I ran a small electrical contracting business in Phoenix for six years, and I watched probably $200,000 in jobs walk out the door because nobody answered the phone on Friday evenings or weekends. I tried two different answering services. Both times the operator had no idea what a 200-amp panel upgrade meant, and customers just hung up and called someone else. I sold the business in 2021 and spent two years building dispatch software for field service companies, which is how I got obsessed with the after-hours gap. Most service businesses are one missed call away from losing a job to whoever picks up next.
I spent six years as a controller at a Series B SaaS company, and every quarter close was the same story: two weeks of chaos, reconciling spreadsheets at midnight, hoping nothing had slipped through. When we went through a formal audit for our Series C raise, the auditors surfaced three months of misclassified expenses that took a full week to unwind. That nearly killed the deal timing. After it closed I kept thinking: we should not have been finding this during the audit. We should have caught it in January, not June. That is what Ledgerline does. It keeps your books audit-ready all the time, not just when someone is counting on you.
My dad ran a two-bay shop in Tucson for 22 years. When I was finishing my CS degree at ASU, I'd go in on weekends and watch him write estimates by hand, lose track of parts orders in a spiral notebook, and miss calls because he was under a car. He knew everything about engines but spent half his time on paperwork. I spent three years at a fleet management startup after graduation, which is where I started seeing what real operational data could do for a shop. I came back to Tucson in 2022, sat with him for a month, and realized the tools built for dealerships were completely useless for independent guys. Auto-AI is what I wish he'd had.
I ran my own bookkeeping practice for eight years before building this. Every quarter I burned hours writing the same explanations to clients about why their books looked the way they did, what documents I still needed, why their P&L didn't match what they expected. My clients were small business owners who needed hand-holding, not spreadsheets, and I genuinely loved helping them understand their numbers. But the writing was eating my Fridays. After I sold my practice to a larger firm in 2022, I kept thinking about all the solo bookkeepers still drowning in that same inbox grind and figured AI should be solving exactly this.
I spent six years as a project manager at a boutique branding firm in Chicago, and the thing that wrecked every project was never the work itself. It was always the brief. We'd kick off a rebrand, three months in the client would say 'we never agreed to that,' and I'd be digging through email chains trying to prove otherwise. After the third time a client dispute killed a project we'd actually done well, I started building a tool just for our team. When I showed it to two agency friends and both of them teared up a little, I figured it was worth doing for real.
I ran my family's printing shop in Leeds for four years after my dad had a stroke and couldn't manage it anymore. We were profitable every year on paper, but twice I couldn't make payroll because all our receivables landed in the last week of the month. My accountant would show me the quarterly reports and I'd think everything was fine, then I'd be scrambling for a bridge loan at 11pm on a Thursday. I eventually sold the shop, went back to school for CS, and spent three years at a payments fintech in Manchester. I kept thinking someone must have already built this properly. Nobody had.
I grew up watching my mom spend half her Thursdays putting together our church bulletin, and by college I assumed every church just had some magic system for that. Then I spent seven years in B2B content marketing and realized the tools I used every day would completely transform how a church admin spends her week. A friend who planted a church in Phoenix was writing his own announcements, social posts, and welcome emails every Sunday morning at 6 AM. I thought I could fix that in a weekend. Two years later, we're serving 340 churches and I've never been more motivated to get to work.
I grew up watching my dad run a two-physician internal medicine practice in Evanston. He was a great doctor, but by the time I was in high school I noticed he was staying up past midnight most nights, not studying, just doing paperwork. I spent six years as a healthcare IT consultant after that, and the story was the same everywhere: brilliant clinicians drowning in documentation and scheduling chaos. I tried pitching better tooling to two of the big EHR vendors and got nowhere both times. So my co-founder Raj and I decided to stop waiting and just build it ourselves.
I spent eight years teaching group fitness and working one-on-one with clients at three different gyms in Phoenix before I finally went independent. The transition was exciting until I realized I had no system for anything. I was texting workout plans at midnight, keeping notes in a random Notes app folder, and tracking macros for twelve people in a spreadsheet that made no sense to anyone but me. My husband, who works in software, kept asking why there wasn't a proper tool for this and I didn't have a good answer. So we started building one.
