Why This Family Office Built Its Own Hospitality Tool

Francois Botha: Household management, guest arrivals, servicing, and running a smooth ship. Today we hear how one chief of staff built a single place for all of that in one day. Welcome to Family Office Builders, a new series from Simple, where in each episode, Someone tells the story of something that they built using the latest AI technology, either for their own office or for the family offices that they work closely with.

We cover the problem, how they thought about the solution, and the journey that they took to find it. And then see what came out of it. I'm Francois. I'm the founder and CEO here at Simple. And Simple partners with family offices on everything from starting up to professionalizing. That covers strategy, direction, operating models, and technology. And of course, today, a big piece of it is AI.

Everything today reflects our guests' own experiences, and Simple does not endorse any product or provider mentioned, and nothing in this session is, of course, investment advice. Today, I'm joined by Alex Sadowski, Chief of Staff at Pardon. Alex built a campus and residence management portal for their hospitality team, one central place for every team member, guest, and every maintenance request. Alex, welcome here with us today.

Alex Sadowski: Hi, Francois. Thank you for the lovely introduction. It's a pleasure to be here with you.

Francois Botha: Alex, I've had the pleasure of seeing some of what you've built in person before and really interesting, but maybe can you give us just to kick us off a little bit of a background on just you and your role and how did you end up being responsible for this build?

Alex Sadowski: Sure. So I am chief of staff here at Pardon and I worked very closely with our principal in discovering kind of what vision it is that needs to be transcribed and translated into actionable items that the team can work on and kind of move forward. And Part of that multifaceted role beyond just working with the venture side, studio side, it's also hospitality.

We have quite a few physical spaces here on campus in Newport Beach, California, and for us, How we intentionally manage those spaces and ensure that they are always in tip-top order, ready for guests VIP arrivals and just really representing the family office at the highest of levels was an increasingly important function for us. And so for us, it's always been a very hands-on kind of person-oriented job of kind of managing the spaces.

And so for me, it kind of dawned on me one day that, especially with AI, tech could really provide a wonderful solution for us to kind of elevate our team's roles in managing the spaces and ensuring that they're taken care of.

Francois Botha: Maybe can you tell us a little bit more about the spaces and also just who you built for within this project?

Alex Sadowski: So our home is in Cannery Village, which is a neighborhood within Newport Beach, California. And we have a couple of spaces on campus here. So we have our Wave House, which is our main HQ. We also have what we call the Clubhouse and the Treehouse, which are two interesting spaces because they both have Residences upstairs, and then they have, um, officing and gallery spaces beneath. And so really we have about six or seven discrete spaces just here in the village.

In addition to two other spaces that are, that are outside of Newport, but still considered local. And so. For us, we have a team that kind of oversees hospitality, meaning the care for the spaces, any maintenance, repairs, ensuring that the spaces are turned over if there's guests that come in. And so we do have to kind of operate at a very high level of efficiency, both on the guest side, but then also in the gallery and office spaces beneath them.

Francois Botha: Amazing. It sounds like you've got a lot of people coming through. Can you just tell us a little bit about the size of Pardon? What's the kind of company size? How many people are we talking about here?

Alex Sadowski: So across Pardon with multiple ventures and arts and lifestyle, we are a team of about 40 people. We use the spaces to allow team members who are visiting to stay in them while they're here. We also have VIP guests of the family office or visitors who we may be working with also come and stay in the spaces. So it's very common for us to have Both of our residences occupied any given week.

For us, it matters a lot to ensure that we can very timely take care of the spaces, especially between guest visits, if there's certain repairs that need to be done or simply just overall cleaning. And so as seen In the demo, one of the great things of the portal I built was a calendar view and specifically by day, it will highlight who is entering or staying with us for how many days and then who may be departing on that given day as well.

And that's probably one of the most critical things because it can then allow us to schedule around and ensure that the spaces could be attended to.

Francois Botha: Amazing. I think before we get into the demo, and we definitely want to see what you've built, maybe let's talk a little bit about the problem. So often as we build new tools and solutions, it starts with some kind of problem. What was that one problem where you thought, hold on, maybe we can build something custom here to solve this problem?

Alex Sadowski: It's definitely a great question because I would definitely say that was the kind of light bulb moment for us where we were using several, I'll say tools to manage the spaces and by tools, I mean spreadsheets and calendar entries, notes, shared documents for procedures, and it functioned fairly well for us.

