Isaac: Welcome to Never Rewrites. I'm Isaac Askew.

Jeffrey Sherman: And I'm Jeffrey Sherman. And today we're going to talk about pushing the philosophy of thesy theseus shipping outside of the tech stack. The so the thing that Isaac and I were are hoping to bring today is we're going to talk about stories that evolved out of shadowing people, but how the idea of, hey, I see something's wrong, I see friction, and I'm going to fix it, not because anybody asks me to, but because I can and I see it. And outside of the tech stack. ⁓ 'cause one ⁓ it's not a flaw, but one limitation of these shipping is if you're not talking outside of your own outside of the IT groups and you're just talking to other developers, the only things you're gonna fix, the only friction you're gonna remove are in the tech stack itself. You're gonna make the tech better, but that doesn't necessarily translate to actually making the outcomes better. If you you know, if people are living with a god awful workflow and you make that god awful workflow thirty percent more efficient, it's still a god awful workflow because that was the efficiency wasn't the why it was god awful. ⁓ so Isaac, I I know you have a great story about ⁓ an onboarding. You you got you were asked to help with onboarding and you helped this lady. So tell us that one. Kick us off.

Isaac: Okay. Mm-hmm. Well, I think originally the ask was I was just more of trying to understand what somebody's role at the company was. ⁓ I think I was new to the company and ⁓ I was just going around and like kinda meeting people and then once I saw someone's workflow, I was just like, ⁓ wait, wait, what are you what's going on here? ⁓ and so I guess you you know how it is like whenever you started a company and you just have like a w one way that you start doing something, right? It doesn't even matter if it's ⁓ technology technology related. It could just be that you're working with one particular flow and then you never really change that flow because you never really have a reason to change that flow. And then after, you know, three years goes by, then someone else comes in and goes like what it what's going on here? ⁓

Jeffrey Sherman: Mm-hmm. Mm-hmm. Mm-hmm.

Isaac: Because you just get used to the way you do things. Like you there's there's no need for the self reflection, so to speak. Yeah.

Jeffrey Sherman: Right, there's the way you've always done things and if you don't have if you don't go and seek a reason to change it, it takes a long time until somebody comes along, you know, usually your grandkid's like, Grandpa, why are you doing it? That would be Why grandpa, why are you doing it this old fashioned way? There's so much better ways to do it now.

Isaac: Yeah. Yeah. Right. And y if it ain't broke, don't fix it, right? ⁓ and so sometimes you don't know that maybe the thinging is broken in the sense that it's taking a long time, but y it just feels very familiar, right? So when I was ⁓ asking to to sh ⁓ shadow this person, I just basically just I wanted to ⁓ back in the day you'd sit at their desk and watch how they did things, but this was remote. So I just helped someone call and had her share the desktop. ⁓ and what the workflow was is for this particular company,

Jeffrey Sherman: Right. Mm-hmm.

Isaac: They would work with hospitals and some hospitals owned other hospitals. So you'd have like a different that we called them payes. ⁓ and whenever we partnered with those companies to help them pay their bills, we would have to put all of their rec all their pay records in our system. ⁓ and then they would they would come in a spreadsheet of like, okay, here's our new partner and all their affiliated clinics or whatnot.

Jeffrey Sherman: Mm-hmm.

Isaac: ⁓ and there was even an internal tool that was built that helped people add the pay. But it helped you add one pay at a time, right? ⁓ so back in the Well, I mean when they started, probably not very many, which is probably why they built a tool and they just did one at a time. ⁓ but at this particular flow, there was like 300 on there. ⁓ and so I'm like,

Jeffrey Sherman: Okay. And how many lines would be on a spreadsheet? That did ⁓ one at a time. Yeah. Mm.

Isaac: So y your your week is essentially going through one by one and inserting these payes. And ⁓ they didn't have a better flow for that at the time. So I'm like, all right, well this is one of those things where we can we can build a script to write a script to automate some of this. And it sounded like at some point in the past somebody actually had a script to do this, but I don't know if it it the format of the spreadsheet changed or something was off. I I forget.

Jeffrey Sherman: Well Mm-hmm. Mm.

Isaac: Some of the details, but they just stopped doing it that way anyway. ⁓ so it wa this was a good opportunity for me to be like, okay, well, I don't know if I have the bandwidth to say I can just jump in and fix this for you. But at the time, nobody was against the idea, so I just suggested it. And I ended up correcting this this script ⁓ and helping import all these payes and you know, just a one one time thing. ⁓ so it saved her a ton of hours throughout the week to do this. And this is one of those areas where, like, whenever I got done, I kind of stepped back and thought, I wonder how many other areas are like this in the company where someone's just, you know, used to, I mean, used to doing things a really old way and not realizing there's a better way. And this is especially applicable in like the AI age, I think too, where we can generate very bespoke quick scripts that solve problems a lot faster than we could.

