00] with, uh, Bix- Bixbop Jigabop," I'm just making shit up right now,

Matt: What?

Mikhail: there is no integration with it. And he's gonna come to you, the person that is promoting this, like, "Hey, let's replace SaaSes," and he's gonna be like, "It's not working. I, I did this in Notion. I clicked these three buttons and it was working, and it's not working here." And you're gonna be like, "Okay, well maybe I can fix that."

And you're gonna go in and fix that, and then Joe will come down the street and be like, "Um, you know what? My integration with, uh, the, uh, bespoke calendar service X- XB- FR2P-DRM-DRG isn't working as well." And you're gonna be like, "Okay, so I'll have to integrate that." And okay, and then someone else like the CEO will ring you and be like, "Hey, uh, it's 3:00 AM and I can't access my calendar, uh, on Notion bec- and that's where I access everything, and I only work at 3:00 AM and this is really my only work time.

So you're actually costing the company, uh, $7 million a second. Uh, can you get it back up?" And so like this is going to be [00:53:00] the life of the person that's gonna be in charge of replacing the internal SaaS applications that, uh, have been around for a long time. And I'm not saying this out of like comedy. I'm not saying this out of anything.

I'm saying this out of 100% reality. is going to be the case. So

Matt: Of course

Mikhail: internally trying to push, "Hey, I'll replace our Notion sub," or, "I'll replace our thing," think about it a little bit more. Think about the, I'm, I don't wanna say this too crassly, the most, the least technic- technical competent employee that uses the piece of software that you're gonna be replacing. And think about them calling you on a daily basis for stuff that is probably working but they don't understand because you're the one in charge of replacing that software. Not AI, are. And even if you say like, "Okay, I'll just pipe it into AI," well, if something goes wrong in that loop, you're still the one that has to interact with that loop. So gonna be as easy [00:54:00] as just replacing it. Honestly, these demos are about 10% of what it requ- of what's required from a SaaS. The rest of it is literally support, maintenance, security, integrations. Like I was saying, Bibbap Jaboo, Jaboo. The, the integration with Bibbap Jaboo is really important. I-- and if you ever worked at an enterprise, these names sound weird, but I can guarantee you've heard of a random name like that that was super important to Joe down, down the hall, and you're gonna be like, "You've never heard of this software before, but this guy can't live without it," or, "this integration before, but this guy can't live without it and it's costing the company millions of dollars a second every time he can't use that random integration." The, this is the, the moat. The moat is it's not just like a, a, a software product. It's a maintenance software product. The maintenance around it is very complicated. These big companies like Notion have hundreds of people just dedicated to that shit, and they've gone through the wringer. They have gone through the wringer.

Trust me on [00:55:00] this. The calls that they've had, the support tickets that they've had are crazy compared to what you imagine them to be, and they've solved for a lot of those cases. And now they have a fairly stable product that they've invested millions of dollars into, and they have a machine, a support machine around that product that is able to up the phone for a, for, at, at 3:00 a.m.

and talk to your CEO and actually solve their problem without having to, you know, wake up the entire company or whatever. That's the reality that we're, that we're, that we're, going in. Not only that, the other part of this and the mo- the other most important part of this is from an existing service to another service.

Let's say you decide to make the, the Notion switch, You're a, you're a Fortune 500 company and you're like, "We're, we're gonna switch from Notion to our own internal Notion tool that we've just generated in, in three minutes." Migrate the data

Matt: There's a reason why mainframes are still, still exist. Not for, not for Notion clearly, but for massive amounts of [00:56:00] data and stuff. And, and there's a reason why that hasn't been migrated away. I mean, if you have a new company, it, I g- uh, what's a company on a mainframe? Give me a big company

Mikhail: Oh, like banks, like any, any bank.

Matt: Sure. So like, I'll,

Mikhail: Walmart.

Matt: TD Canada Trust.

It's a local bank. Let's just, like I'll just assume they're on mainframe. I have no insight. They have a mainframe in the background, whatever, blah, blah, blah. If you today, if Mike and I were like, "We're gonna create a new bank," we would not get a mainframe. My point is, is that mainframes are replaceable, but people have not done the migration for that yet, and there's good reason for it

Mikhail: Like, and the migration isn't just data. okay, you've made a Notion clone. Is it one-to-one?

Matt: What's it, what's it all connected to?

Mikhail: yeah, is it, is it the same URL? 'Cause again, Bill down the hall only knows how to go to notion.io.

Matt: Correct

Mikhail: He can't write Notion L. That's impossible. You are asking a [00:57:00] ludicrous-- Y- you're asking them to change their entire lives. And I'm

Matt: I,

Mikhail: bit facetious, but,

Matt: well,

Mikhail: of the situation.

Matt: you're not being facetious to me

Mikhail: they will react.

Matt: Like I, I, I was in IT, like we had, you know, I, I had a guy call me one time, allegedly a networking expert anyway, calls me up and he goes, "There's something really wrong with the server." I said, "Oh, oh geez. Like what, what's going on?" I go look at it, everything looks fine.

