Getting Hands-On With Data-Driven, GenAI-Enabled Resource Management

​[00:00:00]

[00:00:00] Galen Low: Hey, folks, welcome to our hands on session on how to get started with data driven gen AI enabled resource management for your projects and your operations at large.

We do events like this once every month as a way for our members and our VIP guests to To engage directly with the experts who contribute and collaborate with us here at the digital project manager. Uh, for those who don't know, my name is Galen. I'm the co founder of the digital project manager, and I will be your host for today.

And I've also got with me an amazing crew of some of the top agency operations experts I know, Jessica Tiwari. Ann Campea, who will be with us shortly, but running from a meeting, just like any good project operations person, uh, Marcel Petitpas and Grant Hultgren.

we do have a few VIP guests in the audience today. So if that's you welcome, this is just one of a series of monthly sessions. We hold for our members who get access to a number of other benefits, including our entire back catalog of session recordings, our library of templates, resources, [00:01:00] and mini courses, as well as our flagship certification course, mastering digital project management.

Uh, you can join the fun by heading over to the digital project, manager. com slash membership. All right. We had a lot to cover today. So like that was me rushing through my intro and reading off my script apologies. And also you're welcome. Let's let's dive in. Um, today's session is all about exploring some practical, exploring some practical ways to use your project and people data to make better resourcing decisions.

We're gonna be using a little bit of math, a bit of software and a sprinkling of machine learning and generative Uh, and we're going to be walking you through it on screen so that you can see the concepts in action. Uh, so I thought maybe we could meet our panelists. I'll hold Anne's intro to the end. Um, but maybe we could start with Jessica Tiwari, who's the chief product officer at Parallax.

Jessica, you've been involved in digital products. Yes, I've been [00:02:00] creeping you on linked in. You've been involved in digital projects, products ranging from Upwork to Pantheon to Panorama and now Parallax. I'm just wondering what is it about data and digital experiences and education that gets you out of bed in the morning?

[00:02:15] Jessica Tiwari: Yeah, I think at root I am just a product person. I've as much as I would love to escape and have a different career. Like it is really just something that I've always enjoyed trying to find. the one thing that makes everyone's life a little bit easier. And, you know, one thing that works for everybody, uh, to our earlier conversation of why this is so challenging.

Uh, but yeah, it's, it's particularly been a joy to work with agencies because I love, um, when customers care about the craft, you know, like I love how. How much, how important the quality is, how important the experience is, how important the value delivery is to this particular audience. So that's been something that's been a real fun thing for me to follow around from product to [00:03:00] product.

[00:03:00] Galen Low: I love that. I love that you use the word escape, but like we need you so much that we won't let you out of product.

[00:03:06] Jessica Tiwari: My brain won't let me out of product. So, you know, we're in luck or doomed.

[00:03:11] Galen Low: Maybe both. Oh, great to have you here with us today. Uh, next up, why don't I go to Grant after that? Uh, so Grant, uh, is the head of customer success at Parallax.

Uh, and Grant, you've been all up in my feed. Uh, you've been giving talks and sharing your knowledge about agency operations and growth mode and the way that software can help. I was wondering if you could give us the inside track, like what is the most common way that smaller agencies are managing to still grow in 2025?

[00:03:36] Grant Hultgren: Yeah. Um, Yeah, it's, it's a tough one. Right. But I think at the same time, uh, most of our ICPs are actually these smaller boutique agencies. And the one thing a lot of them have in common. Yeah. And, and this part of my story too, is when I went into digital project management, way back when cutting my teeth, I brought like that relational AE experience.

And I think that is still so true [00:04:00] for a lot of our customers today. Yeah. They're problem solving. Like Jessica saying like, okay, I'm, I'm, I've got my product hat on today. They're problem solvers, but they're relationship builders. And so they're nimble. They're able to go and meet these customers who trust them, that they've built a relationship with and leverage that to build things that are unique, that are solving problems that are driving value.

And you're that like, don't lose that nimbleness. I know we talk about agile and stuff like that, but like that agility that you have as a smaller agency, not having thousands of people that you have to retrain at, like, Just embrace that. That is such a fun part of it. Um, and I really miss those days. So I'd say lean in there, lean into those relationships.

That's where you're going to find success.

[00:04:42] Galen Low: Solid answer. I love that. I love the sort of like the nimbleness, but like, yeah, being fast on your feet. Because you have that closest of relationships. Yeah, maybe there is hope. Uh, gonna move over to our next panelist, Marcel Petitpas. Uh, Marcel is the co founder and CEO of Paraquito.

[00:05:00] Uh, and Marcel, you also wrote a book. I know it's got not as much to do with agencies and resource management, uh, but I do love the title. It's called Software as a Science. Um, It's kind of all about unlocking, uh, limitless recurring revenue in the software space without losing control. Uh, but you collaborated with a bunch of rather big names and yourself in my mind, I think you're a big name.

Uh, what was that collaboration process like? And, uh, you know, how did it all work out? What was your favorite part about working on the book?

[00:05:27] Marcel Petitpas: Um, you know what, working on the book was a lot harder than I thought. Um, we went into it. Like we funded this project, uh, it was backed by Dan Martell. And so we had, we had the whole team, everything that we needed to make it as easy as possible on us, including the ghost writer and the editor and, you know, all of those things.

And, uh, unfortunately what ended up happening is that we had an extremely high standard. for the quality of the book. And every time that we tried to have a ghostwriter take our ideas and write them for us, we ended up 80 or to 90 percent to where we wanted to be. [00:06:00] So after seven rewrites of the book, two of us that were still basically like, like really committed to making sure this project, uh, went through rolled up our sleeves and basically rewrote every single chapter from the ground up.

And so we ended up getting very little leverage from that whole process, but I'm very proud of. what we produced. And my favorite part of that whole experience was it deepened my relationship with three of my closest friends. And I think we gained a lot of respect for each other because there was some moments in that project where it felt like we're in the foxhole together here.

And like, it's, we really have to grind this out. But all four of us were Completely unwilling to compromise on quality. Um, and, and what we produce is something I'm really proud of. And I think we'll be proud of for decades to come. Boom. I love that. It sounds like my relationship

[00:06:45] Galen Low: with chat GPT, where I'm like, that's pretty good.

It's 80 percent of the way there, but I'm going to, I'm going to rewrite it from scratch and completely decimate the ROI of using AI, but yeah, feel good about the quality. Love that. I feel good

[00:06:56] Marcel Petitpas: about chat GPT until it starts gaslighting me about lies. It's [00:07:00] telling me. And then I, and then I can't do it anymore.

