Projects

How to Spot a Project in Trouble Before It’s Too Late

Projects rarely fail overnight: they send signals weeks in advance. Eight warning signs, how to confirm them, and what to do in the first two weeks.

By Randall Cascante ValverdePublished 3 min read

Almost no project goes from green to red overnight. What usually happens is a slow slide: a dependency slips, an assumption nobody validated turns out false, a team works overtime to hold a date that’s no longer realistic. Meanwhile, the status report stays green, because nobody wants to be the one delivering bad news.

The problem isn’t that projects run into trouble. It’s discovering the trouble when there are no cheap options left. Spotting the signs early is the difference between a scope adjustment and a crisis in front of leadership.

Eight signs a project is heading for red

1. The report is always green. A complex project with no amber in months isn’t a healthy project: it’s a project where the report stopped being a management tool. Be wary of uniformity.

2. The sponsor has disappeared. When decisions wait weeks because the sponsor isn’t available, the project runs on momentum. Unanswered questions pile up and turn into rework.

3. The date didn’t move, but the scope did. If requirements were added and the date stayed the same, someone is absorbing the difference, usually the team with overtime or testing with less time.

4. Nobody owns the dependencies. Enterprise projects depend on other teams, vendors, and projects. If those dependencies have no owner and no date, the delay will come from where nobody was looking.

5. The plan is a document, not a tool. A plan last updated a month ago says nothing. Ask the team what they’re working on this week and compare it to the plan.

6. Testing is getting squeezed. When development runs late and the date is fixed, testing is the first thing cut. It’s the surest way to push the problem into go-live.

7. The risks haven’t changed in months. A risk log that never changes means nobody is managing it. Real risks change: they appear, materialize, or close.

8. The team is tired. Turnover, absences, and silence in meetings are early signs. An exhausted team doesn’t raise problems: it hides them.

How to confirm the diagnosis

One isolated sign doesn’t doom a project. Three or more justify a review. To do it objectively:

  • Talk to people, not just reports. Short interviews with the sponsor, the lead, two or three team members, and key vendors tell you more than any dashboard.
  • Check the plan against the evidence. What’s actually done? What’s left? How long did similar tasks take?
  • Review governance. Who decides, with what information, and how often? Many crises are, at heart, decisions that never got made.
  • Assess each dimension separately: scope, schedule, cost, team, and risk. A project can be fine on budget and badly off on schedule.

For a quick first read, our project health self-assessment gives you a result in five minutes.

What to do in the first two weeks

Once the problem is confirmed, the temptation is to speed up: more people, more hours, more meetings. It almost never works. What does work:

  1. Tell leadership the truth. A red project acknowledged in time has options; a hidden one doesn’t. Present the real status with causes and alternatives, not just the problem.
  2. Stabilize before replanning. Freeze non-essential scope changes and make sure the team works on what matters most.
  3. Replan with data. Use the team’s actual pace, not the hoped-for one. If the date isn’t negotiable, the scope has to be.
  4. Strengthen governance. A short weekly check-in with the sponsor, a risk and dependency log with owners, and documented decisions.
  5. Look after the team. Recognize the effort, remove work that adds no value, and shield the team from interruptions.

Prevention is cheaper than recovery

The signs in this article can be watched from day one with simple practices: a charter with agreed goals and scope, a RAID log reviewed weekly, and an honest executive report. You can download our project charter, RAID log, and status report templates for free.

And if you already have a project that worries you, a two-week health check will tell you, with evidence, whether it will land, what’s holding it back, and what to decide first.

Back to insights
Let's talk