Picture a founder who hires an agency to build a customer portal. The scope is clear, the budget is fixed and the timeline is three months. At the one-month check-in, the agency shows polished mockups and says everything is on schedule. At two months, it demos a login screen and a dashboard with placeholder data and says the work is mostly done.
Two months past the deadline, the budget is gone and the portal still cannot process a single real transaction. An outside developer reviews the code and finds that most of the backend was never built. The polished screens had very little working behind them, and every update along the way had sounded like progress.
Software projects rarely fail overnight. They slip a little at a time, and the warning signs show up early, but they are easy to miss if you do not know what "off track" looks like from the outside.
The updates are getting vaguer
Early in a healthy project, your development team gives specific updates: "We finished the user authentication flow. Next week we're building the payment integration with Stripe." You can track an update like that.
When a project starts to slide, the language changes to lines like "We're working through some technical challenges" or "We're making good progress on the backend." They sound like updates, but they tell you nothing.
Updates turn vague because the team has no good news to report and hopes to catch up before you notice. It is rarely that the team has forgotten how to communicate.
After every status update, ask yourself whether you could explain what was accomplished that week, in one sentence, to your co-founder or board. If you cannot, treat it as a warning sign rather than a matter of writing style.
You can't see it working
This is the most reliable check we know of, and it requires no technical knowledge: ask your development team to show you the actual product doing real work. Mockups and wireframes do not count. You want the real software, processing real data and doing what it is supposed to do.
A healthy project produces working software at regular intervals. At two weeks, you might see a user log in and land on a real dashboard. At four weeks, basic transactions might work end to end. The features do not need to be polished, but they do need to work. If your team is three months in and can only show you screens with hardcoded data and buttons that do not connect to anything, the project has a problem.
That is the trap in the customer portal example. The agency showed polished UI every two weeks, but the login screen did not connect to a real database and the dashboard graphs were static images. Visual progress looked like engineering progress, and it was not.
Ask for a live demo every two weeks. A slide deck or a screen recording polished before it was sent is not enough. You want an interactive walkthrough where you click around yourself and see what happens. If the team resists ("it's not ready to show yet," "we're focused on the infrastructure right now"), pay attention. Infrastructure work is real, but it does not take three months before anything can be demonstrated.
The timeline keeps moving but the explanation doesn't change
Delays happen in every software project, so a delay by itself is not the warning sign. The warning sign is a timeline that keeps moving while the reason stays the same.
"We hit some unexpected complexity in the integration" is a reasonable explanation once. If you hear it three times in a row, the cause is no longer unexpected complexity. More likely, the team underestimated the work and has not changed its approach.
Take an integration with a supplier's API that the team estimates at two weeks. Four weeks later it is "almost done." Eight weeks later the answer is the same, and so is the explanation: the supplier's documentation is incomplete.
A detailed breakdown often shows a different cause. The developer on the integration has never built one like it, and the two-week estimate assumed someone who had. Nobody told the founder.
A healthy team adjusts and says so: "This is taking longer than we estimated. Here's why. Here's our revised timeline." Compare that with "We're still working on it. Should be done soon," which gives you no plan at all.
The team says yes to everything
A team that agrees to every feature request and never pushes back seems responsive and accommodating, which makes this sign easy to overlook.
In practice, a team that never says no is not managing scope or being honest about tradeoffs. Each new feature it accepts pushes back the ship date, and a team that will not tell you so is building a project that will go over budget. Some of this is structural. In a typical dev shop engagement, more scope means more billing, so telling a client not to build something costs the agency money.
Strong technical teams push back regularly: "We can do that, but it'll add two weeks. Do you want that more than launching on time?" or "That feature is a nice-to-have. Can we ship without it and add it in v2?"
If your team never has that conversation with you, one of two things is happening. Either it padded the timeline so heavily that every addition fits, which means you are overpaying, or it is quietly absorbing scope, and the timeline will eventually give way under everything it agreed to.
What to do when you spot these signs
Start by asking direct questions:
- "Can you show me this working in a live demo?"
- "What specifically got done this week?"
- "Why has this estimate changed three times?"
- "What would you cut if we needed to ship in four weeks instead of eight?"
If the answers are clear, specific and honest, even when the news is bad, you are probably working with a good team going through a difficult stretch. Bad news delivered early is manageable.
If the answers are defensive, vague or keep changing, consider getting an outside opinion. A short code review from an independent senior developer can tell you whether the project can be saved or whether more money will only extend the losses. It costs far less than another month of building in the wrong direction, and it is most useful before the situation feels urgent.
Spotting the problems is easier than acting on them before the sunk cost feels too large to walk away from. A project that is $40,000 over budget and three months behind does not get better by spending another $40,000. Sometimes the right move is to stop, assess and change course, using the same judgment you apply to every other part of your business.