[00:07:02] Galen Low: Well, feed me a better prop, Marcel. Uh, and the timing worked out perfectly because last but not least, I'd like to introduce, uh, and campaign, uh, and is the VP of project management operations at true sense marketing, and you've been showing us behind the scenes. I know you can't show today, but you've been working with your team to reinvent some of your resource planning and other operational SOPs.

Uh, what's like the biggest pain point you're solving right now? And how close are you to solving it?

[00:07:30] Ann Campea: Oh, I'm so glad that you flipped this question to me before asking it live. Uh, right now it's, it's really just accounting for kind of the different philosophies and how we look at this. There's the human element that I brought up when we all first got together as a gang and that, you know, my time for 40 hours a week is spent doing this.

X, Y, and Z. That's the human element of the person saying, I'm reporting my time as this, taking that, uh, [00:08:00] trying to remove the subjectivity, creating some sort of, uh, quantifiable data point that we can then translate into some sort of workload capacity, uh, tracker or balancer and just see kind of where we net out from that.

So it's kind of those two competing philosophies of. What's my data? I think Marcel had pointed out when we, when we originally spoke about doing a part two is there's no perfect data. I love when you kept saying that, but it's fighting the no perfect data. Portion and then trying to build something that can help quantify what we're doing from a resource perspective.

[00:08:37] Galen Low: That's such timely tension, right? This kind of like, dashboard with data and also humans, like, work in the interface. Actually, even just before the session, uh, Carol Osterweil posted on LinkedIn, replied to my post, and she's like, check out these folks who are doing human centric data. I haven't looked at it yet.

Sorry, Carol. Uh, but I think it's a really interesting, uh, idea. Um, I think we'll kind of dig in there today. [00:09:00] Um, and actually, maybe with that all being said, let me just. Tee this up. Um, first of all, I wanted to say because I didn't say it earlier for some reason. Um, this is the digital project manager. We do not feel like it's rude if you have a whole side conversation in the chat that can be its own experience.

You can talk about your favorite recipe. It doesn't have to have anything that to do with what we're talking about. That is fine. We might even jump in there as well. Um, so have at it. Uh, don't be shy. Um, okay, let me kind of like frame this problem. Um, because we've been referring to a previous session and we did.

Yeah, we did a pre, uh, a previous session late last year. Um, and we started this conversation around like what the enablers are for achieving a more intelligent resourcing workflow that takes away some of the finger in the air and hope for the best Hail Mary style techniques that we use when we're staffing a project or an initiative, if you miss that session, don't worry, you will still get the full benefit of this session.

And we will send a link to both session recordings after the event. But the bottom line was this, uh, it was that without data cleanliness, right? We called [00:10:00] that out without data cleanliness and without a healthy team culture around data and the way that it's used, getting to this, like, sort of Nirvana state of data driven gen AI enabled resource management was basically a non starter.

Um, it did leave us with a question, though, which was once you do have all the data pieces in place and the culture pieces in place. What does the workflow actually look like? What is the right mix of data, tools, and conversations that can make upfront resource planning and also those inevitable like resourcing pivots along the way, just a little less painful.

So, that's what we're going to explore today. Um, very loosely, well specifically, but also loosely, we're going to talk through three use cases. Um, first we're going to talk about Uh, how we can use some simple ish math, linear regression and some tools you probably already have in your possession to chart like historical data and make better resource predictions when you're estimating a project, then we're going to explore how to gather and just interpret data within your resource management [00:11:00] or project management tool so that you can notice trends and make some real time decisions when things change during your project.

And then lastly, we're going to give a sneak peek into what's to come as Gen AI powered agentic workflows start finding their way into our resource management practice. Um, tall promises. Hopefully I haven't over promised for my panelists here, but uh, I'm excited. So let's, uh, let's buckle up and dive in.

Uh, I thought maybe we'd start with just like upfront resource planning. Um, and like, one of the biggest influences to your resource plan is often your project plan, and in a lot of cases, that project plan is created when you know the least about how that project is going to go. But if you have good, clean historical data from past projects, you might have a fighting chance of mitigating risk.

Uh, that risk of being like really, really, really off in your resource plan. Um, Marcel, you were showing me some stuff. You schooled me in, uh, in, in algebra. You took me right back to my high school days. Could you show [00:12:00] us like the way that you help your clients, um, get to the state, um, and maybe just like your overall approach to leveraging historical data when you're doing some upfront planning.

[00:12:11] Marcel Petitpas: Yeah. Um, perfect. So I'll tee this up. Um, with just a little bit of context. So early we talked about the fact that I wrote a book. Um, we didn't quite mention was that what I do all day every day at parakeeto was help agencies measure and improve their profitability. Um, and we do that at parakeeto using a combination of three things, education on our framework, which really defines like, what should you measure?

How should you calculate those things? What do they actually mean if they go up or they go down? And how does that Actually influence your decision making number two is software that we built to help you actually get clean data, which is as I'm sure a lot of people listening to this is can be a big challenge, especially in that data spread out across a whole bunch of different tools and everyone is constantly messing up putting that data in a clean and consistent way, which is not a problem that's going away, in my opinion.

And the number three is some coaching and consulting so that when you're running these [00:13:00] reports and looking at the data, you have some help to interpret it and make the right decision. So that's what we do. That informs us. You know what I'm about to share and a lot of what we do that is different than, um, a lot of the project management kind of focused consultancies and software out there is we really focus on executive level visibility.

So the first thing to keep in mind is that a lot of our philosophy is to intentionally simplify and abstract things away from how they're often thought about, which is highly detailed. Task based estimates and individual assignments to people that get built up to these higher level estimates. Um, there's a lot of usefulness and validity to that, especially to project managers that are managing the day to day, but it can often be challenging to answer questions like what are the next six months look like if These 10 projects do or don't close in any degree of combination, it's difficult to work through those kinds of discussions using those tools often.

And so that's been our focus is to say, how do we enable those [00:14:00] higher level, you know, more ephemeral situations that are inherently uncertain? Right? So I just want to make it clear that I think the answer that we get a lot when we're like, Hey, well, What if we can't predict what's going to happen in the future?

Often the answer is we'll make your business more predictable. I think that's kind of a shitty answer. Um, right. Like, cause, uh, a, that's in some cases just not possible and be inherently business is going to come with a whole bunch of uncertainty. So our whole philosophy has been, how do we embrace and enable, you know, better decision making, more alignment among stakeholders with that being the case.

