The Dirty Jobs That Actually Compound for Indie Founders (And the Ones That Don't)
TL;DR: Everyone talks about working on your business. Nobody talks about which work actually compounds. The dirty jobs, the unsexy ones, the ones that feel like maintenance rather than progress — those are the only ones that create leverage over time. Everything else is busywork dressed up as productivity.
I spent three weeks going through comment threads, founder interviews, and every piece of honest post-mortem data I could find. The pattern was ugly and clear. Most indie founders are busy with the wrong things. The ones who break through are doing the same dirty jobs, over and over, long after everyone else has moved on to something shinier.
Here is what I found.
The Compounding Dirt
Most founders hear "compound interest" and think about revenue. They should be thinking about work. Some tasks get better the more you do them. Some tasks make other tasks faster. Some tasks create momentum that doesn't disappear when you take a week off. Those are the dirty jobs worth doing.
Customer Discovery, Repeatedly
This is the one every founder knows they should do. This is the one every founder does once, then stops because it feels slow, uncomfortable, and doesn't produce immediate numbers.
The data is unambiguous. Founders who do 10+ customer interviews before building anything fail less often. Founders who keep doing discovery after launch catch problems earlier. Founders who do it weekly catch shifts in what their users actually need versus what they said they needed three months ago.
The dirty part: it requires talking to people who might not like your product. It requires hearing that your baby is ugly. It requires sitting with discomfort instead of retreating to building.
The compounding part: each conversation makes the next one easier. You learn to ask better questions. You develop a calibration for what "real user" sounds like versus "polite listener." After 50 conversations, you have a sixth sense for when someone is telling you what you want to hear. That sense is worth more than any feature you've built.
Writing Changelog Updates Like a Human
Most changelog entries are "Fixed bug" or "Improved performance." Boring. Meaningless. Nobody bookmarks them. The founders who use changelogs as a compound channel write them like they would a letter to a friend who also happens to use their product.
The changelog becomes content. The changelog becomes an email to power users. The changelog becomes a signal to people watching your product that someone is actually paying attention to what they're saying. Over months, this builds a narrative that is nearly impossible to replicate through a blog alone.
The dirty part: it takes time. Each update has to be specific. Each update has to explain not just what changed but why it changed. Why you made the call you made. What tradeoffs you considered. That level of transparency feels risky.
The compounding part: it creates an archive. A new user lands on your changelog page and, for the first time, understands what kind of company you are. A power user reads the changelog and feels seen. A potential customer finds it through search and decides to try the product. All of this happens without you writing a single blog post.
Responding to Every Support Request Personally
This sounds insane for a scaling business. It is insane for a scaling business. It is also the single fastest way to find out what your users actually need versus what you think they need.
Indie hackers who do this for the first 100 customers build something close to a sixth sense for user pain. They know which questions keep coming back. They know which feature is genuinely missing versus which one people are asking for out of confusion. They know which frustrations are worth addressing immediately versus which ones are noise.
The dirty part: it takes hours. You are answering the same questions repeatedly. You are holding the user's hand through things that feel basic. You are doing work that, by all rational measure, should be automated or delegated.
The compounding part: the feedback loop is so tight that you catch problems before they become retention issues. You also catch opportunities. Every support conversation is a market research session you're being paid to run. Most founders pay for this research and don't even know it.
The Dirty Job Nobody Talks About: The First Customer Email
Every founder knows cold emails are a channel. Most founders write five, get no responses, and conclude cold email is dead. The founders who have this work for them do it differently.
They write to one person. They write about that person's specific problem. They write as if they have no product to sell, just a question they want answered. They send it as an email, not a pitch. They follow up once, three days later, with a simple question mark. Nothing else.
The dirty part: it requires research. Real research, not just "I looked at their LinkedIn." It requires understanding their business well enough to say something useful. It requires being comfortable with a 5% response rate and treating that as information rather than failure.
The compounding part: every email teaches you something about your pitch. Every response teaches you something about the market. After 100 of these, you know exactly what language works and what doesn't. This knowledge compounds faster than any channel hack.