Jeffrey Sherman: Mm-hmm. Yeah.

Isaac: ⁓

Jeffrey Sherman: Yeah, in my head I was thinking, man, in today's world you could have that script done in under an hour with with AI. Yeah, with tests and

Isaac: ⁓ easily. With tests and everything. Yeah. And and better historical knowledge. Like for me, there was w as I was ⁓ correcting this script, I found a bunch of different columns that needed to be set to true or false, or have these other kind of things be adjusted. And since I was new to the company, I didn't know what some of the columns meant. And some of the columns were named very strange things, like

Jeffrey Sherman: Mm-hmm. Always.

Isaac: BLP ⁓ you know, whatever, you know, and so you're like looking at the comm, confused at what it does, and you have to ask around what it does and kind of infer. And then after a long goose chase, suddenly you realize that com actually isn't even in use anymore, but it's just historically is always added there to the spreadsheet, that kind of thing. Whereas AI today could scan all that and you can have it pull in confluence docs and slot conversations and everything it can tell you. ⁓ that one's not being used anymore. And here's why. And here's how you can go investigate yourself if you don't believe me. Because Claude has been listening a lot lately. All right. So ⁓ that actually inspired me to put a a bi-weekly placeholder on my calendar for myself that was just called like I just it was like a 15-minute slot thing that said metacognition. So think about the way you think. And every time I saw that on my calendar, I'd think, okay, did I do anything these last two weeks?

Jeffrey Sherman: Yeah, I wouldn't believe it. Mm.

Isaac: that I am just doing because that's how I've always done it. Like let me go through all of my processes and think if there's one thing I'm doing here that can be improved now, ⁓ to where I don't have to spend this it ha especially things that are repetitive. In my own workflows, I is there a pattern in how I keep re-engaging with my company? And if so, is that pattern automatable or if it's is it just an artifact of the job I have to deal with?

Jeffrey Sherman: Nice, yeah. had a similar experience where I was shadowing someone and ⁓ they were in a sales role and so I was watching them ⁓ doing the smile and dial, and every time they were gonna call a new person, they had to open up a form. And the form took forty-five seconds to load. And I was like, Well, that's weird. Why does this form take forty-five seconds to load? It's you know, there's literally a thousand salespeople at this company.

Isaac: Cheese.

Jeffrey Sherman: And that's a lot of time to be spending on loading a form if they're gonna be loading it all day, every day they they talk to a customer. And and so I went back and I looked, and the reason it was doing that is because there was a drop down field to say there were several drop down fields that had a list of everyone in the entire company so that they could assign things.

Isaac: Mm-hmm. Mm-hmm.

Jeffrey Sherman: Like, you know, who who who referred this, who you who's gonna do the follow up, blah blah blah, like CRM kind of things. But because there were a thousand salespeople in this company, it loaded a thousand records from the database and plopped it into this dropdown and then for the next one loaded a thousand records into the drop down. Right, so first thing I did is I quickly made it one look up and then reused it three times and that s took like twenty seconds off and people were like, ⁓ my god, you've

Isaac: Okay.

Jeffrey Sherman: They noticed. People noticed. ⁓ and again the the and then later, ⁓ 'cause that was more work, I changed it to a type ahead where it's like this shouldn't if you've got more than ten elements, you probably shouldn't be a drop down anyway. And a thousand is certainly way too many. So like it just it was just the wrong structure entirely. And they didn't even use the form field all that much. It wasn't even like it was a critical field.

Isaac: Mm-hmm. Yeah. Right. Or just lazy load it in general. Yeah, just load the form first. I imagine putting that in is, you know, second to getting all the information from a client. Hmm.

Jeffrey Sherman: Yeah, it it the form load went from like forty five seconds to like three.

Isaac: Yeah. Nice.

Jeffrey Sherman: It saved, you know, duh hour hundreds of hours a day ⁓ of employee time and just people were better. And again, nobody it wasn't important enough they would ever make it over the line of you know, some manager from operations or the sales talks to the manager of IT and says, Hey, this is slow, can you fix it? No, this was me, I was shadowing someone. And I saw that there was a problem, and I didn't go out and fix everything, because that I mean that form was a real piece of garbage. And over the years I had at least eight iterations on it that nobody asked me to do to to make it from a steaming pile of garbage to ⁓ something that fair enough. Where

Isaac: A contained basement pile of garbage that's not steaming.

Jeffrey Sherman: Yeah, like another thing that that I fixed cause after watching them I like, ⁓ the tabbing order was wrong. It was i under the covers it was an HTML form, and at some point some developer had screwed up the tabbing and then other developers had never fixed it. And so there's like two hundred fields on this form and the tabbing was all messed up and the guys were like Boop boop boop boop boop shift tab tap tab and it is like they had muscle memory at this point. Right, yeah, like from first name to last name was five tabs.