So. That's my rant. What I'm going to share now is just a simple idea that comes from having a very deliberate structure, uh, in particular in terms of how we estimate projects and then how we track time against those projects. And if we can do a good job of maintaining some cleanliness around that data, we can use, to your point, linear regression to estimate future projects.

So let me just share [00:15:00] my screen, make sure I'm doing this correctly. Here we go. So what I'm sharing here is a really simple concept. Um, it's a tool. It's an example of a tool that we could use to bring in our very structured data and say, okay, here's a list of projects that we've completed that are all a similar type of project.

Um, and so. For this particular client, it's like we want to look at all of our website projects that we've completed in the last little while. And we have some basic information about these projects. So what did we get paid? How many hours were planned? How many hours actually got spent? And then in particular, how much time on each of these projects was spent In these very structured role categories.

So this is like a big idea in terms of how we think about structuring data is it's fine. If in the project management tool and in the time tracking data, there's 600 variations of these four things, right? Designer, UX designer, UX design, uh, whatever you I design design team. But if we can have a process where [00:16:00] we're normalizing all of that to, okay, broad strokes, how much design time got spent, then we can end up with.

These really consistent building blocks across all of our different projects. And then we can start to build a linear regression model here that helps us understand the relationship in this case between budget and time and how those things scale. And in particular, we can start to see how that scales for each of the role categories that we've defined in this example.

So, um, what this tool allows us to do is say, in this case, we have a new client, they're coming in, they have 150, 000 budget for this website. How much time should I expect to need from the design team? And it says, well, 236 hours is what this linear regression model tells us. And I might say, well, this feels a little bit more like the project Jscope.

So I'm going to actually estimate 300 hours instead. And now I can see how this compares to those other projects as well as to the line of best fit. And I can start to kind of plan. And compare my gut feel assumptions or my, you [00:17:00] know, this client's maybe a bit of a pain in the ass. So I'm going to, I'm going to buffer this a little bit, compare that to real data.

The other thing that I think is interesting about this kind of visualization is you can tell a lot about where the risk lives in a certain type of project based on how far away from the line of best fit these dots are. So you can see here that with design. These dots don't diverge much from the line of best fit.

So that tells us that we have actually a very reliable algorithm here for how time on design scales with project budgets, whereas development surprise, surprise, there's a lot more divergence from the line of best fit here on the dev team, which makes sense. Development is hard. There's a lot of uncertainty.

There's unknown unknowns. And so there's a lot more risk here, and we can just see that based on how much more spread there is across a bunch of projects that theoretically had a similar scope. So we might want to price more risk in here. We want to add more contingency, and the data really helps us inform those decisions.

And all of this is really kind of the [00:18:00] baseline that we could use to build a machine learning system, for example, which I think for something like this, which is much more deterministic, a lot of people get ML and AI confused, but I think ML is actually a lot better for this kind of a use case because it's not as much of a black box where you're kind of like, I'm sure we've all had this experience.

We give a, an LM model or a large language model, a similar prompt, and we get a completely different answer every time, right? That's good for some use cases that are more conversational and more judgment based, but when we're trying to use. data and have a deterministic output, something like this is much more, um, suited to a machine learning algorithm.

And, uh, that is essentially where this is going to or what we're trying to work on with this kind of math. So that's me talking a lot about, you know, this thing that is actually quite simple. But I think the thing that you surfaced earlier, which is really important is this is only possible. If we have clean and structured data, and [00:19:00] ideally what this can start to inform is, you know, this wasn't on the script, but like, if we now are also thinking about our capacity in the same way, these broad buckets that are exactly the same of how we think about time, then.

Resource planning, not at an individual level, but just at this kind of broader, more abstracted level gets a lot simpler because we don't have to now go update 500 task assignments to individual people to get an understanding of like, Oh, our design team is going to be absolutely slammed in September.

We can much more quickly get to that high level insight. And it's usually something that we're doing much earlier in the project life cycle before we have any of the information really that we would need to go and do that more detailed resource plan. So, um, those are my 2 big takeaways is structured data can give you a lot of power to use that data to inform decision making.

But the other thing that's important is that the way that you. Predict things when there's a lot of uncertainty doesn't have to be very precise. And in fact, there's a lot of utility and saying, let's deliberately simplify this and make the edges rounder so that it's [00:20:00] quicker and more fluid. But we can still get a directionally accurate sense of where we need to be paying attention.

Um, so. That is it. I'm going to stop talking now. I like the rounded

[00:20:08] Galen Low: edges bit. So two things I love about this as well. A, uh, you are in Google Sheets. Um, and which is, you know, a tool that a lot of us are using or we're using Excel or something like that. It's not, uh, you know, fancy and hyper proprietary.

Uh, but also, Ann, I was thinking about you as I was looking at this. I was like, you know, I like this fidelity of data. So easy to get like way too deep into the weeds of like, you know, this five minute task, here's our historical data on, you know, 400 people who've done it in the past, and then you can't really cut it, you know, you can't really do anything with the data, but, and I'm thinking about kind of what you're doing with your team right now, like.

Uh, like, is this sort of how you're approaching it in terms of, like, team sort of departments, groupings of projects?

[00:20:49] Ann Campea: Uh, yes. In an ideal scenario, yes, this is how we are approaching it. I, uh, would also mention that we've kind of tried to sandwich our two [00:21:00] approaches, which is similar to what Marcel's showing, right?

Predictive, trying to just look into the future based on past historicals. But we've also done this scenario planning where. What does it actually cost in terms of what we're charging? What are the hours that we, you know, kind of working back for those that are familiar with this exercise? What are the scenarios that we would work back from in terms of what are you selling as a business to your client?

How long does it actually take? projected to take and then understanding how the fees bump up against the planned hours within your scope of work. It was one opportunity we looked at to see if we could reduce the human factor that we've all been talking about for quite some time, um, from that scenario planning.

So it's a combination of both. We've tried to kind of sandwich them both together to figure out what that perfect formula could be. And perfection, however, Marcel [00:22:00] and Grant wanted to find it. And I don't know how you guys want to define perfection, but somewhere in between all of that is where we're trying to find ourselves.

[00:22:10] Galen Low: I love that. Marcel, you have some questions in my chat and also here. Uh, folks are wondering like, is this a template that you're able to share with folks?

[00:22:20] Marcel Petitpas: Uh, it's not. Um, but if getting this kind of reporting system installed into your business is something that you're interested in, then I would encourage you to reach out to us.

