UDSKRIFT 🔗
Isaac: Welcome to Never Rewrite. I'm Isaac Askew.
Jeffrey Sherman: And I'm Jeffrey Sherman, and today we're gonna revisit the SAS Pocalypse and ask: Is reliability a new moat for a SaaS company? So going through the numbers at the end of the SAS pocalypse episode, and we'll include a link to that one,
Isaac: Mm-hmm.
Jeffrey Sherman: the basic breakdown that we came up with is you could replace a fully featured emails can like a marketing email provider with about two hundred fifty dollars in cloud credits and save yourself ten dollars a month. So like a two-year
Isaac: Mm-hmm.
Jeffrey Sherman: back ROI. And It it wouldn't be for everyone, but for anyone who has a lot of developers on staff, this could be a thing where, in the SAS pocalyp world of, hey, does it make sense? Could we get rid of our SAS providers? Because we could do it ourselves, we could vibe code a similar thing and then switch from a full featured provider to a basic provider a and save the money. And we came up with kinda. But we didn't consider the option the reality of reliability, right? If you vibe code something, you have to maintain it, and that costs
Isaac: Mm-hmm.
Jeffrey Sherman: time. And just from an economic standpoint, it would be nearly impossible to save to s to save ten dollars a month while still maintaining code in any way, shape, or form. developer, having a a develop the first time the
Isaac: Mm-hmm.
Jeffrey Sherman: developer has to go in and fix something. You've lost your ten dollars a month for two years.
Isaac: Well, maybe.
Jeffrey Sherman: Maybe.
Isaac: It depends. if you have maybe you have something automated set up to to give maintenance to itself. Like to upgrade its own packages or
Jeffrey Sherman: Mm-hmm.
Isaac: run tests against itself to make sure it still works. You know, like the maintenance maybe you could automate the maintenance part. Or do a self hosted L L you know, that can do the maintenance for you.
Jeffrey Sherman: Well, I was gonna say you y if it's an internal tool, you don't have to expose it to the internet, so you don't have to main up maintain your packages inherently. Right?
Isaac: Mm-hmm, yeah.
Jeffrey Sherman: 'Cause you realize, well, it's not connected and so the fact that it's insecure is whatever. Is you're you're running oop Yeah.
Isaac: But even still you could automate the security of it. You could still upgrade it, And then aga again you could have a self-hosted Mac Studio that has an LLM on it that, is you you use that and you're just paying for the cost to run the the self-hosted thing. So maybe. But I I think the larger point you're trying to make is the reliability is Something w people might still want to pay for because of the the headache of setting this up. Because we're we're tech savvy. We
Jeffrey Sherman: Yeah. Right.
Isaac: know how to do all this stuff and set up servers and, host things like that locally. Some people aren't. And this is the same people that are buying things like active campaign software, where they're like, hey, this makes my email marketing easy. I'll pay
Jeffrey Sherman: Mm-hmm.
Isaac: for it. Rather than, let's there there's a young kid on staff here that is good with technology. I wonder if he can just set up all this locally and figure it out for us. The headache of that I think well, I think it gets further complicated too because you're gonna like hoppy cats making the same software, but being just as reliable but charging half the price. And I think you'll see maybe the price is coming down or some of the other tiers
Jeffrey Sherman: Mm.
Isaac: coming down. And if it becomes cheap enough, then like why would you unless you just don't want to manage like thirty different subscriptions, why would you not just pay three bucks a month instead of nineteen bucks a month for a software that does just the one thing you cared about? the email sending part of active campaign instead of the automations. Right? What if you manually sent out your own blast and you just didn't need an automation? And you're like, well, this other software doesn't have automations, but it's three dollars. That serves my thing, it's reliable, let me go with that. Right? So
Jeffrey Sherman: Mm-hmm.
Isaac: I think there's complexity there. I think you'll see like a whole variety of people, some paying for the standard one because they're used to paying for that and they don't know they're not tech savvy enough to switch. Some people who are tech savvy enough and want to host it themselves. And a combination of people who host it themselves and just set up their own automations to keep it up to date.
Jeffrey Sherman: So you don't think that reliability in and of itself is a moat for SaaS companies to to ward off vibe coded I guess not vibe coded competitors,
Isaac: I won I won't say that.
Jeffrey Sherman: but just vibe coded independent people, like customers. okay.
Isaac: N no, I think it is a mode. no, I I I I I because I mean even now, like again, I've I've done my fair share of vibe coding and there's things where like I'll have alerts when things go down and occasionally my own stuff goes down and I get like a dead man switch thing letting me know it went down. Like that that wasn't reliable. I coded something that wasn't reliable, it's a vibe coded. In the same sense if somebody vibe codes a competitor and their API isn't down. Even if they're three dollars a month but they're 80% reliable, I might pay $19 a month for 100% or 99.99 whatever percent reliability. Right? Yeah.
Jeffrey Sherman: I definitely would. It like it's
Isaac: So it's still it's still a moat. I just I'm not convinced that the when we say vibe coded, like vibe coded now versus vibe coded from two years from now, we've seen some crazy improvements on how well vibe coded things can work and how reliable they can be in the fail safes. So I'm I think some vibe coded projects will. give give some SaaS products a run for their money when it comes to reliability. and then some
Load full transcript
won't. And then like there'll be a whole pro just like in the AI, sorry, the app store on Apple on on phones, like there's just it's inundated with a bunch of slop software. Some will rise
Jeffrey Sherman: to be
Isaac: to the top.
Jeffrey Sherman: Yeah, and they were inundated with slop even before AI.
Isaac: Right.
Jeffrey Sherman: Where it was very common to they would come up with a game and then they would reskin the game and then they would slightly tweak one thing and they they so you'd have a company like every month they'd be releasing a game but it was still the same.
