As I’ve gotten older, I’ve progressively gotten busier and busier. My work now feels more real, so I'm working more than ever before. I have many more hobbies now than I ever have before, and I'm also more into them than I used to be. I have many great friends who I want to keep in touch with. I have more responsibility over my own life; I need to think much more about things like housing and bills and money and what I will eat today. The list goes on. As someone who likes doing things, this to me initially seemed pretty great. I like the feeling of always working on something, of always making forward progress in some domain. Always being occupied and always feeling productive feels great...

Until it doesn't. In this post, I enumerate a bunch of thoughts on what goes wrong when you try and do too much.

Problem 1: Doing many things in parallel has costs.

We all have heard the phrase "don't spread yourself too thin". This rationally made sense to me, but I now also viscerally feel why it is bad to do so. Doing many things decreases immersion in each individual thing. It's extremely hard to obsess over anything if you have 10 other things you also need to be thinking about during some given time period.

I care a lot about self-improvement, about getting better at the things that I do. I find doing many things mitigates my ability to utilize this muscle, and makes everything go fine, but not amazingly. Paul Graham argues that in order to do a really good job on anything, it's probably important for it to live at the top of your mind. Many things cannot live there simultaneously. If I'm doing many things, I actually find ~nothing lives at the top of my mind, because my mind is always distracted by the next task on my stack.

Problem 2: Not having slack has costs.

As I got busier, I eventually filled all hours in the week with stuff. I had no slack. Here are some problems with not having slack:

• (a) Suppose you work on many things, and because you are human, suffer from the planning fallacy—the phenomenon where humans reliably predict things will take less time than they actually do. If you have no slack and you suffer from the planning fallacy, this just means you have to drop some balls. You either finish the task and drop something else or half finish the task and move on.

• (b) Some 'emergency' comes up, by which I just mean an unforeseen kind of urgent thing, and you must spend time dealing with it. Because you have no slack, you once again have to drop some balls to deal with it.

In my experience either (a) or (b) happens most weeks. Both of these issues stem from our poor ability to make accurate forecasts of the future. Our lives are unpredictable, so slack is important.

A further issue I notice in myself is that when balls get dropped, a certain kind of ball is dropped first. When insufficient time is budgeted to some project, what happens is the easy to measure metrics saturate, but the fuzzier harder to quantify aspects of the project disappear. An example here might be dating. It's very easy to measure "how many dates am I going on per week". But that's not all that's needed. To date successfully, one first needs to source dates, then go on the dates, then reflect on the dates, then reflect on the process that generated the dates — and then if dating goes extremely well, dedicate time to making the relationship work. You might not initially realise this is all required, or have good models of how long each part of the process might take, but if you have no slack you will find it very hard to both realise this and do something about it.

Problem 3: You actually just don't know what you want or need.

The way I personally naturally end up being busy is by blocking out time on some recurring schedule to do each of the things I want to spend my time doing. A problem with being naively busy in this way is that you might end up reward hacking yourself by choosing the wrong things to spend your time doing. You should model yourself as having uncertainty over your goals and your preferences over how you should spend your time. Consider also that various parts of your life might contain instances of Chesterton's fence -- that is, that there might exist things in your life that are good for you for reasons you are unable to explain. To give an example, I was recently reflecting on why it is that I haven't made many new close friends recently. I think the reason is that I haven't had that much slack. Forming close friendships requires at least one person to exert some agency to create circumstances where bonding can occur. I model myself as having some fixed optimization budget, and was not dedicating any of it to making new friends. But back in the day when my calendar was less full, this is a thing I would sort of randomly and naturally end up doing through random exploration.

What I'm doing

Here's what I'm doing to combat these problems:

(1) I think hard before taking on some new thing. I never really used to do this. I used to find taking on new things pretty universally fun, perhaps due to some novelty bias that I have. I now viscerally feel the opportunity cost of taking on or committing to new things.

(2) I have a taxonomy of things I do, containing 3 priorities.

• (a) Active priorities: These are things I currently care about getting better at and making real progress in. Most new things I do will start out here, as some up front investment is often necessary to build habits / become sufficiently competent in some new domain. This set must be very small.
• (b) Maintenance: These are things that have often been active priorities in the past, where I am reasonably happy with the current state of things, and think things will go reasonably without active intervention. It is important to accept here that progress here will be slow / non existent, but that is OK. I find that many things in my life can eventually end up here.
• (c) As needed: These are things I will do sometimes, but mostly reactively. I will not set aside regular time to do this, but will have to every so often. These get almost no mind space. I find this to be in good balance between wanting and needing to do a lot of things, but also dedicating enough effort to some small set of important things.

(3) I have weekly regularly scheduled ‘slack’. I'm allowed to fill this slack, but only in the week prior. This is time set aside for planning fallacies or emergencies.

(4) I have revived my daily journalling habit. In some sense, this is also 'slack'. It is time set aside to introspect, and to think about whatever seems most important that day. I often end up realising I am randomly dropping some ball or goodharting some goal in these sessions. Another way of thinking about this time is as something like "exploration" time, to consider modifications to my life routine that might improve it on the margin.

I wrote the first draft of this blog post by hand in my journal several days ago. Thanks to Gemini 3.0 Pro for digitizing it.

Thanks also to Neel Nanda, Jeevan Fernando and Josh Engels for helpful feedback and discussion.