Um, we do have a bunch of free tools that we can share with you. They're not exactly these, but, um, this is also something that we can work with you to get installed. Um, there is a lot of right, and this is my issue with a lot of, um, ML demos and AI demos is what you're seeing here is the end result. What you're not seeing is all of the work that has to happen to get here.

And there is a lot of work that has to happen to get here in order to determine what the structure looks like for your firm, clean up your historical data, make sure your systems are designed to create this data going forward. So, um, this is simple, [00:23:00] but it's definitely not easy. Unfortunately, that's true about most things.

And. Operations and project management, as you all know,

[00:23:06] Galen Low: that's, that's absolutely fair. And I do also like just like the, um, the categorization of information I think is useful because I think that's kind of a difficult place to start. And, um, without taking too much time on it, I did want to ask grant just from like, you know, you've been that operations guy at an agency, right.

And creating other agencies as well. Uh, I'm just thinking of that exercise of like, getting people, having the conversations with people to be like, listen, we're going to put your historical data up on this, on this chart. And that's how we're going to find that gray zone to estimate. Well, uh, how does that conversation normally go?

[00:23:38] Grant Hultgren: Well, I mean, right? Like I think it's a conversation that has to be had and using the data to have a structured conversation is absolutely, I think the best way to do that. Um, so that it's not filled with emotions or well, yeah, because the customer hit me with a scope creep, you know, like, and it's like, that's true.

All those things happen, right? Like, and that, like for our audience here, right? That is [00:24:00] generally not a digital project manager's fault, right? Like there are variables and like, Marcel started off saying like, we got to wrestle with uncertainty that comes with the gig, right? Like, and I had a conversation with another CEO is like, what do we do?

Like, what's the, what's the formula for solving this going forward? It's like your, your job as CEO is not to find the formula, but to wrestle with this daily and take all the new inputs you've got in a structured way to lead your team. And I would say the same is true for a digital project manager, anybody who's leading a project, like you don't control all the inputs and outputs.

Sometimes the dev is, it's going to take longer, but How are we managing the expectation against the values? And I think that goes like the human centric part that Ann's getting at. Like, what are we delivering? What are we focusing on? Yes, we want to like perform financially, but if the like customer doesn't like it, who cares, right?

Like you're going to lose the relationship. So it's just all of that. Like I, I like Marcel's approach because it gives us a structured way to have meaningful conversation. And to take action. And like, even [00:25:00] if we're really good, like AB test, can we sell dev differently? Can we sell design differently? Like those are conversations leaders need to have to set their teams up for success.

And so I think that's the important part. Like. That I love like seeing in his diagrams.

[00:25:16] Galen Low: Yeah, I love that. Like the job is to tackle the variability because predictability and perfection isn't really achievable. Um, in, well at least in this business. Um, I, I wanted to shift on to um, the second use case because I think that's really like, I can absolutely picture myself yeah, after having done all the massive legwork to get all the data in order but using that to sort of plan something up front.

Um, but then we were talking actually, this, this group when we were prepping for this, we were talking about Yeah, but like nothing ever stays that way. You can plan, you can have your plan. It's going to be perfect. And then tomorrow or five minutes from now, it's going to be out of date and you need to pivot.

Um, and we were talking about like how we can make real time decisions, um, because things happen, right? Maybe you're down a team member and you need to replace with someone who's maybe [00:26:00] not as specialized. Of a team member, or maybe you're just trending towards missing your deadline and you're going over budget and you need to reforecast, maybe you're just making the case to like add headcount to achieve a goal.

Um, I was wondering if I can get maybe grant and, and Jess together, just walking us through just like what it looks like to use a resource management tool and, and, and how you would approach mobilizing some of the insights from our projects, like recent history and use them as an input to, you know. The little and also really big pivots that have to happen along the way.

[00:26:30] Michael Mordak: Yeah.

[00:26:31] Grant Hultgren: Yeah. And I'm going to kind of lead off and. Listen, like I'm a project manager from my craft into like operations and try to approach like, I'm not a salesperson, so I hope this really doesn't come off this way, but I want to demonstrate how our tool parallax works relative to some of the conversation and viewpoints that even Marcel brought up and, and frankly, and too, in that, you know, Marcel is saying like, Hey, well, high level executives, like we've got to, yeah, we got around the corners, right?

Like we can't go and [00:27:00] diagnose every piece of data. And I would say the true is actually. Yeah. It's true kind of for us too, except for we're trying to go bottom up. So we're trying to say, Hey, actually like if we can get the granular data and we can get the people who are closest to the work, which is like how I try to operate, like I had a high degree of trust with my project managers.

If you think you can do this within 10 percent margin, like you have that flexibility to do that. Go and deliver great work. And I trust that we're going to be able to generate the revenue long term from the, the. Success of those relationships more than we ever would on the margin on a week by week basis, and I can demonstrate that with the tools and stuff that we had.

But, like, that's our standpoint of, hey, how do we empower those frontline workers who are doing the work who know it best, whether it's a developer, designer or project To correlate, communicate and understand that not every project survives, you know, first contact with a customer, right? Like, so I'll, I'll show kind of like base how our [00:28:00] tool works today.

And then I'm going to hand it off to Jess, who's going to talk about a particular use case where we think we can enable or even decrease the friction that we think is occurring today. Um, but I I'm hoping it kind of has a lively conversation. So let's dive in here. Uh, this is a new one for me. Let me Share my screen.

All right. So, you all should be seeing what we call Shaper. And so, when, um, Marcel is talking about, hey, these high level views of, of a project and how it all orient, or, or, uh, orients across the line of best fit, think of this as, like, one of those projects. We're seeing some red. We got some green. That's good.

I'm, I'm all for that. Like, that, that's good. You know, some stuff is going off the tracks here. And the goal here within our shaper tool is to help demonstrate both an enumerator, the granularity, um, which is our plan. That's what we thought it was going to take. Uh, then that's the project manager, maybe talking with the salesperson, sitting there saying, yeah, no, like, uh, let's take this top row.

This account [00:29:00] person, Nicole, like we've got them 15 hours across the project. Pretty standard, right? Like strategy, consulting, helping, but we can also see from the denominator that early on it, it, it taxed Nicole, she was going 22 hours over a 15, right? The youngling, and we're probably burning the project a little bit hot now, like that sort of degree of saying, Hey, that's a time sheet coming in.

And the biggest piece here that I think gets into like Anne's point about the values of like that human component is like. Nicole could have been like, no, I only spent 15 hours on that, right? When in reality, she's like burning herself out, right? Like, and she's spending 22 or she's spending even up beyond that.

