The Hourglass Method for building SaaS

Hey y’all, Thanks for trusting me with your inbox! I'm really enjoying this style of newsletter. Hopefully you are enjoying it as well. If you have any ideas, please feel free to reach out and let me know --- ### Thoughts from the week If you didn’t catch this week’s [Mostly Technical](https://mostlytechnical.com/episodes/74-the-hourglass-w-jordan-gal), it was with a guest named [Jordan Gal](https://x.com/JordanGal). Jordan’s been an internet friend of mine for a long time, and since Ian was out at Disney, I had Jordan on to hang out and it was amazing. One of the things we talked about was an idea he calls the Hourglass Method. It’s a system for developing a SaaS product when you’re not totally sure what features your end users are actually going to want. His theory is that your feature set should be shaped like an hourglass: You start super wide, building out all the features that help check the boxes for potential users. Then, once people are inside the product, you watch what they actually use, narrow down to just a few features, and go deep on those. After you’ve nailed the core offering, you can expand again—adding more features now that you’ve got real traction. To me, this feels a bit like the explore/exploit idea. You explore broadly to figure out what people want, then exploit that knowledge by going deep. The difference is that at the end you explore again! It was a very good episode, I encourage you to go listen. --- ### Tweets from the week [](https://x.com/tferriss/status/1902063410952728652) One of the fun parts about working with Steve is that we have very different strengths. Some of our strengths actually overlap…I think we both pay close attention to detail, we have a good eye for whether something looks good, and we’re both very technical. But we also complement each other in important ways. Steve is great at following up with people. He enjoys doing cold outbound for sales, negotiating deals, all that stuff that, honestly, I’m not super great at. Working with Kelsey is similar, she’s just incredibly on top of things. I can let emails or DMs go unanswered for way too long, start to feel guilty about it, and then take even longer to respond. But Kelsey? She just gets things done. Because of that, I can focus on the things I’m really good at—the things that move our business forward—and Steve and Kelsey can focus on the things they’re really good at, which also move the business forward. It doesn’t always work out this way. Sometimes you end up with a co-founder who’s good at the exact same things you are and bad at the exact same things. Then you’re stuck. Nobody wants to respond to the email. Nobody wants to negotiate the deal. Nobody wants to do the marketing. You both just want to sit in the corner and program and that usually doesn’t work out. --- [](https://x.com/nikitabier/status/1902355813584232736) I’ve experienced this firsthand and have tried to shout it from the mountaintops: publishing your work increases your luck. It doesn't have to be Twitter, but it has to be somewhere. As Nikita says, you get to skip past 10 rungs just by putting your work out there. You break out of the old, strict hierarchy simply by hitting “publish.” --- [](https://x.com/Jibran_05/status/1901366949466284398) I saw this one after I saw Nikita's tweet and saved it as another example of publishing your work increases your luck. I've got dozens of these tweets saved in my bookmarks for the day when I eventually write something long-form about this. (A book 👀) --- [](https://x.com/Kpaxs/status/1901851683627376655) This is kind of devastating—sorry for dropping it into a light Friday newsletter, but I think it’s true. I feel this even on a day-to-day basis. It’s part of what pushed me to switch to an explore/exploit schedule, because I hate ending the day feeling like, "Oh, I should have been working on something different." It’s especially true when I’ve wasted time scrolling Twitter and finish the day with that sinking feeling of, "This is not what you were meant for. You were meant to do something bigger." --- [](https://x.com/CastanyMiquel/status/1902019260475449622) I don’t do any TikTok at this point, but this was an actually useful video about testing a bunch of different hooks while keeping the same demo in the middle. It’s not something I’d ever thought about before. Most tweets or videos like this are just hype, but this one was very tactical and very practical. --- [](https://x.com/Netlify/status/1902040654085275938) Not too long ago, I [tweeted something about Peak React](https://x.com/aarondfrancis/status/1713639140837765208) being behind us. I think I was directionally correct, but missed the mark a little bit. What I was actually seeing online was more frustration with Next.js, not necessarily React itself. Sure, some folks are frustrated with how fast React moves and then there’s a whole counter-contingent saying it actually doesn’t move that fast. So who knows what to believe. But I do think people are starting to turn a little bit on the complexity of Next.js. This move from Netlify strikes me as incredibly smart. Netlify is obviously competing with Vercel, and Vercel has Next.js. From what I understand, TanStack is super popular, and now Netlify becoming the official deployment partner for TanStack gives them a real rival to the Next.js behemoth. TanStack also seems widely loved by the community, so beyond the technical angle, this feels like a strong goodwill play. I think it’s going to pay off big time for Netlify. --- [](https://www.youtube.com/watch?v=mEWJEhgN8So) This is an interesting [podcast](https://podcasts.apple.com/us/podcast/dan-shipper-the-first-multimodal-media-company/id1709773028?i=1000699009475) from the folks over at Every, where I’m a happy subscriber. It’s the first time I’ve heard someone talk about a multimodal media company, which I thought was super interesting. And if there’s any company that Try Hard aspires to be, Every is definitely directionally correct. I think Every focuses a lot on writing, which we don't do that at Try Hard Studios, but we are a multimodal company. If that is a term, we do open source, we do video courses, at some point we will do software, we do YouTube videos, and we do podcasts like [Mostly Technical](https://mostlytechnical.com/). Every serves as an inspiration as I think about how can we build a company that does several things, but importantly does several things very well. --- [](https://x.com/visakanv/status/1900884930152067553) If you're not following Visa yet, he may be one of the top three accounts on Twitter that I would recommend, so definitely go follow him. This tweet is interesting to me because it touches on something I wrote about a while ago in my [Publishing Your Work](https://github.com/readme/guides/publishing-your-work) article: when you consistently put stuff out there about a topic, people start to associate you with that topic. Then, when they see something related, they tag you—and that creates a kind of self-fulfilling prophecy. The more you talk about XYZ, the more people think of you when they see XYZ. That’s also why I think it’s helpful to have little monikers or catchphrases—pithy things like “You can just do things”, “try hard”, or “maximum effort era”. Stuff like that sticks, and it reinforces the association. --- ### This week at Try Hard Studios **🖥 Solo Terminal Dumps Package** I've been working on [Solo](https://github.com/soloterm/solo)—a tool for running multiple commands simultaneously in local Laravel development. As part of that, I extracted the feature that intercepts and collects `dump()` calls, displaying them in a dedicated terminal window. This keeps your API responses clean and your debug output organized. [Give it a try](https://github.com/soloterm/dumps) and let me know what you think! **🎙️ Podcast Appearances** I joined Robbie and Chuck on the Whiskey Web and Whatnot podcast for a spirited chat over Green River Bourbon. We chatted about Laravel vs Rails, my dream of terraforming land, plus a jovial discussion on minivans, twins, and building a home office that’s not at home! [Give it a listen here!](https://whiskey.fm/why-most-developers-overcomplicate-everything-w-aaron-francis) 🎟️ **Conferences** Next week I am headed to Utah to speak at Kent C. Dodds' [Epic Web Conference](https://www.epicweb.dev/conf/2025). Steve is gonna head over from Boise, so we'll get to hang out in person for a little bit, which is always a delight. I'm sure my talk will be posted at some point, and if it is, I will add a link here. --- That’s it for this week! If you saw something interesting lately, hit reply and share—I’d love to see it. Talk soon, Aaron {{ ENV.unsubscribe\_personal | default: visitor.unsubscribe\_url | hyperlink: "Click here to unsubscribe." }}