THE OMNI SHOW

Andrew J. Mason: You're listening to The Omni Show where we connect with the amazing community surrounding the Omni Group's award-winning products. My name's Andrew J. Mason. And today we learn how Pete Edstrom uses OmniFocus. Well, welcome, everybody, to this episode of The Omni Show. My name's Andrew J. Mason. And today we are excited to have Pete Edstrom with us. He's into generative AI prompt engineering, and he's the director of technical product at Optum. Thanks for joining us, Pete.

Pete Edstrom: Yeah. Thank you so much.

Andrew J. Mason: It's our honor. And before we get going too much, talk to us a little bit more about Optum the company.

Pete Edstrom: Absolutely. Yeah. So I'm at Optum. It's really part of a larger group, UnitedHealth Group, UnitedHealthcare and Optum. It's like 350,000 people and we serve about 150 million people, members, throughout the country.

Andrew J. Mason: I think the first question from me, Pete, is when I first hear the term director of technical product, do you mind breaking that down for us? What does that entail? What does somebody who has a career as a director of technical product end up doing?

Pete Edstrom: Yeah. I can't tell you all the secrets, but yeah. So I lead a team of product managers. And product managers are about prioritizing the work, about figuring out what the right work is to do that's going to drive the most value for the organization. So we're working with our customers. We're working with our engineers to say what is the right next thing to be working on. So if you think about a project manager, they're just going to make sure that the work gets done well. And the product manager makes sure that you're doing the right work. So that's where my team focuses their at work, and we're part of a larger organization that is building search for UnitedHealth Group. So think of it as a team creating Google for all of the UnitedHealth Group companies and properties and needs.

Andrew J. Mason: Wow. So my next question was going to be how does AI play a role in all of that for you? But man, I do see some of the challenge of integrating search across different disparate platforms would be for you.

Pete Edstrom: Yeah. I mean, you're seeing in the market a lot of search applications starting to blend with all the generative AI things happening, and we're definitely there trying to think about how we can do that with all of our properties. Even when we're not thinking about generative AI, there's a lot of machine learning that happens in terms of how do you prioritize the right pieces of content to put to the top, that relevance ranking. And so the team I work with spends a lot of time figuring out what data can we pull in, how can we use that to show the right things at the top when people search for things, and then how do we continue to improve over time?

Andrew J. Mason: So is all of this really in the name of ease of use, putting the right thing in the right context in front of the right person at the right time and maybe even intuiting based on patterns that we're not even taking a look at? "Hey. This is what we believe you might need to see here."

Pete Edstrom: Absolutely. And people don't necessarily even... They just want it to work magically. So if you go out to Google and you search for a thing, there are 1000 things that you did not tell Google that you meant when you search for those three words or two words that you typed into the Google Search box. And so it's our job to figure out what are all those things that you probably meant, how can we figure out to infer those. So for example, one of the searches that we were involved with recently is the provider search for UHC. And it was actually announced as part of the investor conference that the company did last year, which was fantastic and it was great to have our work showcased. But we're digging into that huge, vast number of providers that are available in this country. And if somebody searches for those, we want to show you only ones that are in your neighborhood. We want to show you only ones that are actually part of your provider network and plan so that if you pick one and you go to it, you're assured that when the expenses come through that they're going to actually be covered by insurance. We want to show ones that are actually performing. So if you search for, say, hair loss, you want providers to not just say that they are working in that area but that are actually performing those procedures, those consultations, that kind of thing, and returning those results towards the top. So we try to give you as smart of a response as possible.

Andrew J. Mason: Okay. So that's really helpful because it's helping to draw that line between... There's filtering in search already. And as long as you know what you're looking for, you can whittle away all the extra results so that you can see what you need to see in front of you. But the determining factor there is that you actually know what it is you're truly looking for. So if you don't, it's helpful to have something that intuits that for you and says, "Here's what we believe your intentions are based on other data that we have."