And like the reliability of the data input comes down to how are we valuing and empowering those people to give us the accuracy of the data and also not like the stick for going over, right? Like we shouldn't be, that's a great way to make sure that all of your numerators and denominators are aligned.

And guess what? Zero value to me. [00:30:00] I, I have, I can't do anything with that data. So what we're trying to do here is demonstrate that, hey, okay, projects don't always go as planned. How can we then update them to make sure that we're actually accounting for this? And in this case, like I'm doing a really bad job.

I haven't really probably updated this all this much in terms of our plan. You can see that I've got, um, very like static amounts, you know, this one week down here, I. Dropped it to five hours. Why did I do that? Uh, when Sean, when 10 hours, you know, like there's maybe some conversations happening there, but it's hard.

Like I could be in Jira. I could be in Asana. I could be in monday. com trying to say, man, look at all these tasks that I'm trying to juggle this team around with all these different roles and they're saying, oh, it takes five hours over here and 10 hours over here and. Do I have all my tasks actually listed out?

Have I estimated that all should I estimate that like all this stuff? I think that your digital project managers are like. Yeah, every Monday, right? Like, looking forward to my scrum meeting. Gonna get [00:31:00] that information from the team. I hope it's accurate this time. Um, and so, you know, like, those are all inputs into this tool, but we found that when you focus on the ICs and those individual contributors, give them and empower them to own their projects.

You can start to actually make incremental changes along the way that start demonstrating how a project is trending. If it's trending over, great, there might be an intervention. If it's trending under, are we delivering value to the, the contract that we have? And that might be a different kind of intervention.

But really empower those people to, you know, have that narrative, while also understanding like, that's a, that's a lot for our, our project managers these days to like juggle and to own. And I think, like, I'll pause there, because that was a very brief kind of, uh, uh, like, Hopefully not too, too much, but, um, look at Shaper and there's obviously more to the tool, but this is the core of it where we're really trying to empower folks, um, to, to [00:32:00] plan effectively, but like, yeah, if there's questions or Galen, like happy to, to, to point out anything here further.

[00:32:06] Galen Low: Now, what I love about this is like, because we used to do this as well, we used to say, you know, account manager or project manager roles get a percentage of the budget. And it's just like whatever X number of hours a week straight across the board, which, you know, as many people know, is not exactly how the role works.

Um, so like that Nicole example, what I love about it is that, you know, it's this opportunity to have a conversation where Nicole can say, yeah, but beginnings of projects are always heavier on the PM, you know, and then at the end and in the middle, we're going to make it up. And then like, my question to kind of grant would be like, cool.

And then like, can we like look, uh, across projects for a certain role? And like this kind of similar view to kind of see where that smile curve is and if that's consistent. And then we're like, okay, yeah, now we can like make a case. Uh, and yeah, no one's going to get the stick and no one's going to get punished for this because we understand the work better from what happens in the front lines.

And we know that it's like, pretty much consistently [00:33:00] the case. Um, and it might not even be worth then like adjusting that because it was going to take you an extra hour or two to be like, okay, well, let's, you know, get really granular about the PM tasks at the beginning of a project, or you could just kind of, you know, to Marcel's point, like round out the edges to be like, yeah, on average, it kind of like it averages out, we know it's going to be, we have a tolerance of being this much heavier at the beginning and this much heavier at the end, as long as we do this in the middle.

[00:33:24] Grant Hultgren: Yeah. And what I would say to that to Galen is What I appreciate about that and like even those instances is like for us, the way I kind of like talk to our prospects or our customers that are coming on board is like, listen, your organization is a house and let's just imagine that we're trying to build that brick by brick.

Every project is that brick. And if you don't have that data, guess what? It's really drafty in that house. So like, that's the importance of like getting all these projects built up so that we can have a really complete picture. And some of those bricks might be crumbling and that's just reality. Like, and we got to fix it and that's okay.

Like, but [00:34:00] now we know what we need to fix, like, and that sort of like, okay, now let's actually do something. And I think like. Man, it's such a privilege to get to a point where I could be a COO of an organization because that's a total overhead role, right? Like, I would try to help sales and I would try to help deliver, but like, man, I was a cost burden.

Like, how am I actually proving out as a leader that like, I'm effective at this and like helping the organization? And to me, it's like, great, give me the problems then. And like, help me help a project manager who's like, well, I don't know what to do. Like, I know how to run this project well. Great. Then I'll go talk to the customer and be like, you're way out of line on scope.

Right? Like what, what's the 10 percent that's the hardest that I can do to like help them do that 90 percent really well. Cause that's when deliveries like, and people align and I just think it gets a lot easier. But, um, yeah, that's that's what the tools meant to do doesn't always work out that way, but that's that's what we're trying to do.

[00:34:54] Galen Low: I love that perspective. And I love that it's called shaper because that actually is kind of what it feels like as you go through a project [00:35:00] need to kind of like shape your resource plan. And and I remember you showed me something you you internally, you guys are using work front, I believe. Is that right?

[00:35:09] Ann Campea: Yes, yes, we are. It's a component within work front called workload balancer. Um, this is a lot prettier. Don't talk to me. I said that. Uh, but yes, it kind of gives us a similar flavor of you're able to actually get into the projects themselves and what tasks are driving that time by resource. So that's 1 of the things we like is that connectivity between the projects, but.

As Grant was highlighting, I think at the core of it is looking at how things are balanced. And from a project manager perspective, my expectation of my team is that they don't have to know all the nuances of how this was built or what's going into it, but they need to know how to read it. So I love that.

This is, you know, very obvious, deeply read. You're in trouble. I'm guessing green. You're pretty good when it comes to capacity

[00:35:59] Grant Hultgren: a good old [00:36:00] stoplight, right? Yeah

[00:36:01] Ann Campea: Very simple. I love it.

[00:36:04] Galen Low: Oh, that's excellent. I am Gosh, I could spend days on this. Thank you for showing us this grant. Yeah, I wonder though just looking at time I wonder if maybe we could peek into the future because, um, you know, I, I, I often, this is sort of my agency experience, right?

That dashboard view in whatever flavor we built ours in Google Sheets. We almost broke Google Sheets as we cracked the like 120 person mark. Um, but yeah, similar traffic lighting. Uh, and then what we use it for is to like drive those conversations. And, you know, we'd have that traffic manager or that resource manager and her full time job was.