I'm like, "What's, what's, what's the situation? Like, what are you getting?" He's like, "Well, the uptime says two days." I look at the update schedule. It's like, yeah, it went down years or two days ago 'cause we updated it. like, "Yeah, but the uptime," he's like, "the uptime," he's like, "There's something wrong with the CMOS battery."

I'm like, "Something wrong with the CMOS battery? The clock's fine. What are you talking about?" And he goes, "Uptime is the collective uptime of the machine." No, man. No. Uptime is the current consecutive uptime from last power on or restart. Is that what uptime means? I was the co-op. [00:58:00] I was, I was the intern effectively.

Co-op basically means intern. Um And I, when he said it, I was like, "Wait, maybe I'm an idiot." So I went and asked a colleague, and I was like, "Uptime is just a consecutive," you know? And he's like, "Oh, yeah

Mikhail: Yep. And I ha- I have another story about that too. I had to switch a monitor for someone at a, at a... I, I was an intern as well doing IT co-op. the monitor for them. Next day I get a call, "Monitor's not turning on." I asked, "Have you pressed the power button?" And they said, "Yeah, I pressed the power button many times.

It's not turning on." I come downstairs, I press the power button, it turns on. Next day, same thing. "Monitor's not turning on." "Have you pressed the power button?" "Yeah." I told-- Like I, I obviously told them where the power button was that one time, so I came down, pressed the power button. They're like, "Well, it's impossible for me to press the power button.

It's not the same as my other monitor." They can't, they can't press the button. do it. And there's no, like, there's no rebuttal to that. There's no rebuttal to that as a co-op,

Matt: I do have a rebuttal though. I do have a rebuttal though[00:59:00]

Mikhail: no, there, there's no rebuttal to it as a person that's gonna be replacing Notion as well.

Matt: Oh, sure, I guess. Yeah

Mikhail: yeah, like if you're, if you're Random McGee that's, like, a, a part of the engineering team replacing Notion with a, with a vibe-coded application, and stuff isn't exactly as it is, and you don't have the authority to push back on employee Bill, you're gonna be dealing with this exact thing.

"Hey, I have to click three pixels below this, this area that it was in Notion. I can't do that. You're costing me millions of dollars," is going to be the argument against you.

Matt: Yeah.

Mikhail: not, I am not th- I am not joking. That is the reality. So anyone that's replacing these things in a traditional, you know, Fortune 500 company is not gonna be replacing them much.

If they're gonna replace, they're gonna replace one thing, and they're gonna realize it's not worth it, and they're gonna just pay the SaaS price as hopefully it starts to become more competitive, that's it. Small startups, different story. a different story. That's not where the, the most of the money is made on these SaaSes, so I don- I don't think that [01:00:00] they're, like, fully focused on that. But, like, it is, it is an indicator of, like, if it's a small startup, there's, like, 10 people working, and they need a, a Notion alternative, maybe they could co- code it up themselves. There is some argument there.

Matt: Sure

Mikhail: mean? Like, 'cause that could be, like, a, a cool hackathon project that they do internally anyway. And then people will... are more willing to usually suffer through some of, like, the changes, and they're a little bit more adaptable and, and it's a little bit easier to deal with 10 people versus

Matt: I'll, I'll do it for, for WordPress plugins. Like, I'll even tell a client, "Hey man, like, a more professional UI for this, uh, was $50 a month, so I just made this really quick. Hopefully that's okay, 'cause I didn't wanna balloon your budget," kinda thing. And 90% of the time, small business, they're like, "Hell yeah.

Like, I don't wanna pay that." you know, I'm down, I'm down with just having it be a little bit janky. And then of course I could spend more time it nicer, but I don't wanna, you know, up, up their hours, like, 'cause their budget matters. And, and that's it. Like that's, you know...

Mikhail: Yeah,

Matt: So there's, there's realities in, in the real world that come up like that

Mikhail: And that, that's the reality. That's, that's where, like, [01:01:00] the reality is, is being, different from what the market is reacting to. uh, the market is almost reacting to the startups, like, why these companies are still seeing growth in sales. Like the, again, back to, back to Salesforce if we, if we go back there. Uh, the, the number is going down. So if we go back to year-to-date, the number's going down. But the reality is they, they actually have posted profits, like big profits over the, over that one-year period.

Matt: that one, what was it? billion deal we mentioned in the beginning as well, a government deal

Mikhail: deal. That's correct. They, they posted a massive deal with the government. And that's a big deal, but it's not being priced in because for what- not for whatever reason, there is reason for it, but, like, there is... The, the reason that it's being priced in is because, like, they're not seeing the growth. So the, the traditional companies are probably gonna continue to use Salesforce.

They're not gonna all of a sudden switch on a dime to some random Vi- Vibe coded application. [01:02:00] 5 to 10 years down the line, if there's 10 other competitors with Salesforce, that could affect them, right? And then the new companies that are choosing to use Salesforce or not, like Matt was saying with the, with the mainframe systems, might choose to build their own version internally or, again, choose a competitor that is, has less moat now because they are building more rapidly due to AI.