Pete Edstrom: Right. So if you put your what's obvious hat on, that's the bar that we're shooting for. If I search for a thing, what are the obvious things that we would obviously want to include in that search to return results? Now we have all sorts of data and we want to also toe that line of not showing you things that are creepy. So there's a lot there to it in terms of making sure that we're showing you helpful, relevant results without getting into that space of, "Oh, you know too much about me, and I don't like..." People run into this with social media all the time. They'll see an ad pop up and they'll be like, "Why am I seeing this ad? And how do you know that I was just searching for that thing on Google two days ago?" You get into that creepy space and it just makes everybody unhappy. But yeah, building out good relevance, building out those AI models that can figure out the right thing and really try to serve our members and help people get to the information that they want to get to as quickly as possible.

Andrew J. Mason: It really does sound like that sweet spot where you need to make sure it's just right, not too much, not too little. I want to switch gears and then we'll come back around to AI, but first let's talk a little bit more about do you have any recollection or memory of seeing anything to do with the Omni Group, whether as a company or more specifically OmniFocus? Do you remember a first time that you came across the software? Or was it more of a progressive thing over time, you just kind of realized, "Yeah. It's a company that's out there."?

Pete Edstrom: I had to go look this up because it's been a long time, but I... So let me tell you where it started. So I was into those life hacks and 43 Folders and Merlin Man and all of those things. I'm sure you've heard this before. At one point, Merlin Man had written an article about getting started with getting things done. And so that got me into David Allen. I read the book. I'm like, "This is fantastic." And then I heard that Merlin was working with the Omni Group to create OmniFocus. He was a consultant on that. And I'm like, "I love this plan and program here. I love how he's thinking about it." I had even done a bunch of the... Tried to do the inbox zero kind of stuff and totally failed at it, but I strived for it for a very long time. Yeah. And so that's how I got into OmniFocus initially. I was trying to find the receipts for when I first purchased OmniFocus. I couldn't find the very first one because... I found the OmniFocus for iPad receipt version 1.2. I bought that in November of 2010. And I got the iPhone version 1.9 back in April of 2011. But I actually found a blog post I had written, a blog that's long since dead now that I'd written December of 2007 that said I hadn't purchased OmniFocus yet but I was definitely inspired by it. So somewhere between 2007 and 2010 is when I got my very first version of OmniFocus. I got the Mac version, yeah, and then the iPad and the iPhone. And it just built from there. At the time, and I think you still get this some today, there were a million different to-do tracking apps out there. Each one of them had some variation of some novel feature whenever. And you knew that... My background is in computer science. So as a software engineer, one of the very first things you do when you figure out how to write some code, maybe make a web app or an actual app, is you're like, "I'm going to make a to-do app. It's limited scope. It's not too bad. I can have a database with a list of things and some checkboxes," and we're off to the races. Well, the market was flooded with that kind of stuff. And with the Lifehacker thing and the 43 Folders thing, you were just inundated with, "Hey. There's this new to-do app, to-do tracking app, that is going to be just perfect for that one thing." And I was always chasing after that and finding that new thing that's really going to make it sing and it's going to be perfect and everything else. So when I got to OmniFocus, "Okay. This is expensive, but it does everything. I don't even understand all of the things that it does. But if it doesn't do the thing that I want, it's going to be close enough and I can make it work." And the money was so well spent because it meant not only did I have a great to-do app or productivity tracking app or... I'm sure what we call it. But it meant I never had to try another app again in that space. I had made my commitment. This was going to be it. If it worked, it was going to be great. And if it didn't work, it was not because the app didn't work. It was because I wasn't using it right. So we're going to just dig into it real deep. We're going to run with it. And it's been fantastic. I have not tried a single one since.

Andrew J. Mason: Wow. Very few people, I think I can count on one hand that we've interviewed, that just, "Hey. I made the commitment in 2008, 2010. Haven't looked back." Very often, including me, there's been shiny app syndrome where you think, "Man, that one feature in that one app might be the answer to all my problems and all my dreams and fill in the gaps so nicely." But kudos to you for that level of commitment.

Pete Edstrom: Yeah. And part of it was the sunk cost fallacy of like, "Hey. I've spent all the money and so I'm going to keep digging into this," but it's been a fantastic app and it's just kept giving back and it's freed... So when you think about David Allen and using your head for ideas, not for remembering things, it freed my head from being able to never have to think about another alternative again. I can see them come up and then in the feeds and be like, "Oh, yeah. That's super cool. I'm not going to try it because OmniFocus has already got it figured out."