Running around 40 hours a week, sometimes 60, talking to each individual person, doing the work, the project manager is trying to figure out how to like workload balance, how to make all the puzzle pieces fit. Um, and I kind of started this out, this whole series, really, I opened it with the idea of like, okay, well, how can we like make that a little less like painful?

Because it's a lot of time talking to people. It's a lot of puzzle pieces, uh, puzzle pieces fitting together. [00:37:00] Um, and it's, you know, it's, it's, it maybe could be easier. Um, especially in this world of like, yes, you know, machine learning and generative AI and some, some of the agentic stuff. Um, so I thought that maybe we can look at, um, get a sneak peek really at a prototype that Jess has been working on.

Um, maybe I'll just preface this because, uh, I know a lot of you have heard the word agentic being thrown around and I was that person who like was late to the party and I was like, I thought it was someone's name in conversation. I was like, oh, okay. Yeah. Yeah. Mr. Agentic. Um, but basically this idea that like.

Our AI tools are becoming more like proactive agents within our team, rather than being like reactive, prompt based chatbots. Um, some of us have been told that agentic is what's going to take our job. Uh, but I thought maybe we could paint a slightly more optimistic picture. Um, and Jess, I wonder if I could just turn it over to you to kind of like show us what the future could look like in terms of resource planning.

[00:37:58] Jessica Tiwari: Yeah, absolutely. And I think [00:38:00] you're right that there are a lot. There are a lot of places like many, many where machine learning is can make a huge difference and AI can make a huge difference and agent like agents of all forms in our space. There's just so much opportunity because there's so much time spent and friction and manual everything that we have to do to keep these projects on track and on budget and on time.

It's just extremely complicated, and I loved that, you know, everyone's sort of, um, going after that, that panacea of I really want things to be more predictive. They never are. Like, I really want to be able to. Um, to be able to see the future more clearly, but as we've all already discussed, like that, really to do that well, even a little bit better, you need that foundational data to be clean and accurate.

And so we decided with our 1st agent to sort of tackle that root problem in hopes that [00:39:00] if we can even make that planned data as a project is progressing, even just a smidge more accurate. We're then getting ahead instead of, like, having an entire project and have that entire project do inform how I shape the next one or how I plan that or even sell the next one.

I can hopefully fix this 1 in the middle, which is really what we want to do and something that's very difficult to do. If you know that you're ETC hours. Are inaccurate, and this is a very common problem because I think people it's friction. People don't want to get into any tool and update planned hours.

They just want to do the work. Uh, so we're trying to make it as easy as possible for them to do that. Um, and right now, I think all of us are probably familiar. There's like, a million ways that people are trying to do this manually today. There's like, the weekly meetings, there's slacks, there's updates.

PMs are just spending an enormous amount of their time. Talking to all the team members about you on track, you're not on track. And like, let's, and getting that manually back into whatever system you're doing to, to track these things. Um, so I can share [00:40:00] sort of, I got like a little overview slide here to talk about what we're doing.

And then I also have, uh, the prototype itself that I can walk through with you. This is a prototype at the moment. It is not the cleanest thing in the world, but what we've done is we've, uh, built up our first agent. And it is, uh, designed to go out to the I see team members on a project and basically just on a periodic basis, we chose weekly for now.

People can pick when they want to get it, uh, to ask them, like, how here's what you're planned for for the next 8 hours or so, um, does this still look right to you? Uh, so the things that it can help you with are like an overview. Here are all the projects that you're planned on weekly breakdown. Here's all the hours that you're planned on those projects.

It can do a little bit of hours balancing. So if it notices that in a month, you've got a 60 hour. Uh, week you can, and you've only got 30 hours in the upcoming or 20 or 10, it'll suggest that you move some of that time over, uh, it is attached to an LLM behind the scenes. So it can also help you with, like, [00:41:00] the world's information.

So it can do estimation assistance for you at a very general level. It's not super specific right now, but it's, but it knows a whole lot, like, even chat, if you type in, you know, how do I, how am I going to break down this work? How much should each of these things take me to do? It can provide some baseline assistance with that.

Um, and then it's hooked back in this case to parallax, uh, so it can also help you update those hours. If you think that you're going to need more time in the next four weeks to do blah, you just let it know how much time you need, what the reason is, what weeks you want to do that time, and it'll help update those back to your core central data system.

Um, it also helps with impact assessment. And then, because we don't want to cut the PMs out of the loop here, this is very important. These are their projects. All of those updates are going to go through approval routing to make sure that people are comfortable with the, with the changes that are being suggested.

And then, just to show real quick, kind of what it looks like today. There we go. Um, so this would be something [00:42:00] that you just log into using your, um, your email or whatever system you use to log into your core, uh, parallax account. Um, and it knows who I am, it knows what projects I'm assigned to. Eventually, it'll, it'll know more about me, more about my skills, my, um, department, uh, things that will help it with that estimation assistance or that predictability.

Um, but today it just knows basics like how many hours I'm planned. Um, so I can say, what projects am I planned on?

It also can deal with a church's typing, which is very important. Um, I've tried some real doozies on this and it's been pretty, uh, pretty flexible. Um, so I'm going to say maybe, I think I might need some more time.

And that's not, that's clearly not enough information to actually submit a request for more time. So it's asking like, why, which weeks are you going to need? [00:43:00] Um, do I have any plan time?

Let's find out.

I've got 20. I think I need another 10.

So, let's say the reason is I need to make some wireframes. I think it's going to take me an extra 10 hours from what I've currently got planned. It's learning all of this. Figuring out what to do with it. Okay, I'm adding another 10 hours. Um, it thinks it's January right now because beta test. Uh, and then it, it is telling me like, Hey, you also have 20 hours planned.

Do you, that's like a pretty high workload. Do you want to move that to another project? Do you want to move that to another week? Um, I'm going to say no, let's stick with this. Let's go ahead and submit. This [00:44:00]

[00:44:02] Galen Low: is so cool. By the way, I feel like I was born one generation too early to be a project manager or resource manager like this, this thing.

[00:44:11] Jessica Tiwari: It's pretty, we actually hope, so there's a lot of different reasons that people aren't updating their planned hours today. 1 of a lot of them have to do with just unfamiliarity with the tools that we're asking them to go in and use 1 of the things that's nice about these chat based agents is it's just conversational.

There's no, you don't need to have familiarity with anything. And there's a lot sort of like. Lowest common denominator to what future planned hours take, what project information is needed. Like, you can kind of abstract away from parallax here and just say that. You know, it wouldn't really matter what system was behind this.

