When two teams have goals that quietly compete

Two teams are working on the same initiative. Both are busy, both are competent, and every few weeks something lands late or gets done twice. When you ask why, you get two accounts of the same events that are each internally consistent and mutually incompatible.

That is the signature of misaligned goals. Look there first, because it sits upstream of the other three failure modes. Teams that do not share a destination cannot share a route, and no amount of better communication will fix a disagreement about where you are going. If you have not read the argument this sits under, the short version is that this is a structural problem rather than a relational one.

The mistake that costs the most time

The usual response is to schedule a working session and hash it out. It feels productive and it usually produces a list of agreements that nobody can execute, because the thing being negotiated in the room is not the thing that is actually binding.

Misalignment shows up in three distinct forms, and they need different fixes. Diagnose which one you have before you spend anybody's afternoon on it.

Form one: nobody knows who owns what

Your teams disagree about who is responsible for which piece of work.

Write one RACI across both teams, not one per team. Most role confusion survives a documentation effort because each team documented its own half and the two halves do not touch. Every task gets one accountable owner, and that owner sits on one team or the other, not both.

Then write a charter and ask to be corrected. Spell out your team's mission and what you own. Send it to your cross-functional stakeholders and explicitly invite them to disagree with it. The corrections are the deliverable. A charter nobody has argued with is a charter nobody has read.

Agree two or three shared measures. This is the step people skip and it is the one that makes the rest hold. A RACI is a document; a shared metric is an incentive. When both teams are judged on the same two or three numbers, the ownership question stops needing to be relitigated every quarter.

Put a short standing slot in the calendar. Thirty recurring minutes to walk live work and settle who is doing what while it still matters, which is a different meeting from a status update and should not be allowed to become one.

Form two: the goals genuinely compete

This one is harder, because everybody is doing their job correctly and the jobs pull apart.

Run an alignment session, and name the trade-off out loud. Put both sets of goals on one page. Find where they overlap, and find where they truly compete. The second half is the point. Most alignment sessions produce a warm feeling because the participants quietly agree to focus on the overlap and leave the conflict undiscussed. It comes back in six weeks under a new name.

State the goal above both teams. There is a business unit or initiative objective that both of you serve. Say it, then keep saying it often enough that people on both teams can repeat it without looking it up. This sounds like corporate throat-clearing until you watch a disagreement get settled by somebody citing it.

Break it into objectives with a date and an owner. A shared goal that nobody can act on this quarter is a poster.

Here is the hard part, and it is worth knowing before you begin. If the two sets of goals compete because someone above you wrote them that way, no session you run will resolve it. You can surface it, document it, and escalate it. You cannot fix it from your seat. Going in with that clear saves you from mistaking a structural problem for a facilitation problem.

Form three: you cannot settle disagreements

The goals are aligned and you still deadlock every time priorities collide.

Build the prioritization process before you need it. Decide how competing work gets ranked while nothing is at stake, and get senior leaders on both sides to sign off on the method rather than the outcome. Negotiating the rules during a live conflict is how one team ends up feeling overruled.

Say what is realistic. Agree service levels for the work that passes between you, and revise them when the work changes rather than when somebody complains.

Let your team work it out first. Check in on how cross-team work is going and resist the urge to solve it. Step in with your counterpart only when your people are genuinely stuck. A leader who intervenes early trains both teams to escalate everything.

How you know it worked

The signal is that disagreements get settled at a lower level and without you. They will not stop, and a boundary with no disagreement across it usually means one team has given up on the other.

Thirty days in, ask two questions. Did anything get ranked using the new process rather than by whoever pushed hardest? And did a disagreement resolve without reaching you? If the answer to both is no, the diagnosis was probably wrong and one of the other three modes is the real problem.

The exercise

Take twenty minutes with your own team before you take anything to anyone else.

Write down the three pieces of work that cross the boundary most often. For each one, name the single accountable owner. Then send that list to your counterpart on the other team and ask them to correct it.

Where their list differs from yours is the whole problem, and you now have it on one page.