Andrew J. Mason: Let me talk to you about advice that you might have. What shows up for you when you think about people that are 10, 15 years our junior, just entering the workforce or maybe are hitting that space of expanded responsibility, they have this realization that, "My gosh. I cannot do everything that I've committed to do if I'm operating solely out of my head."? What first go-to bits of advice might you have for somebody who finds themself in that place of responsibility expansion?

Pete Edstrom: So I think I already said David Allen said that your head is a great place for having ideas but not holding onto them. Understanding that and believing that is a really great place to start. Just get your mindset into that place. "I'm going to continue to have this series of ideas and thoughts. I need to get them out of my head and write them down somewhere that I can trust, that I can find later." And if you start thinking about like, "Oh, man. I'm going to talk to Andrew in 45 minutes, in 30 minutes, in 15 minutes," it's not a great use of your head because you end up on that loop and it's not a helpful thing to think about. Set yourself a list, a reminder or something else. Free your head to go have breakfast and, "I'll talk to Andrew as soon as my alarm goes off and we're going to be great." But start small. Think about maybe bullet journaling. Think about just using paper and pencil. I think an important part of it is having something with you that you can capture these ideas with. So whether that's Siri with a reminders app or it's actually paper and pencil in your pocket, you're going to have ideas when you're driving, when you're biking, when you're walking around town, when you're having dinner, and being able to take that few seconds to say, "Okay. I've got a great idea. I'm going to write it down and I don't have to think about it now, but I will come back to that tomorrow morning or this weekend or whenever," I think is a great way to start.

Andrew J. Mason: Pete, talk to me about your software stack. What flows into and what flows out of OmniFocus and where it sits in the overall workflow of your software? So does it daisy chain into a larger picture?

Pete Edstrom: So I already mentioned I've got the iPhone, the iPad, the desktop app. I do have it on my watch, but I don't use it there as often. A big part of it is having that ever-present idea capture capability. So I will use the quick entry on my laptop or laptops all the time. I use a Siri to remind me to do a thing and have that sink into the inbox all the time. I did for a while use OmniFocus, I believe they're called, templates with the editorial app on the iPad, where I would pre-build a project for a particular thing and then it would automatically create that with all the dates and the right titles and stuff. I've gotten a little bit away from that. I need to get back to it. So that's one thing that I was doing. I'm not using DEVONthink, any of those tools. I am using Apple Notes really heavily. And so that's maybe a companion tool to OmniFocus, not necessarily feeding things in or out. I also use the Bear app, again as a companion application to OmniFocus. The way I have things set up, Apple Notes is my meeting notes. It's the things that my lists or other things that I need to remember, not necessarily that I need to remember, but I want to have evidence for a thing if I want to go search my memory so for those things. Bear, I have switched over to being exclusively my address book. So when you're working at a company like the UnitedHealth Group with 350,000 employees, you end up working with a lot of different people. And your mind cannot remember all of the names of all the people, much less like what projects they're working on, do they have kids, who is their manager, what is their title, what projects have you already worked with them and collaborated on. So in my Bear app, I've got almost 2,000 records right now. I had to look this one up yesterday too. And it's one record or one note per person, but it's great. I can link them together. If I'm working with you and maybe we met through somebody else, I can put that link in there. I've got a take in there for OmniFocus. So anybody that I've talked to or worked with around OmniFocus, I can click on that and see all the related people. So that part's going on. I don't know that there's any... I mean, we can talk about the output of OmniFocus maybe in the AI space and some of the things that I would like to see done eventually or at least I would like to try to do myself. The other Omni piece is I do use OmniGraffle, primarily for work, but it tends to be more solo work there than collaborative work, so... But every time I pull that one up and put a diagram together, everyone loves it because it looks so much better than any of the other tools out there.

Andrew J. Mason: Very interesting. I actually had never heard of that use case for Bear, almost this CRM that level sets an interaction with somebody, especially if you interact with so many people in a given day. Being able to just like, "Oh, yeah. That's the project I worked on with them," before you have a conversation with them, that's awesome.

Pete Edstrom: The tags and linking between notes was really a game changer for that one. And of course the search. So being able to type something at the top of that bar and have things filtered down real quick is fantastic.