Isaac: Yeah. ⁓ I remember the story. Yeah, yeah. It would memorize the number of tabs you had to do. ⁓ my gosh.

Jeffrey Sherman: Because it jumped all it just and like I I had somebody go through and I fi we fixed it, right? It just then it became order. And but that put the idea in people's head, like, ⁓ this workflow, this thing that's always been this way that is terrible and we've ha always had to deal with it. We could change it. We could fix like it doesn't have to be terrible. We could start thinking about what would we actually like to be here? ⁓ and

Isaac: Yeah. Yeah.

Jeffrey Sherman: That form specifically couldn't be changed 'cause it there was a regul you know, it was a regulated industry and there there were laws about what fields needed to be on this form and

Isaac: Mm.

Jeffrey Sherman: But just simply fixing the form, started putting those ideas in people's heads like, ⁓ we could do this differently.

Isaac: Yeah. So th that got that has me thinking about like I remember I I used to always just wanna just sit at the desk and go go go go go and just knock out stories left and right. And it I would, you know, occasionally run into something that was pretty complicated and required a bit of thought where I had actually I've had to force myself to slow down and really plan it out. And almost every time I paused and went for a walk or did s took a break, I'd have a Eureka moment. I just you know, if I just stopped trying to force the solution, I would have a Eureka moment. So I think so there's something to be said

Jeffrey Sherman: Mm-hmm.

Isaac: 'Cause a lot of times people are not having these conversations with people like from operations because they've got their own bunch of work to do. You got your velocity and your sprint, ⁓ you can't inject a new story into this. Sprint is already packed full of all this. There's no room to address that. And there's some organizations even use like a twenty-five percent tech debt thing to try to tackle that. But that's not even organizational tech debt. That's just tech stack tech debt. The organizational tech debt is this kind of stuff. ⁓ so I find that like

Jeffrey Sherman: Mm-hmm. Mm-hmm.

Isaac: This isn't the first time or the my the story that I told a moment ago isn't the first time that I helped somebody this way. I think pretty much every single time I've worked with the operations team and had a chance to see how they're handling customer support issues, how they're handling their own internal tooling or solving their own problems, or just seeing some very manual looking day-to-day operations. Or even things that like we had really old tools and you just knew kinda like what you're talking about with the tabbing, you just knew, ⁓ you have to scroll here and do that and you have to

Jeffrey Sherman: Mm-hmm.

Isaac: for this when you have to reset this thing and go back to it again to make it work. Thing weird things like that that they just know how to do. And that if you f can finally just sit down with somebody and have the the breathing room in your organization to be creative to solve these problems and not just always be forced to pack the sprint with stories, then you'll you know, you can solve these problems. And I I know working with the ops team like

Jeffrey Sherman: Yeah. Right.

Isaac: It was just a great relationship. Everyone was thrilled every time we put some new stuff out there that really simplified their workflows. Because like you're talking about with the forum shaving off forty-five second load stuff, that stuff gets thrown around all over the place. And everyone's like, ⁓ let's get the new feature out or let's, you know, improve this bottleneck over here. And then all the op stuff gets forgotten because ops got it handled.

Jeffrey Sherman: Mm-hmm.

Isaac: But having that integrated engineer that can kind of pop in every now and then and be like, So how's ops doing lately? You guys need any tech improvements? Having some kind of pause for that I think would be super helpful.

Jeffrey Sherman: Right, a different way that I would kind of think about it is the you need to ⁓ and this is something we've struggled with good way ⁓ ways to make it to say this well, so I'm gonna say it poorly. It's you need to get the time to have tech and people outside of tech to sit at the table and spend the time to understand each other. ⁓ 'cause otherwise the thesis shipping

Isaac: Mm-hmm. Yeah.

Jeffrey Sherman: doesn't work. You you know, I I can go and ⁓ I can sit there and optimize the performance of things all day long. But if it's not the things that make the that matter to the customer, it doesn't really matter. Right? ⁓ and similarly, when you talk to operations, you talk to someone in sales, their you know, somebody actually ⁓ where the rubber meets the road, their desires are going to be very, very different. And their desires are going to be so much smaller than if you hear from

Isaac: Yeah.

Jeffrey Sherman: the VP of sales wants tech to do some major project so that sales can do some major thing. Like that's not something that Eutheseus ship. That's that's a major project. But if you go and talk to a salesperson be like, ⁓ what's your single biggest point of friction? And be, ⁓ you know, when I'm get a live one on the line and I do this, then I have to you know, somebody's supposed to follow up, but they don't always follow up, so there's no way f so then I have to remember to go check.

Isaac: Yeah.