However, as spaces were added or transitioned and as the team kind of shifted a bit, It then really highlighted for me that this was our moment to build something that was fully bespoke to the family office and that would really pay service to the spaces.

And so for us, I also wanted to ensure that I wasn't going to build a tool that was more challenging to adopt, given that all of our teams are very used to these more simple kind of key tools from Spreadsheets and Calendar, I needed to create something that was going to integrate very easily into all of that. And so that was probably the biggest consideration throughout scoping this tool was to ensure that the barrier to use it effectively was going to be low.

And therefore, the team would actually want to engage with it and allow it to be something that would elevate their role.

Francois Botha: So you mentioned also the low barrier to entry, working in the spaces where the team already works. I think those are all very relevant considerations as you're looking at different solutions. Did you look into the buy side of this at all, like buying a solution as being an option? Because I guess there are some solutions out there that could be bought that maybe handles these types of needs, or was that not really a consideration for you because of how the team worked?

Alex Sadowski: Most definitely buying is certainly a great avenue for tooling like this. I think for me, and this highlights one of the things that I particularly love about AI is for me, it always starts with the scope, just sitting down after speaking with the stakeholders, in this case, the principal and a little bit with the hospitality team. After I created a scope of exactly what we needed before I even went to look what tooling existed.

I personally love to just play with AI and see what kind of very fast prototype it can create, which will then inform on, do I continue to build or do I look into a solution that could be acquired? And so for me, that is one of the great kind of pleasures of working with AI is it kind of allows you to become a maker very, very quickly and easily.

And once I kind of had a clear image of what V1 of this wanted to look like, it became really easy for me to just start playing with the tooling and create something very quickly, a wireframe, if you will. In this particular case, once I developed that wireframe, I almost couldn't stop. It was very clear that that was going to be a pathway that would yield a tool that we could use. And for me, I then kind of set my course on finishing the V1, and I was extremely happy with the result.

Francois Botha: Maybe let's rewind, Alex. Before you had this tool, can you give us a very practical example of something not working the way you'd like it to or being a little bit difficult or breaking down?

Alex Sadowski: Yes, for us, one of the unique things with our team is we have a team member who prefers communication in Spanish and things to be physical. And for us, this was a great example of how AI could better accommodate us than an off-the-shelf tool. Because I was able to build into this tool an actual function that would create a one-pager report That I can change the language in and it would turn it into a PDF and I could print it and then provide it to that team member.

And so for me, this is a great example of something that I had a solution in mind that was certainly not conventional and something that would perform better for our team. And it was a great use case of something that AI could build that would actually benefit us.

Francois Botha: Okay, perfect opportunity to maybe take a look at some of what you've built. You've prepared a short little demo. Should we take a look at that and maybe you can just walk us through a quick overview?

Alex Sadowski: Yes, that would be great. I would love to. So in preparation for this, I also had this demo built It's fully on our website that I've created that's fully live for my team to use. And so it always lands on an overview page before going down the menu.

Work orders is probably one of the more critical pages, I would say, because it actually It allows us to not only see what's kind of open in progress, what's been marked as urgent, but it lets us really get into the detail with what's the space, what needs to be done, what's the level of priority. If you have any reference photos, you can add them and you can assign someone.

And I think that is a really critical thing for us, being a kind of small team, yet kind of across these different spaces that have different functions. It's really important that the right task is assigned to the right person so we can ensure that things are done in a very timely way. And of course, beyond the work orders, we have maintenance that we kind of schedule in advance or have kind of across the The month in our spaces.

And so being able to see when that's occurring to again, not interfere with guest stays. And then of course we have our campus views and then also our guest entries. And so everything is synced from our Google calendar. However, if something is needed to be kind of added or appendaged, you're able to go in here and kind of overwrite what's been, what's been included.

The calendar views is very important for us because again, it gives us an overview of what's going on that day, that week, kind of in the coming weeks. And again, to ensure that we can kind of do preventative maintenance and whatnot, it's really important for us to be able to have this future looking look at the campus. And this is the report that I was speaking of previously. So this is where it allows me to change the language. I simply hit print. I can even define a time zone.

So for instance, the week ahead, and then it will just generate a PDF. Paper that I could then print out and provide to a team member. And because this is a very private portal, I've made it invite only. So I have the functionality to invite a team member by name and email. They will basically get an invite to join via email. There'll be a special code that's unique to us and that will give them access to the tool.