Andrew J. Mason: Yeah. I could see where that would really speed up the search process. I feel like this is a good time to start to merge those two threads in our conversation together, where are there any spaces where you see AI colliding with or potentially colliding with OmniFocus here in the future? But before we hit the future, talk to me about what's now? What do you see in the world of AI and OmniFocus?

Pete Edstrom: Yeah. OmniFocus and AI. Well, so there's nothing I'm doing there yet. Let me just back up and say even outside of work, I'm a huge fan of OpenAI and ChatGPT and all of the things that have been happening in that space. It's just a fun side project. I've been doing some lunch and learns on prompt engineering at work and helping people get up to speed on what they can do. And frankly, it's not hard. If you know how to talk, you can talk to an AI. And so part of my work there is just helping people understand that, "Hey. You can do this. It's not scary." So today with those tools, you can, say, talk to a PDF. You can upload a document. You can inquire about all sorts of things in there. There's new tools showing up so that you can chat with an AI about all the files on your hard drive, research papers. I think recently Perplexity introduced Yelp as part of their data sources. So you can chat with Perplexity about restaurants that are in your neighborhood, which is fantastic. Yeah. It's so cool. It starts to merge this conversation with search in a way that's just delightful. I want to be able to talk to my OmniFocus database, all the to-dos. And I've been thinking about this one for a little while and I tried some things and I haven't quite gotten it to work. Questions I would love to be able to ask of my database are, "What projects have I been ignoring for too long?" or, "Which ones should I review more often?" or if I told the AI what some of my larger life goals are, maybe a short paragraph of my one-year plan, my five-year plan, my ten-year plan, if I told it that, what would it recommend that I put my priority on next for my to-dos?

Andrew J. Mason: It's so funny about this dance between autonomy and ease of use. What do they call it? The cognitive fatigue relief that happens. I remember Starbucks came out with this location where we would drive by a Starbucks and it'd be like, "Hey. You usually order a latte this time of day right here. Do you want to just hit a button or say yes?" Some people got freaked out by things like that. So it's interesting to see this dance go back and forth between how much do we use the information we know about you in order to serve you well versus you decide what information you want to keep and not give in our directions so that we don't do those things. To me, it seems very obvious that you don't get to the position that you're in without at some level thinking about how productive you can possibly be and being passionate about getting as much out of yourself as you possibly can. Talk to me about where that comes from for you. What makes you passionate about being productive?

Pete Edstrom: Yeah. That's a fascinating question. When you first posed that to me, I was like, "I don't know." It's kind of like who I am. It's so ingrained. It's hard to pull it out. Part of it is a little bit of a competition with myself, maybe a little bit with others too. Productivity to me is a little bit around not necessarily being the fastest or the shiniest or the best, but it's about almost like, "How can you orchestrate the most a complex situation and do something that's very simple?" And maybe it's a little bit about the best part. So one of the things that my family did when I was a kid, so this is way before GPS, way before digital maps, this would've been in the 1980s, my dad would map out a way to get somewhere, whether it's to my aunt and uncle's house or to a community event or someplace that we're going regularly to. And we had a literal stopwatch and we would time it, "Okay. This time we're going to take Main Street. Next time we're going to take the highway. Maybe we're going to try another route after that." We would gather the data. Which one is the best? Why would we want to do that? Why would we want to take a route that's not the best, right? We'd gather all that data and then we'd narrow it down and say, "Okay. So this is the route. So whenever we go to our aunt and uncle, this is the route we should take because that's the one that's going to get us there the fastest." And we have the receipts. We did the work. And so I don't know that that's the most formative of experiences in my background, but that really led me to thinking about every last thing that I do in terms of efficiency and productivity. It's so stupid. I will empty the dishwasher and I will be proud of myself for using both hands when I empty the dishwasher. So I'm pulling out the basket with all the silverware and opening the drawer for silverware at the same time. I'm using both hands and it's so ridiculous. But I also watch other people do this, and I'm like, "Man, you could do this so much faster if you just use both hands. You've got two hands. Use both hands." And so... I don't know. There's a resource efficiency maybe angle to that that I've always been drawn towards. So I tend to be very future and forward-thinking. I love the idea of trying these new technologies, seeing what those future technologies can help me do today that lets me do it faster, lets me do it more efficiently. There's probably an ego thing in there. Lets me look good in front of other people. "Hey. Look at this cool thing that I did and it only took me five minutes where most people takes an hour." There's a lot of things there. It's always been fun.