The Jobs That Look Like Compound Work But Don't
Here is where it gets uncomfortable. There are things that feel like progress, that feel like the dirty work, that feel like you're doing the hard thing. These are not compound work.
Building in Public Without a Point
Most "building in public" content is "look at what I did today." This is not building in public. This is journaling. The founders who actually get traction from building in public are telling a story with a point. The story has stakes. The story has a problem worth solving. The story involves the audience in the actual decision-making process, not just the output of decisions that have already been made.
Building in public without a point creates noise. It creates the appearance of progress without the substance. Your followers feel engaged because you're posting regularly. None of that engagement converts because nothing was offered worth engaging with.
Chasing Platform Algorithm Changes
Every week there is a new rule about what X, or LinkedIn, or whatever platform wants this week. Founders who spend time on this are doing work that resets every time the algorithm changes. There is no compounding here. There is only maintenance.
The thing that compounds on social is: writing content that people want to read regardless of when they find it. That is hard. That takes months to develop. That does not change when the algorithm does.
Adding Features Because Competitors Added Them
Your competitor added a dark mode. Now you need dark mode. Your competitor added an API. Now you need an API. This is not product development. This is following.
The dirty jobs in product are the ones that create unique value for your specific users. These are the features your users asked for that nobody else thought to build. These are the integrations your users need that aren't sexy but are essential. These are the performance improvements that nobody sees but everyone feels.
None of this shows up in a screenshot. None of it goes viral. All of it compounds.
What This Looks Like Day to Day
The indie founders who do the dirty compound jobs have a specific rhythm. It looks unglamorous from the outside. It looks like they're always in support mode, always doing something that could be automated or delegated. They are. That's the point.
They do discovery calls on Tuesday afternoons. They write changelogs like they have a deadline. They answer the first 50 support emails personally even when they have a support tool. They send cold emails that read like they were written by someone who actually knows something.
The gap between this and what everyone else is doing is enormous. Most founders are doing visible work. The compound work is invisible, unsexy, and takes months to pay off. When it pays off, it pays off in ways that are impossible to replicate with a single viral post or a single growth hack.
The One Question That Tells You If You're Doing the Right Dirty Job
Before you start anything, ask: will this get better the more I do it? Will this teach me something that makes the next one easier? Will this create something that exists after I'm done for the day?
If the answer is no, you're doing maintenance work. Maintenance is fine. Maintenance is necessary. But maintenance does not compound. Maintenance does not create leverage. Maintenance is what happens when the compound work hasn't been done yet.
If the answer is yes, do that thing for six months straight. Don't optimize. Don't improve the process. Just do the thing. Watch what happens.
Luka connects your analytics, error data, reviews, and social signals, finds what they're saying together, and gives you daily focus items matched to where your product actually is. You check it once in the morning and go work on what it tells you. See how Luka works.
Frequently Asked Questions
How do I know if a task is actually compound work?
Ask: will this get better the more I do it? Will doing this 100 times make me faster, smarter, or more accurate at it? If yes, it's compound. If no, it's maintenance.
Should I stop doing everything else to focus on compound work?
No. You still need to maintain the business. The goal is to identify which maintenance tasks can be minimized or eliminated so you have room for the compound work. Most founders discover they've been doing far too much maintenance and far too little compound work.
How long until I see results from compound work?
Months. Not weeks. The work compounds slowly and then all at once. The founders who quit after two months were doing it wrong. The founders who did it for a year had something nobody could replicate quickly.
My competitor is growing faster. What do I do?
Stop looking at your competitor. They might be doing visible work that doesn't compound. They might be doing dirty work you can't see. You don't know. The only thing you control is your own compound work. Do that long enough and the competitor question stops mattering as much.
How many customer conversations should I be having?
Minimum one per week indefinitely. If you're not doing discovery because you think you know what your users need, you're wrong. User needs shift. Markets shift. Your understanding of your users should be updating constantly.
About the Author