I spent six years as a freelance UX designer and I probably lost around $30,000 over that time to clients who found loopholes in contracts I'd cobbled together from Reddit templates. When I finally hired a lawyer to draft me a solid agreement, she charged $800 for something that took her about 20 minutes. I watched the same thing happen to every consultant and small agency owner I knew: either skip the contract and get burned, or pay attorney rates for boilerplate language. My co-founder Jorge and I started Inkwell because that tradeoff felt completely unnecessary in 2024.
My dad ran a roofing crew in Baton Rouge for 22 years. I watched him lose bids because they took too long or had math errors on the material takeoffs. I spent six years as a project manager at a mid-size general contractor in Houston before I realized the software we were using was basically just spreadsheets with a logo on them. After we lost a $400K job because of a scope error in our estimate, I quit and started building what I wish we'd had. I'm not a coder, but I found two who are, and we've been in the field talking to contractors every week since we started.
My dad ran a six-truck concrete delivery company in the Central Valley for 22 years. Every morning he managed his drivers from a whiteboard in the garage and a Nokia. When I left my job at a mid-size logistics software company in 2019, I went home for Thanksgiving and watched him call each driver one by one just to figure out where everyone was. I'd been building routing tools for enterprise carriers the whole time, and the gap was embarrassing. Everything I'd shipped cost $40k a year minimum. Pylon is what I wish I could have handed him back then.
My dad ran a plumbing company in Baton Rouge for 22 years and retired with almost nothing because he never figured out how to price jobs right. When I started my own residential plumbing business in 2018, I did the exact same thing for three years straight. I'd win the bid, get excited, and then discover halfway through the job that I'd already eaten the margin. I tried spreadsheets, I tried ServiceTitan, I even hired a bookkeeper. Nothing actually reasoned through the variables the way you do in your head on a job site. So I started building Plumb.
I spent six years as an events coordinator at a mid-size hotel in Atlanta, and the paperwork almost broke me. Every event had its own folder of Word docs, manually updated spreadsheets, and vendor contracts we'd lose track of between email threads. I watched colleagues redo the same run-of-show document from scratch for every event because nobody trusted the templates. After leaving to join a workflow software startup, I kept thinking about that problem. I built Foyer because I know exactly what it feels like to be three days out from a 400-person gala and not know which version of the BEO is actually current.
I spent four years as a product manager at a mid-size HR tech company where we had Typeform, Pendo, and Intercom all running and nobody read any of it. When I joined a twelve-person startup afterward, our CEO had this habit of emailing five customers every Monday with one question and spending Tuesday fixing whatever they said. Our NPS went from 31 to 67 in six months doing that. I kept waiting for someone to turn that habit into an actual product. Three years later, nobody had, so I quit my job and built it.
I spent six years as a dispatcher at a regional dry-van carrier in Ohio, watching the enterprise software get more bloated and more expensive every time the vendor pushed an update. When my uncle Dayo started his own three-truck operation in 2019, he called me to help him set something up. Nothing on the market fit a fleet his size without costing more per month than his fuel bill, so he was texting drivers from his personal phone and tracking loads in a Google Sheet. I built him a basic tool over a weekend, and four other owner-operators from his church group wanted it by the end of the month. I quit my dispatcher job six weeks later.
I spent nine years as a regional ops director for a 140-unit sandwich franchise, and the thing that killed me every week was knowing exactly which managers were running great locations and not being able to bottle what they were doing. Our best store in Tempe did $180K more annually than our worst store 12 miles away. Same menu, same training, same POS system. I tried spreadsheets, mystery shoppers, weekly calls. Nothing scaled. When my wife's family opened three locations and hit the same wall, I quit and started building what I always wished existed.
I spent six years as a program director at a workforce development nonprofit in Cleveland, and every grant cycle felt like drowning. We had two staff members writing forty-plus applications a year, rewriting the same mission statement seventeen different ways to match each funder's language. When we lost a $180,000 federal grant because our narrative didn't mirror the RFP closely enough, I knew the problem wasn't our programs. It was the ritual. I left to build the tool I wish we'd had. Almsbury isn't about making grants easier to fake. It's about helping good organizations stop losing to better writers.
I spent six years as a product designer at agencies in Chicago, and the last two weeks of every project were always chaos. Figma files dumped in a shared drive, Loom videos nobody watched, Slack threads that vanished. When I went independent in 2022, I started losing clients not because the work was bad but because the handoff was a mess. A developer at a fintech I contracted for spent three days asking me questions I had already answered in a buried Notion doc. I built the first version of Relay just to stop those emails.
