Ask a room of leaders to name the biggest challenge they are facing right now, and one answer keeps coming back. “Do more with less”. Fewer people, a flat or shrinking budget, a hiring freeze that never lifted, contractors gone, and a commitment list that has not lost a single line.
One team leader put it in numbers that are hard to argue with. Her team lost thirty percent of its capacity and zero percent of its scope.
That sentence is the whole problem. The constraint is rarely the issue. Constraints are normal, and good teams have always worked inside them. The issue is that capacity moved and expectations did not, and nobody in the room was willing to say so out loud!
What the phrase usually means
Strip away the framing and doing more with less tends to mean one of three things.
- It means people. A layoff happened, or attrition was never backfilled, and the message underneath is that the same work still has to get done. One UX designer now covers three teams. The specialist who used to unblock everyone is gone.
- It means tools. The budget for the collaboration platform disappeared, so the free whiteboard bundled with the operating system is the new standard. It is not a great product, but it is what is installed, so it is what people use.
- It means roles. Do we really need a Scrum Master and a Product Owner? Do we need a Product Owner and an analyst? Let’s just combine them! The Scrum Master in particular gets treated as an expensive luxury, right up until a transformation goes sideways and somebody asks who is holding the team together. The answer is often that the Product Owner is doing both jobs and neither one well.
The newest version of the argument
There is a fresh flavor to this in 2026, and it is arithmetic being done in spreadsheets rather than with teams. We will keep four humans and add a handful of agents. That is the same as a team of ten, is it not?
Picture a car with its wheels spinning in the mud. Someone suggests the fix is to pile more weight into the trunk. The wheels spin harder and the car sinks deeper. Adding capability to a system that is already stuck does not unstick it. It buries it.
The clearest symptom of this thinking is the proposal to run the daily scrum through an agent so the team does not have to hold it. That does not make an organization more agile. It removes the one place a team synchronizes, replaces it with a report, and calls the result efficiency. It is a sign that the purpose of the event was never understood in the first place.
The same logic applies to frameworks. If team level practice is poor and you scale it, you get more poor practice at greater expense. If organizational agility is poor and you add AI, you get more of what you already had, faster. You do not get agility!
Efficient is not the same as effective
This is the distinction worth writing on a wall.
Output and outcome are different things, and the difference is where all the money hides. AI will get you output faster. It may also build the wrong thing five times quicker than you could have built the right one. Five more features that nobody uses, with excellent documentation, is not doing more with less. It is doing more waste with less time to notice.
Here is the test. If output goes up and outcomes stay flat, waste has gone up. That is the whole measurement, and it explains why velocity has become close to meaningless as a signal. Track cycle time. Track features used against features shipped. Track whether the thing you delivered changed anything for a customer.
Efficiency is doing the work faster. Effectiveness is doing work that matters. A constrained team that only optimizes the first one gets very good at running in the wrong direction.
Where the tooling does earn its keep
None of this is an argument against AI. There is a real and specific list of things it can take off a team’s plate, and most organizations are not using it for any of them.
It can spot patterns in flow that a human would need weeks to see, including where work is queuing and which WIP limits are being quietly ignored. It can find the handoffs in a delivery process that are costing the most time. It can build the dashboard that replaces a standing status meeting. It can absorb the documentation theater that nobody reads and everybody produces. It can pull context together across the six tools a team is forced to work in, so a person does not spend their morning shuttling between a file store, a chat channel, a whiteboard, and their inbox to assemble one picture.
Call it the coordination tax. That tax is real, it is expensive, and it is a legitimate place to hand work to a machine. It can also do a solid first draft of backlog items, which saves a Product Owner an hour of writing and lets them spend that hour on the part that needs judgment.
What it does not do is critical thinking. It knows what has been fed to it. It does not know your customers, your industry, or the conversation someone had at a conference last month that changes how you see a problem. It thinks inside the box we built for it, and it is trying hard to please you, which is not the same as being right. Half of what it produces will be exactly what you wanted, some will be interesting, and some will be nonsense. All of it needs checking, and checking takes the time you thought you saved.
The bill arrives later
The immediate damage is visible if you look for it. Teams absorb the gap in silence because nobody feels safe enough to say pause, we need to prioritize. They put in the hours instead. Burnout climbs. Engagement drops. People quietly disengage long before they quietly leave.
The quality damage is slower and more expensive. Work gets pushed out before it is ready. Output from an agent goes unreviewed. Testing gets compressed because there is no time. None of that saves money. It moves the invoice to next year and adds interest.
And there is a longer bill that almost nobody is pricing. Entry level work is the most obvious thing to automate, so the junior pipeline stops getting filled. Save thirty million over five years, and then ask a harder question. Who is your senior person in 2035? Who has the experience, the judgment, the empathy, and the critical thinking, and where were they supposed to have learned it? You may not need a large human workforce, but you are still building products for humans. Lose the ability to think about a customer as a person and the products follow.
Make the trade-offs visible
The answer to “can you do more with less” is never yes and never no. It is yes, and here is what moves.
- Yes, we can use agents to help here, and this will be the impact on review time and quality.
- Yes, we can deliver with a smaller team, and here is the trade-off we will make on scope.
- Yes, we can keep going, and we need the work prioritized so we are building in that order.
That is not resistance. It is the visibility that agility depends on. Absorbing the gap quietly feels loyal, and it is the single most damaging thing a team can do, because it deletes the only signal leadership has that anything is wrong.
Bring data rather than feelings. We are drowning gets sympathy. Here is our throughput, here is what we lost, and here is the forecast gets a decision.
Do less, better
If the scope has not changed and a quarter of the staff has gone, the plan does not work. Nine people cannot produce a baby in one month, and a turkey does not cook faster at a higher temperature. It burns. Cook it at half the heat and you do not get the same bird either.
So change the plan. Cut the portfolio. If there are five projects, stop two and make the remaining three excellent. Hold WIP limits hard, so cycle time comes down and context switching stops eating the day. Reduce dependencies between teams. Build T shaped people and feature teams. Keep one backlog as the single source of truth. Cut the coordination meetings that exist to manage the complexity you could have removed instead.
None of this is new. That is the point. When something as disruptive as AI arrives, the instinct is to reach for something novel, and the right move is to go back to the basics and do them properly for the first time. This is the moment those practices were designed for.
There is a clean deal to offer the business, and it works. You tell us the priority order. We will meet our commitments. At the end of the sprint the team can say it took in what was agreed, finished all of it, and rolled nothing over. Most friction with stakeholders comes from missed commitments, and a smaller team that reliably delivers a shorter list is in a far stronger position than a larger one that reliably disappoints.
Solve the problem in the container
Network television is a very particular container. There are things you cannot say and things you cannot do, and the job is to make something great inside those limits rather than to complain about them.
Every team has a container. This year yours got smaller. You lost a tool, or a teammate, or three. Rage at the storm for five minutes, because that is fair, and then sit down together and work out how to do the job inside the container you have. That is continuous improvement, and it is what agility looks like when it is not a slogan!
The line to take home
You cannot do more with less. You can do less, better, and being honest about which less is the actual work.
Efficiency will be handed to you. Effectiveness is still a choice, and it is still yours to make!