On a typical Tuesday, I run three streams of work that don’t naturally coexist. Client work pays the rent this month. Content compounds for next year. Admin protects everything from going sideways. None of the three would dominate if I let them. All three would, if I didn’t have a structure for them.
The minimum viable system is three time blocks per stream per week. Nine blocks total. Each block has a fixed slot, a fixed length, and a single output. The blocks are the system. Everything else fits inside them, or doesn’t fit at all.
The three streams
Client work is the now stream. It pays the bills this month. Deadlines are real. Other people are waiting. If you let it grow, it will fill the whole week because it has the strongest external pressure.
Content is the future stream. It compounds. A piece written today produces returns next quarter, next year, and possibly five years from now. It’s also the easiest stream to skip because nothing breaks immediately when you do. The cost shows up later, which is when most solo founders notice they have no audience and a full client calendar.
Admin is the protect-the-baseline stream. Invoicing, taxes, contracts, infrastructure, the bookkeeping you’ll regret skipping. None of it produces revenue on its own. All of it produces operational drag if it’s missing. A solo business with no admin stream is a business that pays cash to its accountant in March and explains itself to a tax authority in April.
The classic solo failure is one stream eating the others. Client work crowds out content. Content overshoots and the invoice goes unsent. Admin gets attention only in quarterly panic. Two of the three streams always get worse when one of them takes over.
The nine-block grid
I run three blocks per stream per week. Nine blocks total. Each block is 90 minutes – long enough for real work, short enough that I can hold the focus without it degrading. Nine times 90 minutes is 13.5 hours of focused output per week. Everything else around the blocks is calls, sorting, communication, and the unstructured space the system needs to absorb whatever turned up that day.
The breakdown looks like this in practice:
- 3 client blocks – project work on the active client deliverables
- 3 content blocks – writing, drafting, editing, or the upstream research the writing requires
- 3 admin blocks – invoicing, taxes, contracts, system maintenance, the bookkeeping batch
The numbers don’t have to be even. Some weeks I run four client blocks and two content blocks. Some weeks I run two client blocks and four content blocks. The total stays around nine because that’s the boundary of focused capacity for one person, in my experience. Designing capacity instead of chasing it covers why nine isn’t twelve or fifteen.
Block placement (the rule that holds it)
Where you put the blocks matters more than how many you run.
Client blocks go early in the day, mid-week. Mornings are the highest-focus window for most people, and mid-week is when client momentum is real. Monday is often spent absorbing inbox. Friday is when deadlines compress. Tuesday and Wednesday mornings are the strongest client-work slots.
Content blocks also go early in the day, but on different days. Mine sit on Sunday afternoon (the long-form draft), Tuesday morning (the editing pass), and Thursday morning (the publication-ready version). Putting two streams in the same morning slot is how streams collide.
Admin blocks go late afternoon or batched into a single longer session once a week. Admin doesn’t need peak focus; it needs reliability. A 4 PM admin block on Friday is more productive than a 10 AM admin block on Tuesday, because the energy cost is lower and the urgency is higher – everything you didn’t admin during the week shows up by Friday afternoon.
The rule that holds the grid: never two streams in the same block.
Paul Graham described the deeper version of this in Maker’s Schedule, Manager’s Schedule. Switching streams mid-block doesn’t just cost the time of the switch. It costs the time of regaining the focus after the switch. Sophie Leroy’s 2009 research on attention residue found that the residue of a previous task contaminates the next task for measurable time. A 90-minute block split between two streams produces less output than either stream would have produced in a 45-minute block alone.
What happens when streams collide
Sometimes a client deadline week is real. Sometimes a content launch demands extra editing. Sometimes admin breaks – an invoice gets disputed, a contract needs rewriting, a tax filing collapses.
When streams collide, the rule is substitution, not addition.
If a client week needs four blocks instead of three, you substitute. You skip one content block or one admin block this week. The grid stays at nine. The total stays inside designed capacity. You don’t add a tenth block by stealing from the evening, the weekend, or the next morning’s sleep.