I spent six years running recruiting for a Series B SaaS company and watched us lose three strong VP candidates to competitors while our hiring committee debated job description wording. When I left to build something of my own, I assumed I'd figure hiring out quickly. I didn't. It took eleven weeks to bring on our first engineer, and most of that time was just chasing feedback from people who couldn't agree on basics. Northstar started as a tool I was building for myself. I wanted something that actually compresses the timeline instead of just putting the old broken process into a nicer interface.
I spent six years at a hotel management consultancy in Houston helping independent properties compete, and I kept seeing the same thing: a 14-room inn with a great location and a terrible guest experience because the owner was answering check-in texts at midnight. My parents ran a small motel in Galveston for twenty years, and I watched my dad personally respond to every review, every late arrival call, every complaint about the Wi-Fi. When I finally had enough technical background to actually do something about it, I left consulting and started building. Independent hotels have soul that chains can't buy, and they shouldn't have to sacrifice sleep to prove it.
I spent six years as a service tech at my uncle's shop in San Antonio before I burned out and moved to Austin to do customer success at a B2B SaaS company. The whole time I was there, I kept thinking about how his shop was still running dispatch off a whiteboard and a group text chain. He lost three solid jobs one summer because nobody followed up after estimates went out. I left that job in late 2022, went back and helped him rebuild his process, and realized the core problem was that phone calls were never getting logged right in the first place. That is what I am fixing.
I spent six years as a freelance UX consultant, and toward the end I had three clients who each owed me over $2,000. I was using Wave because it was free, but every invoice looked like it was made in 2009 and I had no way to know if anyone even opened them. My accountant, who works with about forty independent contractors, said late payments and unprofessional invoices were the most common complaints she hears. I built Vellum because I wanted something that felt like it came from a real professional, not a generic SaaS template, and that actually told you what was happening after you hit send.
I spent five years as an intake coordinator at a mid-size personal injury firm in Houston. Every Monday I'd sort through the weekend voicemails, knowing most of those people had already hired someone else. We tried three different chatbot products over the years. All of them either collected useless information or scared people off with legal jargon before they even left their name. I left the firm in 2022, took a software product role, and kept thinking about that voicemail pile. Counsel is what I kept sketching on whiteboards for two years before I finally just built it.
I spent five years writing menus for a restaurant group in Chicago, and the thing that killed me every time was watching talented chefs hand me a legal pad with ingredient lists and say 'make it sound good.' Most of them barely had time to sleep, let alone write. I left to do freelance food copywriting, and the same problem kept showing up, but at $150 an hour, most independent owners couldn't afford me. I kept thinking there had to be a way to give a two-location taco spot the same quality menu copy as a Michelin-starred restaurant. That's what Saltline is.
I spent six years doing UX contract work for early-stage startups in the Bay Area, and in 2019 a founder I had been working with for three months shipped a product built almost entirely on my wireframes without ever paying my final invoice. I had no NDA. Nothing. After that I started using templates from legal sites, but every one was either fourteen pages long or so vague a client could drive a truck through it. I asked a lawyer friend to review one and she charged me $400 to tighten two paragraphs. There had to be something faster for people who just need a clean, enforceable agreement before a call ends.
I spent four years as a transaction coordinator at a mid-size brokerage in Austin, and every single open house started the same way: the agent sends me a Canva link at 10pm Thursday, the photos are out of order, the square footage is wrong, and we're printing flyers at Kinkos at 7am Saturday. I watched agents spend more time formatting PDF brochures than preparing their actual pitch. When my sister listed her place in 2022, her agent sent buyers a Word doc as the 'property page.' I figured someone should fix this. Nobody had, so I did.
I spent four years as a PM at a healthtech startup in Chicago and watched our CEO burn three weeks building a deck for our Series A. We had real traction and a great product, but the slide process nearly broke him. After that I moved to the venture side and saw the exact same thing from every founder who walked in. The story was always good. The deck was always a mess. I kept thinking someone should fix this, and eventually I stopped waiting and started building Lectern nights and weekends until I couldn't not do it full time.
I ran a four-person IT consulting shop for six years in Austin, and I swear we spent more time writing proposals than doing actual client work. I'd lose deals not because our work was worse, but because the other guy submitted faster and with a slicker doc. My wife finally told me I was spending more nights editing Word templates than I was sleeping. I sold the firm in 2022 and immediately started building the tool I wish I'd had. The goal is simple: let small shops compete on work quality, not on who has a better-looking proposal factory.
