You know the scene.
The steering committee pack lands. Dozens of slides. Status: green. Budget: green. Risks: green. Morale: not even tracked.
The meeting wraps early. Nobody asked the hard questions. And afterwards, in the corridor, someone pulls you aside and says, “You do know that project’s in trouble, right?”
That’s a watermelon project. Green on the outside. Red in the middle.
And here’s the thing. People are hiding the truth from you. Not because they’re dishonest. Not because they’re incompetent. But because somewhere along the way, your organisation made it more expensive to tell the truth than to bury it.
That’s worth sitting with for a moment. The people closest to the work, the ones who can see the cracks forming, have made a rational decision that staying quiet is safer than speaking up.
Understanding why they made that decision is the key to fixing it.
Why watermelon projects happen
Watermelon projects don’t appear overnight. They grow. Layer by layer, silence by silence, until the gap between what’s reported and what’s real becomes too wide to bridge.
Here’s what’s actually going on.
The environment punishes bad news
This is the root of it. I’ve seen it on every troubled project I’ve worked on. Reporting red triggers a predictable chain of events. Audits. Reviews. Escalations. And somewhere in that chain, someone gets blamed.
People learn fast. They’ve watched what happens to the person who puts their hand up and says “we’ve got a problem.” The scrutiny lands on them, not on the problem. So next time, they soften the language. Report amber instead of red. Or stay green and hope they can fix it before anyone notices.
Researchers call this the “mum effect,” a tendency to suppress bad news that’s been documented since the 1970s. One study found that 85% of employees have withheld important information from their manager because they feared the consequences of speaking up.
Leaders aren’t listening, even when people do speak up
I’ve sat in rooms where the warning signs were clearly laid out and watched them get waved away. Not because the executives were careless. Because they’d already committed to the project, sold it to the board, tied their reputation to it. Hearing that it’s in trouble creates a conflict they’d rather not face. So they discount the warning. Dismiss the messenger. Reframe the problem as manageable.
Researchers call this the “deaf effect.” And it gets worse when the warning comes from someone who isn’t in a formal role to raise it. The developer who sees the cracks, the tester who knows the timeline is fantasy, none of them are in the “right” position to deliver the news. Even when someone in the right role does raise it, if the message threatens a decision already made, it often gets filtered out.
Put those two together, the mum effect from below and the deaf effect from above, and you get a feedback loop. The team stays quiet because speaking up hurts. The exec stays comfortable because nobody’s telling them otherwise. And the project rots from the inside while the dashboard stays green.
The middle layer smooths the message
Between the delivery team and the executive sits a management layer that often acts as an unintentional filter. I’ve seen this pattern repeatedly. They see the problems. But they think they can manage their way out of it before it needs to escalate. Not out of malice. Out of hope.
So they round the edges. Present a cleaner picture than reality supports. But frequently they don’t have the skills, the organisational support, or the authority to actually resolve what they’re sitting on. What could have been a two-week conversation becomes a six-month crisis.
The metrics measure the wrong things
Most projects are measured on schedule and budget. Both are lag indicators. By the time they turn red, the damage is already done.
If your governance only asks “are we on time and on budget?” it’s asking the wrong questions. It should be asking, “are we making good decisions? Are risks being escalated? Do the people doing the work feel safe enough to tell us what’s actually happening?”
Our brains work against us
Even without the cultural factors, cognitive bias stacks the deck. Optimism bias. The planning fallacy. And the illusion of control, where experienced leaders can be the worst offenders because they’ve seen similar situations before and mistake familiarity for certainty. I’ve fallen into that trap myself.
None of this is malicious. It’s human. And it’s why you can’t fix watermelon projects by telling people to try harder. You have to change the system they’re operating in.
Why it matters
Every week that a project reports green when it’s actually red is a week where leaders aren’t making the decisions they need to make. Resources aren’t being redirected. Risks aren’t being mitigated. And the cost of recovery is climbing.
The sooner a problem surfaces, the cheaper it is to fix. The longer it’s hidden, the more it costs when reality finally arrives. And it always arrives.
What to do about it: 10 practical steps
1. Redefine what “red” means in your organisation.
Red is not failure. Red is information. Red, caught early, is the cheapest problem you’ll ever solve. If your culture treats red as someone’s fault rather than a signal for help, you’ll keep growing watermelons.
2. Make the consequences of honesty less painful than the consequences of silence.
If people believe that flagging a problem will result in support, they’ll flag it. If they believe it will result in blame, they won’t. Look at what actually happens to the people who raise bad news in your organisation. Not what the policy says. What actually happens.
3. Reward early kills.
Celebrate the person who stops a failing project before it burns another quarter of budget. That person saved you millions. If the only stories that get told at leadership meetings are the success stories, people will keep manufacturing success stories.
4. Ask the question that matters.
Next time your project dashboard is green across the board, ask one question: “What’s the worst thing nobody is telling me?” If there’s silence in the room, that’s your answer.
5. Get closer to the work.
Don’t rely on the dashboard. Go and talk to the people doing the work. Walk the floor. Sit in on a stand-up. Ask the team lead how they’re actually going, not in a governance meeting, but over a coffee. The truth lives in the corridors, not in the steering pack.
6. Use objective criteria for RAG status.
If your RAG ratings are subjective, they’re meaningless. Define what green, amber, and red actually mean in measurable terms. “25% over budget equals red” is clear. “On track” is not. Same definitions, less room for creative interpretation.
7. Create short feedback loops.
The longer the gap between status updates, the more room there is to hide. If a milestone is six weeks away and there’s no interim measure, nobody knows whether the project is on track or not, not even the project manager. Break the work down. Create shorter checkpoints. Make the truth visible more often.
8. Look at how your managers are incentivised.
If middle management’s bonus depends on presenting a clean portfolio, they will present a clean portfolio. Whether it’s accurate or not. Align incentives with honest reporting, not pretty reporting. And don’t just promote people into leadership roles and assume they know how to manage delivery. Invest in training and mentoring them until their track record demonstrates competency.
9. If you’re the one serving the watermelon, stop.
Rip the band-aid off. Lead with your chin. Get things out in the open. You can’t get any help if you can’t tell them there’s a problem. I know it’s scary. That’s why we’re all in this game. But the longer you wait, the worse it gets. The cover-up is always more expensive than the problem.
10. Don’t assume AI will fix this.
Better dashboards. Real-time data. Automated reporting. AI can speed reporting up and make data more transparent. But the challenge with watermelon projects is not better or faster data. It’s the receipt of bad news and how leadership reacts to it. If the culture punishes honesty, people will game AI reports the same way they game manual ones. The technology is not the bottleneck. The culture is.
The bottom line
Watermelon projects are a system problem, not a people problem. The people on your projects are responding rationally to the environment they’re in. If that environment makes it safer to hide the truth than to tell it, that’s exactly what they’ll do.
The fix is not more governance, more reporting, or more technology. It’s making it psychologically safe to tell the truth.
The less bravery it takes to deliver a difficult message, the more of those messages will come through. And those messages are the ones you need to hear.
No responses yet