Isaac: Right. Well, but the speed at which you can release them is crazy. Cause before you had to go set up your Apple developer account and it's like ninety nine dollars a year. then you had to create different keys with different permissions, download those keys, boot up X code, code the thing, upload now I I guess they actually made it like the X cloud kind of thing where you can push to GitHub and have it automatically create a release for you, that kind of thing.
Jeffrey Sherman: nice.
Isaac: but for the vibe coded project I recently did, I just logged in. And said, just control the browser and just publish a brand new iOS app for me. And it created the entire app, did all the permissions for me, set it up, published it. I had to do
Jeffrey Sherman: Ha
Isaac: nothing. Right. So the ease in which you can do all this without having any knowledge of all the different don't don't forget to download the P list key and blah blah so and so and it's crazy. It's very, very user friendly.
Jeffrey Sherman: So if you are running SAS, it it's hard to run on, hey, we're reliable. it's hard to market to the concept of, hey, we're reliable,
Isaac: Yeah.
Jeffrey Sherman: therefore you should pay us. How would you go about showing that or or indicating that without
Isaac: I don't know. I wonder if that's just something people will have to experience. They'll they'll I I don't know if it is hard, 'cause some people run off reliability. AWS is proud of their nine nines of reliability, right? Nine nine point nine nines.
Jeffrey Sherman: Well nine nines? Wow.
Isaac: or something like that. Or I forget which which uptime company it was. I thought it was AWS. Maybe
Jeffrey Sherman: Five nines maybe. But
Isaac: it was five. Either way, they're they're proud of that stat and they sell it on that. this is how reliable we are, and that's impressive to me. So I don't know if that's necessarily true that it's hard to run on reliability, but let's say it is for the sake of argument. And people go away and they have issues. They might go back to active campaign if they want to go swap away from active campaign for something that they think think is vibe coded but cheaper. And it doesn't work. And they go, like like my open source project I released, it is
Jeffrey Sherman: Ha ha ha.
Isaac: it doesn't it doesn't work for them 'cause it was vibe coded and I missed out a lot of stuff I don't you know. And then they go
Jeffrey Sherman: And you didn't go back to it either, it was just a one and done.
Isaac: is one and done, yeah. So then they go back to active campaign 'cause they've got the support, they're familiar with the products. They've got that reliability moat you're talking about. So I don't necessarily think I don't think it's hard to sell, but I guess if is is it the reliability thing
Jeffrey Sherman: It's a different sale.
Isaac: is yeah, it depends on one yeah you gotta be tech savvy enough to want to swap anyway. 'Cause people I mean, this is my experience just dealing with my family. Like when I teach my mom how to use software. She will never change software. She hates technology. She hates having to learn new things on a computer. It's just like
Jeffrey Sherman: She's still on MySpace. That's that's her social network and she's not moving.
Isaac: Facebook, but basically, yeah. and she's not gonna learn a brand new platform. it was hard enough giving her a Mac and having her switch from a PC to a Mac and she's like, I don't understand where everything is right. So like people want they're gonna stick with the things they know and you would really have to do something dramatic, like jack up your cost substantially, or really screw up something for people to be like, I'm switching. It's like same thing for me. Like I let let's say I get on a a support with Comcast. I hate dealing with Comcast, right? But I don't hate them enough to swap. Because it's such a pain to switch. But at
Jeffrey Sherman: Yes.
Isaac: some point I will get so angry that I'm sure
Jeffrey Sherman: The US.
Isaac: I will hit that threshold. They'll finally do it under or like lately they've been laying in new fiber. there's like Google Fiber and some other stuff happening nearby. Maybe they s they make it so easy for me to swap and I remember how much I hated dealing with the support before and there's a plan for getting in on early that I'm finally like, let's do it. There needs to be something some kind of factor, some pushing factor to make people want to swap, I think. And you'll have to hit that before people will want to abuse that mote.
Jeffrey Sherman: So it's not a new moat, but it it is still a moat, I guess, where even when the age of the SAS apocalypse where customers could do it do it themselves, reliability and the cost of switching is still a good moat.
Isaac: Yeah. That's it.
Jeffrey Sherman: But does that encourage dark patterns of making it more difficult to switch? I guess it does. Yeah.
Isaac: Sure. That's always been the case. Check out all the, you know, all those companies that made it so hard to cancel, like Adobe looking at you. speaking of
Jeffrey Sherman: Yeah.
Isaac: which, actually, how this all started when we talked about SAS Pocalypse was the reverse coding or or verse
Jeffrey Sherman: Mm-hmm.
Isaac: reverse engineering w from the the Adobe conversation. apparently, as of a couple of days ago, they put new guardrails in to Claude and maybe some other maybe open I I did this too, but to prevent the reverse engineering of certain softwares. Yeah, that's
Jeffrey Sherman: Really? That's very anti-usual.
Isaac: well, it might be to do with I mean the irony here is that they have stolen so much information and you know and and tried to copyright it, but and now you know, now that is costing other big companies money because you can reverse engineer their they're going, well wait a minute, that's theft. You can't do that. So
Jeffrey Sherman: Ha ha ha
Isaac: But yeah, I don't know. I think it was I think it was anthropic that I'd was talking about garbrails from that recently. But I can I can look up article and put it in the the show notes too for people who are skeptical of my claim.
Jeffrey Sherman: That that is I I believe it, it's wild. Pot calling the kettle black here.
Isaac: Yeah.
Jeffrey Sherman: All right. Well, thank you all for listening. Be reliable. I'm Jeffrey Sherman.
Isaac: And I'm Isaac Askew, and this is Never Rewrite.