DEV Community

opah
opah

Posted on

How I plan a week as a freelance developer with several clients

I'm a freelancer. I own two small companies and I also work as a developer for other companies. For a long time my week was decided by whoever pinged me first.

I still work for more than one company at a time. Each one has its own tasks, its own calendar, and its own idea of what is urgent. If I do not make a plan, the first message of the day wins, and Friday arrives with work I meant to do and never started. The method below is how I plan now, and it does not depend on a particular app.

One list for everything

I cannot plan a week from five different places. If each company's work lives in a different tool, I will only see whatever I happen to open. The rest waits until I remember it, or until someone asks why it is late.

So I put everything I am supposed to do on one list. Work for the companies I own and work I do as a developer for other companies all go in the same place. I mark who each item is for, because a list that mixes clients without a label makes the hours hard to explain later.

The list is for this week, not for every open task I have ever been given. I add work that is due, work I already started, reviews waiting on me, and anything I promised in a call that never became a formal task.

I do not update the list from memory. When something changes, I go back to the place the work actually lives, then I update the list. If a new request comes in on Tuesday, it does not automatically join today's work. It goes on the list first. I decide later whether it fits.

A Sunday plan

Once a week, usually on Sunday, I sit down with that list and with every calendar I actually use. I want to know what must move this week, what can wait, and how much time I can honestly give each company.

I look at the meetings first. They are already real. A day with two calls and a review due is already a full day. I write down the hours I can still use after those meetings, and I divide them on purpose: some for this client, some for that one, some for my own companies.

Then I pick, for each company, a short set of things that should be done by Friday. Not everything that is open. If the list cannot fit into the hours I actually have, I cut the list. I would rather start the week knowing what will slip than discover it on Thursday.

I also leave a little room. Someone will ping me. If I pack the week edge to edge, the first new request breaks the plan. I write the plan down. A plan I keep in my head is too easy to abandon at the first message.

Time blocks per client in the calendar

A list tells me what the week contains. It does not stop me from jumping between companies every half hour, which is a bad way to write software. I lose the thread and come back to the work with less care than it needs.

So I put time on the calendar, per client. After the meetings are in place, I block the remaining time for one company's work at a time: a morning for one client, an afternoon for another, a later block for my own companies. I decide in advance whose work this block belongs to.

If a day is already full of calls, I do not also plan a deep piece of work for a third client into it. When something new and urgent arrives, I can change the blocks, but I change them on purpose. I move or cut something, and I put the new work in a real slot. I do not let it eat whatever I happened to be doing when the message landed.

The calendar is also how I protect the companies I own. Client work is good at filling every gap. If I do not give my own companies a block, they get whatever is left, which is often nothing.

Tracking time as you go

I do not wait until Friday and try to remember the week. I track time as I go, on the task I am actually doing.

When I start a piece of work, I start a timer or I write down the time, against that task, not against a vague bucket with a client's name on it. When I stop, I stop the entry. If I switch companies, I switch the entry too. The point is to know, midweek, where the hours already went.

This helps with invoicing, and it also helps when I am not billing by the hour. Midweek I can see that I have already used the time I set aside for one company, and that another has had almost none. That is the moment to adjust the remaining blocks, not Friday afternoon.

At the end of the week I look at those lines before I start the next plan. If the Sunday plan and the tracked time disagree, I believe the time.

Notes next to tasks

Work for several companies produces a lot of small decisions: what was agreed on a call, what I said I would do next, why I chose one approach over another. If those notes live in a random document, or only in my head, they are gone when I need them.

So I keep notes next to the task they belong to. After a call, I write down what we decided and what I owe, on that piece of work. After a review, I write the next step. If I stop mid-task, I leave a line about where I am, so I can start again without rebuilding the context from scratch. I do the same for my own companies.

If I open the task tomorrow, I should be able to see the state of it without searching my inbox. When I close the week, I can see what moved, what I owe someone, and what should go on next week's list.

I ended up building a tool for this, opah (https://opah.dev). You can import your work from Jira, Linear, GitHub, GitLab or Notion, and it's free during early access. It'll cost $8 a month later, with 30 days' notice. There's a demo with no account at https://opah.dev/demo.

Top comments (0)