Most solo founders try addition. They keep all their normal blocks and stack the client overflow on top. That works for one week. It produces a deficit by the second week. By the third week, the deficit has compounded enough that all three streams are losing output simultaneously.
Substitution is uncomfortable because it looks like giving up on something. It isn’t. It’s acknowledging that the week’s capacity is fixed and the streams have to take turns when demand spikes. The skipped block returns next week. The deficit doesn’t.
The system block
One of the nine blocks is for the system itself.
Mine sits on Friday afternoon: 90 minutes for the weekly review (what worked, what didn’t, what changes), the four decisions for next week, and the capacity check (is next week reasonable or am I planning to peak?). This isn’t extra work. It’s the Review part of the Solo Operating System – the loop that adjusts the system so the next week is better than this one.
Skipping the system block is the common move because it doesn’t feel productive. The blocks that produce visible output feel productive. The block that maintains the structure that produces the output doesn’t. The skip is the most expensive choice in the grid. Without the system block, the grid drifts. Decisions don’t get made. Capacity isn’t checked. The same gaps reappear week after week without anyone noticing the pattern.
The system block stays in the nine. It counts as one of the admin blocks – maintenance of the system is admin, not bonus.
What scales and what doesn’t
The grid scales down. If your week is tighter – seasonal slow period, illness, a different season of the business – run six blocks instead of nine. Two per stream. The structure holds at a lower volume. The work happens at a smaller scale, which is fine.
The grid does not scale up. Twelve or fifteen blocks per week doesn’t mean more output. It means more switching cost and less focus per block. The constraint is not your calendar. It’s the depth of focus you can produce in a day. Beyond about three deep blocks per day, the quality degrades fast. A solo founder who runs twelve blocks for a quarter doesn’t ship more than one who runs nine. They produce more meetings about the work and less of the work itself.
The grid also isn’t the whole week. The blocks are the structured core. The unstructured time around them absorbs messages, sorting, follow-up, casual reading, conversations, and the slow background work that finishes projects in the next quarter. That unstructured time is part of the design, not waste.
The rest of the Solo Business archive – including the cluster closer on rebuilding the system – lives at the Solo Business category.
FAQ
Why 90 minutes per block?
Long enough for a single output – a section of writing, a piece of client work, a batch of admin. Short enough that focus doesn’t degrade in the back half. The 90-minute heuristic comes out of ultradian rhythm research, but in practice it’s just the longest stretch most people sustain without a quality drop. Adjust by 15 minutes either way if your data says different.
What if I have multiple clients?
The client stream isn’t per client. It’s the total stream. Three client blocks distributed across two or three clients works fine. The rule is that one block doesn’t span two clients – pick one client per block and finish a discrete output for them.
How does this interact with the four decisions?
The four decisions are made on Sunday and define what fills the blocks. Decision 1 (what the week is for) tells you which stream gets the extra block this week. Decision 2 (three things to finish) often maps directly to three of the nine blocks. The grid is the execution layer. The decisions are the upstream layer.
What if I’m not a content business?
Replace the content stream with whatever your future-stream is – product development, audience growth, business development, partnership outreach. The principle is the same. Three streams, each with a different time horizon (now, future, protect-baseline), each with a protected slot.
Do I need to time-block in a calendar app?
No. I keep the grid on paper, beside my keyboard, replaced each Sunday. A calendar app works fine if you already use one. The format isn’t the system. The visibility is. Whatever lets you see the nine blocks at a glance on a Tuesday afternoon is the right tool.
Containers, not work
The block isn’t the work. The block is the container that lets the work happen. When the container is solid, the work fills it without leaking into the next one. When the container is loose, every stream tries to fill the whole week.
Most solo founders don’t have a stream problem. They have a container problem.
The Solo OS Starter Kit. A printable four-part frame, a five-question audit, and a 15-minute weekly routine. The container the nine blocks sit inside. Download the kit →
