Why finishing work matters more than starting new work
You open your board on Monday morning. There are twelve cards in progress. Two more in review. Nothing moved to done last week. You feel busy. But nothing is shipping. The board tells the story clearly: too much started, not enough finished.
Finish work before starting new WIP means checking your board for completable items before pulling anything new into progress. It's a habit, not a methodology. Before you grab the next task from the backlog, scan your in-progress column and ask: can I close something first? If yes, do that. The goal is to keep work moving forward instead of spreading your attention wider.
TL;DR: Starting new tasks feels productive but finishing delivers value. Before pulling a new task, check if anything in progress can be finished first. Group your board by status to spot stuck work. Build the habit of closing before opening.
Why does starting new work feel more productive than finishing?
Starting is easy. You read the brief, open a fresh file, sketch the first idea. There's momentum and novelty. Your brain gets a dopamine bump from something new.
Finishing is harder. The last 20% of a task involves the tedious parts: edge cases, reviews, documentation, final approvals. It's less fun. So you unconsciously avoid it by picking up something new instead.
The result is a board full of half-done work. Each card in progress represents a promise. When none of those promises get delivered, the project stalls even though people are working all day.
This isn't a character flaw. It's a pattern that emerges naturally when there's no habit to counteract it.
What does a board look like when nothing gets finished?
Picture a website redesign project with five team members. The board has four status columns: Backlog, In Progress, In Review and Done.
Before (the overloaded board):
| Backlog (6) | In Progress (14) | In Review (5) | Done (2) |
|---|---|---|---|
| SEO audit | Homepage layout | Logo options | Project brief |
| Blog migration | Navigation redesign | Color palette | Hosting setup |
| Analytics setup | Footer design | Mobile wireframes | |
| Contact form | Hero section copy | Icon set | |
| Testimonials page | About page copy | Typography specs | |
| FAQ page | Responsive breakpoints | ||
| Image sourcing | |||
| CMS templates | |||
| Performance audit | |||
| Accessibility review | |||
| Social media assets | |||
| Email template | |||
| Client portal page | |||
| Pricing page layout |
Fourteen cards in progress. Five in review. Two done. That's 19 items started and only 2 completed. Every person on the team has at least three cards open. Nobody has capacity to review the five items waiting for feedback. So those cards sit in review for days, blocking the people who submitted them.
After (the same project, with the habit applied):
| Backlog (12) | In Progress (5) | In Review (2) | Done (8) |
|---|---|---|---|
| Blog migration | CMS templates | Pricing page layout | Project brief |
| Analytics setup | Responsive breakpoints | Email template | Hosting setup |
| Contact form | Accessibility review | Homepage layout | |
| Testimonials page | Social media assets | Navigation redesign | |
| FAQ page | Client portal page | Logo options | |
| SEO audit | Color palette | ||
| About page copy | Mobile wireframes | ||
| Image sourcing | Footer design | ||
| Hero section copy | Icon set | ||
| Performance audit | Typography specs | ||
| Pricing page layout | |||
| Email template |
Five cards in progress. Two in review. Ten done. The total number of tasks is the same. But work is flowing through the board instead of piling up in the middle. People finish before they start something new.
How do you build the habit to finish work before starting new WIP?
This is a daily habit, not a process overhaul. It takes about two minutes at the start of your day and costs nothing to try.
Step 1: Open your board grouped by status
Before you do anything else, open your project board. Group it by status so you can see every column at once. In Eigenfocus, your board view groups by status by default. You see Backlog, In Progress, In Review and Done (or whatever custom statuses you've set up) side by side.
Look at the In Progress and In Review columns first. Not the backlog.
Step 2: Ask three questions
For each card you own in progress or in review:
- Can I finish this today? If yes, that's your first task. Not the interesting new thing in the backlog.
- Is this blocked? If it's waiting on someone else, flag it. Add a comment, ping the reviewer, do whatever unblocks it. A card that sits in review for a week is work that nobody benefits from.
- Should this go back to backlog? Sometimes you started something and realized it's not the right time. That's fine. Move it back. Honest backlog is better than a crowded in-progress column.
Step 3: Only then pull from backlog
Once you've addressed everything in progress and in review, check the backlog. Pick one item. Just one. Start it and work on it until it's done or until you've made meaningful progress.
The constraint is simple: don't start a second new task until the first one moves forward.
A real morning routine using this habit
Here's what this looks like in practice for someone on a product development project.
8:55. Open Eigenfocus. Open the board view for "Q3 Platform Update."
9:00. Scan In Progress. You have two cards: "API endpoint for user preferences" and "Update onboarding flow." The API endpoint is 90% done. You just need to write the test and push it. The onboarding flow is maybe 30% done.
9:05. Scan In Review. One card: "Database migration script" has been in review since Thursday. Your colleague hasn't looked at it yet. You ping them on Slack with a link to the PR.