Francois Botha: Great to see this overview. I think it's great to see just a fresh take. It's very clean in terms of the use case that you build for, and it's quite evident to see how that plays out in the tool. My first question maybe is just talking a little bit about the building around the existing calendar rather than an entirely new system. Why did you focus on that? And what was that experience like?

Alex Sadowski: It's a great question. For us, as an example, There's plenty of wonderful tools out there, but they're only as good as someone is willing to adopt them. For me, Notion comes to mind that it's a wonderful tool to use, but for like a first time user, it might not be the most intuitive to build pages and project management portals within Notion. And so for me, I wanted to make sure that this was going to be something that was truly an aid to everyone.

And so if everyone has already been used to calendar entry for guest visits and for scheduled maintenance, I didn't want to necessarily circumvent that or have them do something different. And so the tool does intake all of that information and it structures it within the portal, which is perfect because it can kind of coalesce with work orders that are added or things that are manually added to the portal that are completely kind of net new unlocks.

And so for me, that was kind of like a win-win scenario where there's this tool that gives you the ability to do these new things or even some of the old things, but it's going to work with the old systems as well. Because I think that was the biggest feedback point that I received from my team who was actually going to be using this.

Particularly seeing them kind of interact with other tools or trying or trying demos with other things, not in this space necessarily, but just software in general, it became very evident that that was going to be a need.

Francois Botha: I think you make a good point there about using the tools that people already use. So Calendar being one of those tools, it's just so obvious that that is something that they use. And I think the nice thing is also this modern approach of rather than rebuilding entirely new tools, actually integrating with some tools that are already in place. So I think there's a great integration use case as well. But maybe, can you speak a little bit about the tools that you actually use to build this?

Do you have a technical background? How did you approach this build? You mentioned the prototype earlier, but what else did you use to actually get this build off the ground?

Alex Sadowski: Great question. The tooling that was actually used or that is currently used to allow this site to function is a very short list. It's essentially a ChatGPT hosted site that points to a subdomain of ours through a Cloudflare, which is a DNS hosting portal. And so those are really the only two things that allow this to run and that allowed me to build the tool to begin with.

And for me, I would say that I don't have a technical background, but for me, I've always been one, very curious and two, I think very eager to create. And for me, AI has just been this huge unlock to, to kind of be able to really lean into that. And so. I would also say that for me, I have most success with AI tools when I properly brief them and provide as much context as I'm able to. And so I almost don't even treat it like an AI tool.

I really treat it as if I'm giving a product brief to a technical person or someone else who's going to assist me to build this tool. And I think for one, that is probably what sets me up for success the most in building the tool. And then Being able to kind of iterate from there with the AI tool.

The only other thing that I would add to that is I then later connected a platform called Resend, which is an email sender essentially, because that was kind of the next unlock for the tool was to be able to send emails, whether If a team member forgot their password and they need to reset it, as we're all kind of used to, to being able to receive a kind of weekly heads up email that's going to, on Monday morning, go out early and just share what the week ahead would look like on campus.

Francois Botha: Okay, so that's a very practical view on what you used to build, but maybe let's just get an idea on how long did it actually take to get to a version one? Are we talking, you mentioned like detailed context and things like that. Did it take, I mean, days, weeks, months? What are we talking here?

Alex Sadowski: I would say v1 was a much faster Process than building out V2. V1 for me took effectively a day to do. And again, it kind of started half that time was just sitting down, scoping, getting all the detail on paper, and then kind of working with the AI agent to kind of massage and continue to build the platform. So literally by the end of about eight to nine hours, I had a functioning hosted site that worked and that I could share with the team.

Francois Botha: And V2 would have been the email sending that you added onto this or what went into V2?

Alex Sadowski: Yes, V2 included email and that was kind of the really big challenge, I would say, because email is kind of its own separate animal and it has its own Difficulties, or I should say challenges to integrate, especially with like an AI tool like this. And so that took a lot more fiddling and just learning and work. However, again, something that the AI tools can really help guide you through it.

But again, kind of going back to having this be a very easeful tool and something that is really of benefit to my team. Email was a required component to have because it really Push the envelope with how this tool could be of service. Every week they get a weekly update. They don't even have to use the tool necessarily, but they're still getting the benefit of what the tool is providing.

Francois Botha: So I guess two questions that I have just hearing this build, I guess the one question I have is really around maybe the technical aspects of this in terms of whether you had any technical background going into this, because one big thing is often around AI tools and AI builds not looking Or considering the various security or privacy aspects. And I think you need to be quite opinionated on that when you do it right. So maybe let's just focus on those two topics for a bit.

