Everyone is busy. Nothing ships. Sound familiar?
High activity, zero progress. The team looks productive by every metric except the one that matters: working software in users' hands.
Why this really happens
- Work is being done on things that don't matter to the actual goal
- Too many priorities means nothing is actually prioritized
- Process overhead consumes more energy than the actual work
- Nobody has the authority or courage to say no to incoming requests
Sprints are full. Standups happen every morning. Jira is meticulously updated. The burndown chart shows velocity. And yet — when someone asks “what shipped this quarter?” the room goes quiet.
This is the most insidious project failure mode because it doesn’t look like failure. It looks like work.
Why activity isn’t progress
Too many priorities. When everything is priority one, nothing is. The team spreads across eight initiatives and makes 10% progress on each. After three months, you have eight things that are 30% done and zero things that are 100% done.
Process has become the product. Sprint planning, backlog grooming, retrospectives, architecture reviews, design critiques, stakeholder syncs. Count the hours your developers spend in meetings versus writing code. If it’s more than 30%, the process is the problem.
Nobody can say no. Every stakeholder has “just one small thing.” The VP of sales needs a demo feature. Marketing needs a landing page tweak. Support needs a workaround for a bug. Each request is small. Together, they consume the entire team’s bandwidth. And the actual project sits in the backlog, waiting.
The goal is unclear. Ask five people on the team what success looks like for this project. If you get five different answers, there’s your problem. People are working toward different goals, and the aggregate effort cancels out.
How to diagnose it
The Friday test. Every Friday, ask: “What is different about the product today versus last Friday?” If the answer is “nothing a user would notice” for more than two weeks in a row, you have this problem.
The priority audit. List everything the team is currently working on. Everything. Now stack-rank it. If you can’t — if stakeholders can’t agree on what matters most — that’s the root cause right there.
The meeting audit. Add up the hours each developer spends in meetings per week. Multiply by the team size. That number is your overhead. If it exceeds 25% of total available hours, cut meetings before changing anything else.
The fix
Cut scope brutally. Pick the one thing that matters most. Not three things. One. Put everything else in a parking lot. Tell the stakeholders who are waiting that their thing will happen — after this one thing ships.
Shield the team. Someone needs to be the “no” person. Every incoming request goes through one person who evaluates it against the current priority. If it’s not more important than what the team is already doing, it waits.
Make progress visible. Deploy something — anything — to production every week. Even if it’s small. Visible progress creates momentum, and momentum creates more progress. A team that ships something every week has very different energy than a team that hasn’t shipped in a month.
This isn’t about working harder. It’s about working on the right thing.
Common questions
- Is this a management problem or a technical problem?
- Almost always management. The developers are doing what they're told — the issue is what they're being told to do, and by how many different people.
- How do I tell my team they're not being productive without destroying morale?
- Don't frame it as a productivity problem. Frame it as a focus problem. The team is productive — they're just productive on the wrong things. That's a leadership failure, not an individual one.
- Should I add more people to get the project moving?
- No. More people amplify the existing dysfunction. Fix the focus first, then assess whether you actually need more capacity.
Symptoms you might recognize
Sound familiar?
I help unstick software projects like this. No slide decks — just a conversation about what's actually wrong.
Tell me what's stuck