So that, where we're seeing the market react to, but the reality of the situation internally in these companies is they're still seeing growth, they're still seeing a ton of usage because nothing's gonna shift on a dime, and it's impossible to ask a massive company to Vibe code all of their internal applications that all of their users are using because the upheaval and the uproar that's gonna happen in- inside will cause every engineer in there to quit. So it's, it's a ta- tale of two kind of si- tale of two sides of the coin, but, like, it's not... [01:03:00] It's also not like, you know, this, this is exactly what's happening and, you know, one side is, uh, is right or the other side is wrong. There is some merit to the stock going down. Maybe not as much because, again, they're still making a ton of money, so in my opinion, the market is reacting too much, and on the other side it's like it does have to react to the fact that SaaS needs to adapt

Matt: Well, despite, despite it being a fool's errand to predict the future, I mean, I, I, I'll give you a prediction. My prediction is the, this is going to be similar to website builders and how disruptive they were. Were they disruptive? Yes. Was it apocalyptic? No. And I mean, SaaS, again, had some writing on the wall, had some cracks in it.

People were sick of paying for subscriptions, so then new business methods come out. I mean, SaaS was a new business method. really started when we were in college. There were SaaSes before and there are certainly SaaSes after, but now almost everyone's a SaaS after. And there comes a point where people are just not gonna wanna [01:04:00] pay per month.

And I will say this, is, I would assume, and it is definitely an assumption, that more people now are becoming financially literate due to finance content on YouTube and things. Caleb Hammer and stuff, love him or hate him, he's, he has a big show, Financial Audit, as well as several other different, uh, podcasts and things that cover business and, and personal finance and things.

And one of the very first things I hear from people that are worried about debt and people going into debt is, "Check your subscriptions and get rid of the ones you don't need."

And that's almost like a little mini war, not necessarily groundbreaking or anything, but it's almost like a little mini war against SaaS.

Mike and I pay probably, like, hundreds of dollars through the business more than we need to 'cause of SaaS. 'Cause it's like, well, like, if I get rid of this subscription, it's gonna go up 50% 'cause I'm on a grandfathered-in plan and I use it twice a year, so now what?

Mikhail: Yep. maybe, like, and, and this is what leads me to my final point here, and something that, like, kind of a challenge that I'm putting down on myself, I guess, that I'll be doing [01:05:00] publicly, is I wanna pick a couple of subscriptions that we have or a couple of pieces of software that we have, I wanna see if what people are saying is true. Maybe I can... I'm not gonna, I'm not gonna say I'm gonna one-shot them. I, that, that I think is kind of a sensationalist headline. But maybe I can spend, know, a few hours, let's say, of my time trying to replace them and see if it's replaceable, and if it could save us some money. And then I'll come back on the show, or in fact what I'll do is, because this episode won't be released for two weeks, I think, um, so it's not next week, the week after that y- that you'll be hearing this. In the show notes, so go to the website, go to the h- htmlallthethings.com if you're listening to this, and find the show notes for this episode. In the show notes, I'm gonna have an update section where I'll either have a statement on, "Hey, I was able to do this. Here's the link," or I'll have a little, you know, demo section, or I'll have a statement on being like, "This was bullshit, and we weren't able to replace anything." [01:06:00] So you'll have an update in that if you're listening to this

Matt: Maybe a web news even or something. Like, just depends on how it goes. Yeah

Mikhail: a good idea, actually. There might be a web news for the week of the release of this e- of this podcast. I just need to remind myself. But yeah, um, gonna be some sort of update either in the show notes or in the web news for this we- for this, uh, episode.

Matt: Cool.

Mikhail: But yeah.

Matt: I look forward to it

Mikhail: but that's it. I think that's it.

Matt: That, that's the episode. Uh, I would love to hear from you, the audience, though. Do you think this is... You know, SaaS is... The writing's on the wall? Do you think that this is just a blip and, you know, maybe a bit of a correction? Do you think that they're gonna start skyrocketing upwards again? You know, what do you think?

Are they gonna grow slowly? Are they not gonna grow? What's gonna happen to SaaS, and is it due to AI or not? But if you wanna support episodes like this, do so on Patreon. That's patreon.com/htmlallthethings, check out the tiers and give that a go. And many thanks to [01:07:00] our $3 tier patrons, Tim from The Web Hacker on thewebhacker.com, Jason from Geek Life Radio via geeklifeeradio.com, Garrett Segall, Level Up Financial Planning via www.levelupfinancialplanning.com, Magnus from Yes Web via yesweb.se, Syntaxify from the HTML All the Things Discord server, and Stacy Mostiller from the website swoonworthydesigns.com.

And last but not least, we'd like to give a shout-out to Michael Larocca, a contributing author on htmlallthethings.com. Michael is the author of Self-Taught: The X Generation blog at selftaughttxg.com. And remember, you can get up to 20% off a Scrimba Pro plan using our link that's in the show description and in the show notes with full details on how it works in the show notes on htmlallthethings.com.

Feel free to leave a comment or review on the platform you're listening to this on, and we are signing off