Andrew J. Mason: For whatever reason, just when you said that, it brought to mind this... I don't know if it was a webinar or something. I was listening to David Allen say, and this phrase, he had stuck with me. He said somewhere along the development of the methodology, he came up with this idea that, it was a revelation to him, that you could more or less effectively think about a thing. And so there's a more effective or a less effective manner with which you could think about a given thing. And so that's why it's important to direct your thoughts in a given direction. I think of that, a little bit of an element of that along with this. There's also some fun to it. It's the gamification of it of, "How easy, how fun, how good can we make this?" Also in terms of how you can free up time for other things that really are important to you as well.

Pete Edstrom: Yeah. It's totally fun. And I think mostly it's a game against myself, like, "How can I do more with less and really crank through all..." AI has been fantastic for that. "How can I write this complex email and take as little effort as possible with it?" We sent out a Christmas letter last year. And I asked my family, "Hey. Send me a handful of bullets on what you want to share in there. Don't worry about style or anything. Just give me the content." And I fed that all in and it created this beautiful letter. I ran it through a few different styles before I found the one that I liked. And I'm like, "Great. This is a perfect style." And I'm like, "You know what would make this really cool? Emojis? Let's get emojis in this whole thing." And so then I just filtered it through and it picked out all the... There was probably 50 emojis on this two page document, but it was all the perfect emojis in all the right spots, and it was colorful and fun. And I didn't have to spend any time searching for the perfect emoji for all that. And it was like, "This is great. This is fun."

Andrew J. Mason: Pete, thank you again for spending your time with us. One final question before you go. I love to ask people in your journey thus far, maybe you don't call it a mistake, but what have you come across that as you look back and you say, "If I were able to do this whole thing over again, maybe I would skip that part." And it could be instructional for people that haven't yet traveled the road that you're on, just saying, "You know what? If you find yourself in this situation that I found myself in, you don't have to do it the way I did."

Pete Edstrom: Yeah. So I don't have a great, "This is a section of my life that I would want to redo," but there's an attitude that I wish I could remove. So I spent a lot of time, a lot of my time, earlier in life, early career kind of time, complaining about the justices or the wrongnesses of one situation or another. A lot of it was around work and around, "Can't believe the manager did this," or, "Can't believe they're having me do this one thing." And my wife heard ad nauseam about all this kind of stuff. And I'd be raging about work. Sometimes it'd be about the government. It'd be about other people. And if I could redo that, I absolutely would, take a step back and say, "Well, are these situations that I can control? Or are these situations that I have no control over? I might as well show up with a good attitude, enjoy the space, the people, the situation that I'm in, and work with that. And on the OmniFocus side, if I can change any of those situations, put a project together to start saying, "Okay. Maybe I can't change my job tomorrow, but I can start looking for a job. I can update my resume. I can solve some of these problems over time. So let's have a good attitude today. And in the meantime for things that I don't like, stop being so vocal about it and just build a project plan and start chipping away at it. You can make a difference in a lot of different things." I'm a big believer in steady pressure over time, and you can make a difference. So enjoy where you are. Enjoy the situation. And for everything else, make a plan to shift it.

Andrew J. Mason: Well said, Pete. Yeah, that's perfect. How can people get in touch with you? If they're interested in finding out more about what you're up to or just even connecting with you, how can they do that?

Pete Edstrom: I can't share any of the workshops. Those are all internal at work. Email is great. And LinkedIn is also fine. If you connect to me on LinkedIn, just let me know that you've heard about me from this podcast and I'm far more likely to accept the connection. I tend to not say yes to every last person that just reaches out, so... But if you give me something to grasp onto and for us to chat about, I'm more than happy to do it. Let's make the connection.

Andrew J. Mason: And speaking of that, thank you for connecting with us, Pete. So grateful for your time, I know you're busy, and just being present with us here today. I think it's been valuable.

Pete Edstrom: Awesome. Thank you so much for having me on. This was a lot of fun. And yeah, I hope everybody has had a good time here.

Andrew J. Mason: Hey. And thank all of you for listening today too. You can find us on Mastodon at The Omni Show at omnigroup.com. You can also find out everything that's happening with the Omni Group at omnigroup.com/blog.