This is going to help them get the information that they need and get us the information that we need to help the project go back on track. And then it's going to go ahead and submit that over to the project manager along with the reason code, and they will be able to say whether or not to prove that [00:45:00] time.

So what we've got now is, um, an actual reason, which I think will really help us with predictability in the future every time I need to change planned hours, rather than some of the other solutions, which are like, why don't you task out absolutely everything in the world that you have to do on this project, assign estimates to each of those specific tasks, and then you can track each task.

That's really unwieldy and very time consuming and unlikely to be something that people will be able to keep up forever. This just puts key reasons on the reason that things change, and that's really going to help us with the next project, the project after that, and with any sort of future machine learning, predictability, or, you know, agents on the PM side, um, to be more informed.

So that's what we've got right now. I'm really excited about where it might go. It's obviously like a pretty, pretty. It's not very smart yet. Like, to be frank, there's a lot of work that we need to do there. Um, but it's, it's um, I think pretty promising and could save our PMs an enormous amount of time once we can get it tuned.

[00:45:58] Galen Low: And I love, the other thing I love about it is [00:46:00] that it's kind of a conversational training for like that individual contributor. To like better understand their work and how long it takes. Oftentimes we go and we ask our team to like provide estimates. We've never trained them to estimate. We've never actually sat down and been like, okay, well, like, yeah.

And like, how, how, how big is the work? Like, what does that look like? And it's okay that it's different for, you know, you or someone else. Because now we can plan. We have visibility and grant to your point. We're having the conversation. No, one's getting the stick. We just want visibility into how we can solve some of the challenges in our business.

And I was thinking of you too, because I'm like, you know, this finds that balance, I think, right. Between it's human ish. Yes. You're chatting to a chat bot, but it's like transmitting information. And I think it's. fostering a dialogue is kind of what I see. It's not like great. You get that time. I'll just, you know, inform the project manager that they're going to go 400 hours over budget.

Um, bye.

[00:46:54] Ann Campea: Absolutely. I really gravitate. Thank you, Jessica, for sharing that. I really gravitated to the concept of you're [00:47:00] asking people who are not trained to time estimate to time estimate. That really is a sticking point for a lot of us in this agency type work where we're trying to do this type of estimation is I don't tell my boss.

I have the time feel like I'm guessing because we don't simply have the time to create the accurate time estimate. So I love kind of the full circle that of what Jessica showing and to your point, Gail, in the conversation that it. That, uh, brings up.

[00:47:30] Galen Low: Yeah, the goal is getting better at guessing. Maybe not perfect at predicting the future.

No crystal balls, but better at guessing. Um, yeah, thank you for showing that I was like, man, I feel like this is the Wizard of Oz and peek behind the curtain. I'm seeing how it all works. I know it's early stages, but I like where that's going. And you can just imagine, um, like the sort of Um, being, uh, being like a voice chat as well.

Um, you know, where the technology is at to be like, cool, good morning. And you're just kind of like making coffee, [00:48:00] chatting with this thing. Um, and then it abstracts that away and actually takes away a lot of that. So the politics and the intimidation of having that conversation with your project manager.

[00:48:09] Jessica Tiwari: Yeah, that was actually one of the things we heard. I mean, no one likes those conversations. They're, they're a little confrontational. They're distracting you from the work that you'd really love to be doing. Like you didn't. Join this agency so that you could do this, but it is. It's such an important part of what you have to do.

And if we can make that easier by kind of like, removing the emotion, like, this is the 2nd or 3rd time that's come up in this, um, in this chat. Like, it's, uh, I think it, it, it does that for people and should make their lives a little easier.

[00:48:34] Galen Low: Like it's so cool. I can't wait to see it as it rolls out. Um, looking at the time.

Thank you for my panelists being wonderful at helping me stay on time because I was way behind at the beginning. We've got about 10 minutes left. I know some folks may be peeling away to their next meeting. Um, if that's you, I just wanted to say thank you for joining us today. We are going to do some Q and a, uh, with this last little bit.

Um, also wanted to [00:49:00] mention that if you are loving this, uh, I'd love to see you at our next event, which is, uh, about debunking myths around. Profitability through automation and productization. Uh, and that panel is going to include Melissa Morris, uh, from agency authority, uh, Joe DiPaolo from Osello and Brian Kessman from, uh, Lodestar Agency Consulting.

You can RSVP at that link that Michael just posted in the chat. Um, and again, if you're a guest today and you want to continue this this discussion. Um, consider joining our membership. Um, you can check us out at the digital project manager dot com slash membership. Um, and maybe let's get into some QA. I saw some topics kind of go through the chat, like especially around work front and things like that.

Um, And, uh, let me just double check and see kind of what questions we got. But if you do have any questions for the panelists of what you've seen today, feel free to pop them into the chat and we will get to as many as we can.

Um, I have this one question [00:50:00] here and hopefully I will do it justice, um, which is, uh, what are your respective views to the panelists? What are your respective views on how to, uh, on how buffer plays a role in this, um, At the project planning and execution stage, um, and if I'm understanding it correctly, it's kind of like, yes, like rounding the edges, knowing that we can't be perfect.

So let's just like build in some extra time here and there based on what we know. It's

a

[00:50:29] Marcel Petitpas: great question. Um, I'll start by addressing it just like from a profitability perspective. I think that one of the Key issues that we see in a lot of agencies when we first start working with them is that the way that they have architected their estimation and pricing process creates an intrinsic link between time and money.

And so it makes this conversation far more challenging than it needs to be. And so I think that the biggest idea around this is like, absolutely, that's a thing that you should [00:51:00] do right contingency, you know that there's risk, you should build that risk in and I think it's critically important that the document.

And in some cases, even the way that you represent the numbers around a project should be distinct in terms of what you're using to inform the expectations internally at the agency about like, what do we actually think this project is going to take, which is often completely different. Then what you're putting in front of the client to say, Hey, this is how much it's going to cost and the scope and how that scope is being defined.

And creating one document that tries to do both of those jobs is, is a massive trap because it often creates the situation where you're like, well, we can't lower the scope. Unless we lower the number of hours, but the scope hasn't actually changed. It's just that we're giving the client to discount, but the way that we've got this system set up, like we can't really discount things or it's considered a bad thing if we discount things.

And then six months go by and everyone's like, Oh shit, we went way over budget on this project. But in reality, it's just that we lied to ourselves about what we actually thought it was going to take. And we forgot that that was the case. And now this feels like a bad [00:52:00] situation. And so, um, that doesn't speak like super directly to buffer to me.

