Two Things Every App Needs
AI is making the barrier to building software nearly zero. So how do you stand out? The answer lies in two simple things: a changelog and a feedback page.
AI is slowly making the barrier to building software nearly zero. Which means if you're a solo developer or a SaaS startup, you're competing with thousands of other people who can also ship some basic version of what you're doing using Lovable faster than you sometimes.
So how do you stand out?
For a while now, developers have been slowly turning into content creators. You used to just ship code and not have to deal with marketing and sales because there were teams for that (mostly). This has been changing because it's now easier than ever to get started on social media, and more importantly, building in public is the new way of building anything. It's easier than ever to see small and large organizations shipping so-called beta's of whatever product it is.
We're in a strange period. It's become normal for companies to ship things that don't work yet. The Fisker car that Marques Brownlee reviewed had so many issues he titled it "the worst car I've ever reviewed." The Rabbit R1 didn't work as a device and it's slowly dying. The Humane AI pin is already dead and sold to HP. But this isn't only apparent in physical products - it's in software too. It feels more than ever as if all software is broken. Cursor, my favorite IDE, crashes on my machine constantly. I'm pretty sure you're familiar with the Cloudflare stories. Even big orgs like Google and Microsoft constantly ships things that don't work according to their promises.
It used to be that products had to work fully out of the box. Now? Companies ship the dream first, then build toward it publicly. WhatsApp (of all things), which used to be rock solid, has bugs now. Everyone's shipping more code, and everyone's dealing with more bugs because of it.
So what separates the products that gain users from the ones that don't? Simple. TRUST and a Great Dream of what your product could be. The ones winning are inviting users into the journey and showing them that they are here to stay through commercials and ads that show how amazing the product could be or is (although half the features usually don't work as expected). Users are doing more work shaping what products evolve to be and if you want to do the same as well the best way to do this is to show two things: you're committed and shipping regularly, and you welcome feedback and iterate quickly based on it. Let's dive deeper.
1. Changelog
The Browser Company (makers of Arc) probably brought this to life. They did weekly videos showing what they shipped, and they had a changelog page that pops up every time you opened Arc. It built a community of people who flocked to Arc because those guys were shipping like crazy, and people genuinely loved seeing that.
Perplexity does this every other week. Zen browser does it. OpenAI with CODEX does it. Many more have changelog pages now. It's become a thing, especially in this AI era, because it shows "hey, we're actively working on this thing and we hope to make it better. Here's what we're doing, here's what we've done, here's how we're getting you closer to the dream of what this application will eventually do."
A changelog essentially shows everyone you're moving. One thing is for sure: we like companies that ship and innovate. Everyone loves to see their favorite company or startup do weekly updates showing what they shipped because we relate progress to positivity. Add a changelog to every single idea you build from now on because it's the easiest way to become a "code-influencer" without having a mic and camera.
2. Feedback Page
A feedback page shows you're not just building what YOU want. It shows "hey, you want this? that's what I'm gonna do." Make it visible - what's done, what's pending, what people are asking for. Users don't always know what they want, but they love knowing you listened.
It's also a great way to build community around your product because users can meet in one place where they all need the same thing, so discussions flow naturally. Raycast is a master of this. Their feedback system lets users vote on what gets built next. Linear does this too - their roadmap is public, you can see what's being worked on, what just shipped. Google used to do this with Google Labs, where they'd test different ideas publicly and build communities around mini-apps before deciding what to fully develop.
In this age, there are a lot of factors that make a product go viral. Two of which are above and fairly easy to do. The most important one: "having a product that sucks less and delivers on the core functionality that it promises." If your product does those two things, it's not guaranteed to fail (unless you're solving for a really niche set of people or something that isn't worthwhile), but you're guaranteed to learn either way.
"Products fail, founders don't."
The lessons you learn from failure are far more important than ones you learn from successes. Products fail, but founders don't. So ship it and show it. Someone will see.