Jeffrey Sherman: Which means so now I write this down so then every day I look at the list of things that I have to follow up. I have to follow up on our internal people to make sure that they you know, something like that. You're like, ⁓ I I could I could send you an email every day. You know, every morning, here's the the five things that you did, you know, here's the the deals that you got going yesterday that you should check up and make sure that they got handled by the next person right. Very small, very pointed. That will never come from the VP of sales.

Isaac: Yeah. Right. Yeah, those will never make it into stories.

Jeffrey Sherman: But it will be have a huge impact a and it will get those lines of communication going.

Isaac: Yeah. I think those things are super important. ⁓ and I think they're even being made easier, like I mentioned earlier in the age of AI where you can actually spin up some simple fixes to people's problems. ⁓ and sometimes they'll do it themselves. I've met a lot of pretty tech savvy non tech people ⁓ that or non developers I'll say that have already figured out, ⁓ I can actually just tell Codex to spin up a local cron for me that runs every hour on my machine separately from Codex. I'm like, ⁓ yep, nice, okay, cool.

Jeffrey Sherman: Mm-hmm.

Isaac: and it'll automate pooling data or whatever they need to do or sometimes even manually get new like like some like navigate to a a website and pull in, you know, any kind of new news articles or something like that. So I think we're in the age where that communication is getting easier and ⁓ it's gonna be harder to to n to justify not doing that to be honest.

Jeffrey Sherman: Mm-hmm. Yes. Well especially 'cause you have all this free time 'cause the AI is doing the coding. This is the kind of stuff where, you know, when AI is off thinking about a problem for an hour, go t you could go talk to sales, you could go talk to operations.

Isaac: Mm-hmm. So and maybe we'll get into this in the next episode, but I'm talking to somebody on the side that is actually trying to do a similar thing as we're talking about here with operations or meeting with people and saying, Hey, how are you using tech lately? Can we help out in any way? You know, is there an opportunity? ⁓ then just going to different organizations that are not traditional tech organizations, but could benefit from technology, especially now that we can write it faster for very specific situations. And just going, hey, how can we is there anything inefficient here? Like just show me your process. I'm just I'm here, I'm a tech specialist. I automate things. Show me your processes across your thing. And let me see if I can think, ⁓ wow, you just do it that way? I I'm pretty sure I could do it for a fraction of the cost.

Jeffrey Sherman: Mm-hmm.

Isaac: ⁓ so I think it's kind of a really cool business opportunity. I might have more to say about it next time.

Jeffrey Sherman: Hmm. Yeah, when you when you mentioned people using AI to do little one offs, ⁓ it recall made me recall on horror ⁓ I spent a decade on Wall Street, and Wall Street ha people have the their traders, they're fairly cle a lot of them are fairly clever, and even more of them have money. And y It was not uncommon to get called down to the trading floor and be like, hey, so and so has this really great tool that they wrote or that they paid somebody to write, and everybody wants it. So figure out how to roll out this thing to the whole company. You go and you look, and it's like a V B six. You know, th this is old stories here, so it's twenty years ago. But still in 2005, 2006, VB six.

Isaac: Mm-hmm.

Jeffrey Sherman: Yeah, and you're like, okay, it's like ⁓ o okay, and it what's it doing? ⁓ it's connecting directly to the database from everybody's computer. Yeah, okay. We can scale this, but it's you know, this is one of those ⁓ the revalid reasons to do a rewrite is totally inappropriate technology. I cannot have V V six connecting to directly to the production databases from every trading computer on the

Isaac: Mm-hmm. Yeah.

Jeffrey Sherman: Floor.

Isaac: This is vibe coding in early two thousands, it sounds like. Yeah.

Jeffrey Sherman: ⁓ five co this is what we were talking about last time in ⁓ with Dustin, where there was always somebody's nephew's cousin's younger brother who is a computer genius. And if you're dealing with Wall Street people, they have money where they're like, screw it, if it's gonna make me you know, I I'm gonna make an extra three or four hundred thousand dollars this year if I pay some kid five grand to write me a tool, they have no problem writing a tool.

Isaac: Mm-hmm. Mm-hmm.

Jeffrey Sherman: Even though you're like, you just want to search security policies. You can't do this. Like, I'm making more money, so now you need to deal with it. Yeah. That's a you problem.

Isaac: With the money. That's a you problem. ⁓ Great.

Jeffrey Sherman: ⁓ so yes. Leave it there of yeah, get out, look for the the friction, find it, get those conversations going and that's how you can start getting the buy in for larger transformations.

Isaac: All right. to your colleagues and see how can I improve your life today with technology.

Jeffrey Sherman: Yes. All right. Thank you all for listening. I'm Jeffrey Sherman.

Isaac: And I'm Isaac Askew, and this is Never Rewrite.