It's like, yes. Do that. That's a good idea if you feel that there's risk. But I think what makes it unnecessarily complicated is this broader issue of there's not a very thoughtful first principles process to how we arrive at assumptions about projects and then present those to clients.

[00:52:20] Galen Low: I also like Greg's note in the chat about like buffer, like over buffering.

Like if everyone's adding buffer, then it's going to look like it takes 200 hours to like, you know, type four words on a page. Well, yeah.

[00:52:33] Marcel Petitpas: Oh, go ahead Marcel. I was going to say the risk with that is you end up in a situation where you've over buffered the project and then you have a bunch of unutilized time.

And so there is a balance to be struck with this. And I think this is where again, having a very honest process around profitability is, is a forcing function where it's like, well, if we ask the client for 500, 000 for this three page website, they're probably going to say no. So it, like, it does force you to kind of reconcile the reality of what you can charge with the [00:53:00] reality of what it's actually going to take.

Well, and I think to

[00:53:02] Grant Hultgren: add to that, right? Like. It's like a deadline. Everybody goes up against the deadline. So if you buffer and you build out that budget, guess what? You're going to use the budget. Like, or you're going to end up in a worst case scenario. Like where you were like, did we did, did we deliver the value?

Like, Oh gosh, should we like do a little bit more to me? Like the way I, it's like estimate what you think it's going to take, have confidence in that. If you're really think the customer's a risk and you like feel like that, you know, they're going to throw a scope in there, like give them the 20%. It's like, we won't touch this.

If you don't touch the scope, right? And it's like, that way, it's like, we can go above, and like, that's sort of a buffer, but it's sort of contingent on them. And it comes down to, yeah, like, yeah, how well do you scope? Are you all in alignment? Of course, like, those things need to be, you know, hashed out, of course, but like, That's the part where it's like, there's ways to do that and make sure that you're not under delivering or you're getting Ripped off in your organization, especially like smaller companies.

Like you have to stand up for yourself. [00:54:00] Typically overscoping for the sake of overscoping or like buffering. I, I just, I always struggled with it because it's their team's going to use the budget. Like they'll, especially if it's time materials, I think we're moving away from time materials in general.

But, um, yeah, it's something we'll, we'll have to work out. It's so

[00:54:15] Galen Low: funny. Cause it's like thematically, you know, we're trying to get cleaner data, but like first line of clean data, like, as you talked about, is that honesty, right? Like if we're not being honest and we haven't created a culture of honesty, you won't even have that beginning data, that seed of data.

To be very good,

[00:54:32] Marcel Petitpas: this is up to me. One of the most important things that undermines data hygiene actually at a first principles level is not being very thoughtful about the incentive structure that the way we talk about certain metrics in the organization actually influences the way that people track their time.

And so like. agencies that are very disciplinary around going over budget, they'll find that people start lying about how much time is actually going into projects and they don't understand how long things actually take because people are hiding that time. [00:55:00] Organizations that are focused on utilization and over index, they're going to see that projects go way over budget.

It doesn't matter as long as people are hitting their utilization numbers like you. And it is very nuanced, unfortunately, it's simple, but it's not easy, but those things matter a lot, the executive level focus and conversation happening in the business.

[00:55:19] Galen Low: And that's also what I like, Jess, about what you showed as well, where it's kind of like, it starts conversationally as sort of, you know, talking about it.

And the more we talk about it, and think about our rationale for it. You know, the less it is about like, I don't know, maybe just add in 10 hours, just cause like, cause, cause I think that someone's going to come at me with something, but also having a reason. So you can, and then like, and if I, if I was thinking at what you're putting down, Jess, like that they have reason codes, right.

So that you can actually Run reports and be like, Oh, yeah, most of the time, like this category of thing is the reason why people need more time. Um, and actually maybe it isn't the client that's, you know, pushing for additional scope. Maybe it's actually just we, we keep underestimating [00:56:00] this thing because we're like, yeah, you know, how long is a piece of string?

Yeah, this long, you know, every time. Uh, and, and, and maybe never is,

[00:56:08] Marcel Petitpas: I think this touches on. An important, I think, message to close all of this out, which is it's very easy for us to get focused on data and we all love data. I'm wearing a hat that says the word data on it, right? Like data is awesome, but it's really important that we remind ourselves that it's actually not the point.

The point is alignment and clarity and confidence so that teams can make decisions, be aligned on those decisions and get things done. And the data is really only there to help facilitate that outcome. Um, and so. I just want to articulate that that's that's really what this is all about and let's not lose sight of that important outcome in all the excitement of the data that we can play with all these incredible tools that we have access to to help us with that.

[00:56:54] Galen Low: And that's like. That's the like lovely, optimistic bit about it. Maybe disappointing for some folks [00:57:00] who just thought, you know, the data was just going to walk itself in, pour itself a cup of tea and fix all your problems. That starts actually with, with us and especially like the demos, right? The tools support and augment what we're doing for now, but it's, it's still augmenting our intelligence, not stepping in and grabbing the wheel.

Necessarily. Huh. Well, this might be a good place to leave it. Um, I just wanted to say thanks again to everybody for another amazing session. I would love your feedback. Michael posted a link to a feedback survey. Maybe we'll post it again for good measure. Um, but this is the first time we've kind of done a session where we're like show tools and like, you know, kind of walk through use cases.

We love being practical. Um, we know it's, you know, it can be, uh, rushed and, you know, obviously I'd want to spend more time, but you know, time is that commodity that we haven't talked about a lot of these days. Um, so feedback. All honest, raw feedback. Welcome. [00:58:00] Um, uh, bring it. Um, and with that, I will say thank you very much for joining us.

Of course, a big thank you to our panelists for volunteering their time today. Uh, this was a lot of fun. This was a really, really good conversation. Um, thank you for sharing and everyone in the chat. Thank you for sharing your insights as well. Um, hopefully we'll see you all at the next one. Thanks for having us back.

Yeah.

[00:58:24] Marcel Petitpas: I'll also just I'm going to plug real quick that we're hiring a junior project manager. So if you or someone, you know, uh, is awesome and wants to work at a company with a very, I don't know, uh, intelligent, charismatic, uh, humble CEO, then, um, be sure to check out parakeeta. com forward slash careers.

Thank you

[00:58:42] Jessica Tiwari: with excellent taste in hats. Pays

[00:58:44] Galen Low: in honey. Yeah, the salary is actually just honey from Marcel's beekeeping, uh, foray. All right, I'll let y'all get back, uh, to your days and, um, we'll chat with y'all soon.

​[00:59:00]