I spent three years doing contract dev work at a digital agency in Austin, mostly shipping Shopify builds for retail clients. When AI coding tools took off I started using Cursor to prototype side projects way faster than before. I shipped a small SaaS tool over a weekend and felt pretty good about it, until a friend pointed out I had my Stripe secret key sitting in a public GitHub repo. I had been so focused on the feature side I never stopped to check the basics. I talked to maybe a dozen other people building with AI and almost all of them had a similar story. There was no simple tool that just scanned your repo before launch and flagged the things that would embarrass you.
I spent six years doing web design on a project basis, and every January felt like starting over from zero. The month my daughter was born, I had three project deposits land and nothing else lined up for the following quarter. I started reworking my service offerings into retainers out of desperation, and within two years I had more predictable income than I ever did chasing one-off clients. I built Standing because what I figured out through trial and error over six years should not take anyone else that long to learn.
I spent six years at a local marketing agency in Phoenix, watching great small businesses lose customers to one-star reviews they didn't even know existed. My uncle ran a tire shop for twenty-two years and had no idea his Google rating had dropped to 3.1 until a loyal customer mentioned it. By then he'd lost three months of business to a competitor across the street. I tried to fix it manually for him and it took weeks of awkward email follow-ups. I knew there had to be a better way, so I left the agency and started building what I wish had existed back then.
I ran a small dev shop in Austin for six years and we were constantly getting eaten alive by scope creep. Clients would say 'just one more thing' every other week, and by the time we shipped, we'd burned 40% more hours than we scoped. I tried Notion templates, fancy contracts, SOW clauses... none of it changed the behavior. My business partner eventually quit, and honestly the burnout traced straight back to us never catching the drift early enough. I kept waiting for someone to build a simple tool that flags it before it compounds. Nobody did, so here I am.
I spent six years managing floor staff at a regional grocery chain in Phoenix, and every Sunday night I was on the phone until 10 PM trying to fill Monday morning gaps. Someone would call out sick and I'd start texting down a list, hoping someone would pick up. We tried two different scheduling apps and they both just digitized the same broken process. When I finally left to build this, my old GM texted me and said 'please make it actually work.' That's the whole pitch.
I ran a small home goods brand on Shopify for almost five years. Writing product descriptions was the thing I dreaded most, and hiring freelancers meant waiting a week for copy that sounded nothing like us. When I sold the brand in 2022, I joined a larger e-commerce agency and watched the same bottleneck play out across forty-plus client stores. Every product launch stalled on copy. I teamed up with my college roommate, who had been doing NLP work at a Series B startup, and we spent six months building Tradewind to fix the one thing that slows every store down.
I ran a small paid ads agency for six years and never once actually knew which clients were worth keeping. I had Stripe open in one tab, GA4 in another, and a spreadsheet I updated maybe monthly when the anxiety got bad enough. When I finally dropped three clients and my revenue went up, I realized I had been flying completely blind the whole time. I sold the agency in 2023 and spent the next year building what I wish I'd had: one screen that connects revenue, traffic, and spend without needing a data analyst to glue it all together.
I spent six years as a property manager at a mid-size residential firm in Phoenix, and when I left to help my uncle manage his four duplexes in Tucson, I was genuinely shocked at what independent landlords are working with. He tracked leases in a spreadsheet, collected rent over Venmo, and handled maintenance requests in a group text thread. When a tenant moved out and disputed her deposit, he had zero paper trail and ended up eating fourteen hundred dollars. That was 2022. I spent the following year talking to about sixty landlords before writing a single line of code, and almost all of them had the exact same horror story.
I taught reading intervention for six years in Phoenix, where most of my students were two grade levels behind and their parents were working double shifts. The kids who had outside tutors caught up. The ones who didn't, usually didn't. I kept thinking the actual teaching part isn't magic, it's just repetition, patience, and knowing exactly where each kid is stuck. I left the classroom in 2021 to join an edtech startup, which mostly taught me what not to build. Marigold is what I kept wishing existed when I was standing in front of thirty kids with thirty completely different needs.
I practiced small animal medicine for nine years at a two-doctor clinic in Columbus. By year four I was spending more time on SOAP notes and insurance prior auths than I was actually with patients. My mentor, who had been practicing for thirty years, retired at 58 just to escape the paperwork. When I missed my daughter's third-grade play because I was still charting at 7pm, I decided I was done waiting for someone else to fix this. I took a leave, did a software bootcamp, and spent two years building the tool I had been begging someone else to build for a decade.