How to end a project early when it's going badly

A project goes wrong in one of two ways. Either it deteriorates slowly — scope shifts, communication stalls, you start dreading the client’s name in your inbox — until both sides pretend it’s still fine. Or it blows up fast and neither of you knows what comes next.

The instinct is to push through. Walking away feels like failure. It’s not. Staying on a project that’s consuming more than it’s worth is the actual cost.

Signs the project is unsalvageable

Not every difficult project should end early. Difficulty is normal. What’s not normal:

• The scope has changed so fundamentally that the original agreement is irrelevant
• The client is consistently unresponsive, then urgent, with no middle ground
• You’ve already absorbed significant overrun and there’s no acknowledgment, let alone compensation
• The relationship has become adversarial — every delivery turns into a dispute
• You’ve documented concerns in writing and nothing has changed

One or two of these might be recoverable with a direct conversation. All of them together is a pattern.

Try the conversation first

Before you exit, have one clear conversation. Not an email thread — a call or a meeting with an agenda.

State what you’re seeing, not what you’re feeling. “We’re three weeks past the original scope and I’ve raised this twice. The project as it stands is no longer viable at the current terms.” That’s a fact. It opens a door to renegotiation, reprice, or a clean stop.

Some clients don’t realise how far things have drifted. The conversation clarifies it. If they respond constructively, you might not need to exit at all. If they don’t, you’ve created a documented record that you raised the issue.

How to exit

Give proper notice. What’s proper depends on the project — a retainer warrants more than a one-off deliverable. Even for projects with no contractual notice period, a week is the minimum. Two weeks is professional.

During the notice period:

• Deliver everything that’s genuinely complete
• Document what’s in progress and where it stands
• Write a handover note: what files exist, where they live, what’s outstanding
• Don’t abandon anything mid-task if you can avoid it

The handover isn’t optional. Even a project that went badly leaves a client who needs to continue the work somehow. A handover protects you from being blamed for what happens after you leave. It also makes payment disputes harder to sustain — you delivered what you said you’d deliver.

What to invoice

Invoice for every billable hour you’ve worked. Don’t negotiate your time away as a concession for leaving.

A common mistake: freelancers discount the final invoice to soften the exit. It doesn’t soften anything. It just means you worked for less than your rate on work the client accepted.

If your contract includes a kill fee clause, that applies here too — in the sense that you may be owed payment for capacity blocked. Read what your contract actually says before you assume nothing is owed beyond hours worked.

Invoice clearly. One line per deliverable or billing period. The more specific the invoice, the harder it is to dispute. If you’ve been using your calendar to log client time throughout the project, you have the record to back it up.

What to say

You don’t owe an explanation beyond what’s professional and accurate. Something like:

I’ve decided to step back from this project. I don’t think we’re aligned on scope and terms, and continuing wouldn’t be in either of our interests. I’ll deliver [X] by [date] and provide a handover document. My final invoice will cover work completed to date.

That’s it. No apology for things that weren’t your fault. No long post-mortem. If they want to discuss it, you can — but the exit itself doesn’t require negotiation.

Protecting what you’re owed

Once you’ve given notice, some clients slow down on payment. They know leverage disappears at delivery.

• Send the final invoice the same day as the exit notice, or immediately after you deliver the final work
• Attach your contract or reference the agreed terms if there’s any question about what’s owed
• Follow your standard escalation sequence if they don’t pay — this is just a late payment situation like any other

The exit doesn’t change what they owe you for work already done. Don’t let the discomfort of the situation make you accept less than that.

After you exit

Two things worth doing:

Write one sentence about why the project went wrong, for yourself. Not to send — just to record. Often the root issue is visible in hindsight and avoidable in future client selection. Vague briefs, scope discussed verbally instead of in writing, a client who pushed back on the contract from day one.

Then move forward. One difficult project doesn’t define your practice. The clients who aren’t worth finishing are also, usually, the ones who wouldn’t have been worth keeping.

If you’ve been tracking client time in your calendar throughout the project, Timescanner gives you the full billing record to close out the invoice accurately — even on a project that didn’t go to plan.