Any technical background and how did you approach privacy and security here?

Alex Sadowski: It's a great question. It's a very important consideration. I would say I don't really have a technical background, especially in this area. However, I would say from experience, I have learned enough about the different kind of safeguards and things in place that you need to kind of look out for. So in this particular example, because we connect into a DNS, we kind of have a lot of our privacy and security functionality on that end, on the domain side.

And then also just working with the AI agent to ensure that kind of everything's password protected on the site and also has the proper safeguards. And so this was another area that I did, again, kind of in my scope kind of outline. And I had to kind of also work with the AI agent to one kind of research and ensure that I was doing everything right on the DNS side. But then also as it was building, it was going to kind of keep these parameters in mind.

Francois Botha: Let's switch gears just around this discussion because the build is really cool to see, but I'm wondering about the uptake. I mean, this tool is live. Let's talk just a little bit about who's using it. What do you think their overall experience is like? Just a reflection on anything that you think is working well or not well. What can you share with us on that?

Alex Sadowski: Yeah, so my whole hospitality team and a few of my colleagues use the tool. One, because it helps them kind of see Kind of lay of the land, who's coming to campus, who's departing, what sorts of things need attention on campus, and also just allowing us to keep an eye on scheduling around things if there's any sort of scheduled maintenance that's planning to happen.

And so when I first kind of introed it to the hospitality team first, their first reaction was, you built this, which I still love, because it was just this like, Shock and awe of what this thing kind of meant to us in terms of really unlocking kind of this rising tide that lifts all boats kind of tooling.

And so I would say that was kind of the first big unlock and I've been sure that I've kind of stayed close with the team and continue to ask them and they have feedback, which I'm grateful for. And I'm excited to kind of implement all that in a V3. But then even like our principal, some of my other team members who aren't so much hospitality focused on the team, for them, it's also great because it just kind of gives them another lens into the campus that we all share.

And so it kind of gives them the ability to future look at who's coming, if they need to be aware of anything going on on campus. And so really team-wide, it's been a tremendously useful tool.

Francois Botha: You talked a little bit about v3, so looking at the future, how do you see the next versions here? Are there any lessons that you have learned from this first build that you are taking into kind of the road mapping, things that you might do different,

Alex Sadowski: Yes, absolutely. For one, everyone asks me for a dark mode, so that is in the works. There's some definitely quality of life updates I need to make. I think one of the challenges with using an AI agent is I would say design is not their forte. So I had to massage it a lot to get it to its current state, but I have a lot of future ideas of how I can Kind of work with it to kind of refine the design from a UI perspective, but also just from a visual kind of appearance perspective as well.

So I'd say that's kind of a big component of it, ensuring it's mobile friendly. But I think what I'm truly aiming for is actually building an app. And that way people can get push notifications. They can take a picture with their phone, upload it very simply. And so for me, that's kind of what I'm currently scoping on. I imagine it's going to take a bit more time to do it properly, but that would be, I think the Tremendous kind of final unlock for the team.

Francois Botha: Amazing. And in terms of feature requests, how are you handling that? Are you handling it in a professional product management kind of way? Do you just document it? What does that look like on your end?

Alex Sadowski: Yes, I have, I have a Notion that I'm kind of running to kind of keep, keep, um, tabs on, on the updates for this system and, and kind of future requests. So I do, if anyone has a feature request or like a bug fix, um, bug fixes, I I try to take care of expediently, but any feature requests, I just ask them to kind of scope it as best they can. And I just slot it into my, my updates.

So I, V3 is going to be my, my big update, but I have had some little like sub updates, um, from kind of quick feature requests that are more smaller or just like very easy, uh, quality of life ones.

Francois Botha: So Alex, thanks so much for sharing that. Any other last reflections that you think you'd like to share with the audience?

Alex Sadowski: Just saying, I would very much encourage any and everyone to just explore with AI. It's such a wonderful opportunity to create something that you're dreaming of or you have an idea for. And even if it's not feasible, I think to learn something from the exploration at the very least. And so I think if you approach these tools with curiosity and a willingness to want to learn, I think you can go really far no matter what.

Francois Botha: That brings us to the end of today's session here with Alex. Alex, thank you so much for joining us today.

Alex Sadowski: Thank you, Francois. It's been a pleasure.