9:10. Decision: finish the API endpoint first. It's close to done. An hour of focused work will move it to Done.
10:15. API endpoint done. Tests passing. Card moved to Done.
10:20. Back to the board. In Progress now has one card (onboarding flow). In Review still has the database migration. You pick up the onboarding flow and work on it until lunch.
13:00. After lunch, the database migration review came back with minor comments. You address them in 15 minutes. Card moves to Done.
13:20. Two cards done today. One still in progress. Now you pull one new task from the backlog: "Design settings page." That's the only new item you start.
By the end of the day, you finished two tasks and made progress on two others. Compare that to starting three new tasks and finishing none. The throughput is better even though it feels slower at first.
If you're setting up a project for the first time and want statuses that support this habit, see how to set up a project workflow from scratch.
Does limiting work in progress actually improve throughput?
Yes. This is one of the most studied ideas in operations management. The principle originates from Toyota's production system in the 1950s and was later formalized in queuing theory.
The logic is straightforward. When you have 14 items in progress and 5 people, each person is splitting attention across roughly 3 tasks. None of those tasks gets full attention. Finishing takes longer. Handoffs take longer. Review queues grow. The total time from start to finish for any single item increases.
When you limit in-progress work to, say, one or two items per person, each item gets full attention. It finishes faster. The review queue shrinks because there's less waiting. And the Done column fills up more consistently.
You don't need to enforce formal WIP limits with a tool. The habit alone does most of the work. Check before you pull. Finish before you start. That's it.
How does your board view help you finish work before starting new WIP?
The key is visibility. If your board doesn't clearly show what's stuck, you won't catch it.
In Eigenfocus, board views group by status by default. You see every status column side by side. Each card shows its assignee, labels and any custom fields you've enabled for that view. When you open the board, the shape of the columns tells you the story immediately. A fat In Progress column and a thin Done column means work is piling up.
You can also group the same board by assignee. This shows each person's cards in their own row, so you can see who has too many items open. If someone has six cards in progress, that's a signal: they need to finish something before taking on more.
For deeper views on how grouping options help, see when one board view is not enough.
What happens when finishing becomes the default?
When a team adopts this habit, a few things change within the first week or two.
Review queues shrink. When people prioritize finishing, they also prioritize reviewing other people's work. The In Review column stops growing because reviews happen faster.
Estimates get more accurate. When you finish tasks instead of leaving them open, you learn how long things actually take. A task you estimated at 4 hours that sits open for 3 days gives you no useful data. A task you estimated at 4 hours and finished in 6 teaches you something.
Standup meetings get shorter. "I'm working on three things" becomes "I finished one thing yesterday and I'm continuing another today." Clearer updates, fewer follow-up questions.
The board becomes trustworthy. When cards move steadily from left to right, the board reflects reality. People start relying on it for status instead of asking each other for updates. That alone saves time every day.
Frequently asked questions
Is this the same as Kanban WIP limits?
Not exactly. Kanban WIP limits are a formal constraint: you set a maximum number of cards allowed in a column and the board enforces it. The habit described here is lighter. It's a personal check you do each morning before pulling new work. You don't need tooling support or team-wide agreement to start. Just ask yourself: can I finish something before I start something new?
What if I'm blocked on everything in progress?
Then pulling a new task is the right move. The habit isn't rigid. It's a check, not a rule. If every in-progress item is genuinely waiting on someone else, start something new. But also spend five minutes unblocking: send the follow-up message, ping the reviewer, escalate the dependency. Blocked work that stays blocked is work that never ships.
How many items should be in progress at once?
There's no fixed number that works for every team. A common guideline is one to two items per person in active progress. If you have five people and fifteen items in progress, that's a sign things aren't finishing. The right number is whatever keeps work flowing without creating a bottleneck at review.
Does this apply to teams or just individuals?
Both. Individuals can adopt the habit on their own, today. Teams benefit more when everyone does it because the review queue effect compounds. When the whole team prioritizes finishing, reviews happen faster, which unblocks more work, which leads to more finished items. It creates a positive cycle.
Won't this slow down the team if people wait instead of starting new work?
It feels slower at first because you're starting fewer things. But finishing faster means delivering faster. A team that starts 20 tasks and finishes 5 per week delivers less than a team that starts 8 and finishes 7. The second team ships more value even though they started fewer things.
Start finishing
The habit is simple. Every morning, open your board. Look at what's in progress. Ask if anything can be done today. Do that first. Then pull new work.
You don't need a new methodology. You don't need to restructure your workflow. You just need to check before you pull.
Eigenfocus gives you the board views to make this visible. Group by status to see where work is stuck. Group by assignee to see who's overloaded. Custom statuses to define your real workflow stages. It's built for teams that want to see what's happening and act on it.
Explore how teams use Eigenfocus for kanban, project management or software development.