How to Split a Freelance Project Into Milestones
Milestones are the backbone of most freelance projects. Done well, they give the client regular proof of progress and give the freelancer regular payment. Done badly, they become arbitrary dates that everyone quietly ignores.
Here is the checklist I follow when splitting a project, whether I'm the client writing a brief or the freelancer writing a proposal.
Step 1: Work backwards from "done"
Start at the end. Write one or two sentences describing the finished result in terms you can check.
Then list what has to exist for that result to be true. For a small web app, that might be:
- Agreed designs for key screens.
- Working sign-up and login.
- Core feature working end to end.
- Payments connected.
- Deployed to production with monitoring.
That list is your first draft of milestones. It's rarely the final version, but it's grounded in the outcome rather than in arbitrary weeks.
Step 2: Make each milestone testable
A milestone should end with something the client can look at, click or read. Run each one through these checks:
Visible deliverable. A demo, a file, a deployed URL, a merged pull request.
Clear acceptance. One or two lines describing what "approved" means.
One owner. It's clear who delivers and who approves.
No hidden dependencies. If the client must provide content or access first, it's written down.
Independent value. If the project stopped after this milestone, the client would still have something useful.
"Backend work" fails these checks. "API for bookings, documented and deployed to staging" passes.
Step 3: Size and order them
Size matters as much as content. Some rules of thumb:
- Keep the first milestone small. A few days of work, not a few weeks. It tests the relationship cheaply on both sides.
- Aim for one to three weeks per milestone after that. Longer stretches hide problems.
- Put the riskiest work early. If one integration might not work, find out in milestone two, not milestone six.
- Leave polish for later. Visual refinements and nice-to-haves go near the end.
- Include a final handover milestone. Documentation, access transfer and a walkthrough deserve their own line.
If you end up with more than six or seven milestones for a modest project, some can probably be merged.
Step 4: Attach dates and money
Every milestone gets a delivery date and, if payment is staged, an amount.
Dates:
- Include review time for the client, not just build time for the freelancer.
- Leave a little buffer before anything with an external deadline.
- Note dates that depend on the client delivering something first.
Money:
- Weight amounts by effort, not evenly by count.
- Consider a small deposit before the first milestone for new working relationships.
- Agree what happens if a milestone needs rework: included revisions versus new work.
Dated milestones also leave a useful trail. Some freelance platforms turn them into a record; on Workli, proposals are built around milestones with delivery dates, and on-time delivery in a freelancer's verified history is calculated from those dated milestones.
Step 5: Review together and adjust as you go
Before work starts, read the milestone plan together and check:
Both sides agree on every deliverable and acceptance line.
Dates are realistic given both schedules.
Payment amounts and timing are clear.
The change process is agreed: what happens when scope shifts mid-project.
Once the project is running, treat milestones as a living plan, within limits. It's fine to split a milestone that turned out bigger than expected, or to reorder two that don't depend on each other. What you shouldn't do is quietly move dates without saying so. If a date changes, write down the new date and the reason, and confirm it with the other side.
A few warning signs that the plan needs attention:
- Two milestones in a row are late.
- A milestone is "almost done" for more than a week.
- The client hasn't reviewed the last deliverable.
- New requests are landing inside an existing milestone without a conversation.
Good milestones make a project feel smaller and safer for everyone. Spend the time to get them right at the start, and you'll spend much less time arguing about progress later.


Comments 0
Replies to this note, oldest first.
No replies yet. Found it useful? Say so below.