Most mid-market organizations reach a point where the functional structure that helped them scale from fifty to two hundred employees starts actively working against them. Sales and marketing are not aligned on pipeline definitions. Product and operations are solving problems the other team already addressed. Finance and IT cannot agree on project prioritization because neither has visibility into the other’s constraints. The instinctive response from leadership is to consider a reorganization — perhaps a matrix structure, perhaps a centre of excellence, perhaps a new layer of integration managers. In our experience, that instinct is almost always wrong. Reorganizing is expensive, disruptive, and frequently reproduces the same siloed behavior inside the new structure within twelve to eighteen months. The underlying problem is not the org chart. It is the operating model: specifically, how decisions are made, how accountability is assigned, how objectives are shared, and whether the organization has built any structural mechanisms for cross-functional relationship formation.

Why functional silos persist despite good intentions

The standard diagnosis is cultural: people are territorial, leaders are protective of headcount, functions optimize for their own metrics. That is partially true, but it mistakes symptom for cause. In organizations we work with, functional silos persist for three structural reasons that culture change cannot address on its own.

First, incentive structures are misaligned. When a VP of Operations is evaluated on cost reduction and a VP of Sales is evaluated on revenue growth, they are not being irrational when they fail to collaborate on a customer experience initiative that requires one function to absorb cost for the other’s benefit. The system is working exactly as designed. No amount of townhall messaging about collaboration changes the underlying calculation.

Second, decision rights are ambiguous. When it is unclear who has the authority to decide — not just who has input, but who actually makes the call — decisions escalate to the level where authority is clear: the executive team. In mid-market companies with executive teams of five to eight people, this creates a bottleneck that slows every cross-functional initiative and trains the organization to avoid cross-functional work because it takes too long to get resolved.

Third, there are no structural mechanisms for relationship formation. Cross-functional collaboration depends on trust, and trust at organizational scale depends on repeated interaction. When functions sit in separate parts of the building, attend separate meetings, and measure separate metrics, the informal relationships that enable fast, low-friction collaboration simply do not develop. This is not a personality problem. It is a design problem.

The accountability architecture: assigning ownership for shared outcomes

The first intervention in an effective cross-functional operating model is the creation of what we call an accountability architecture — a formal assignment of ownership for outcomes that span functional boundaries. This is distinct from responsibility assignment matrices like RACI, which are useful for project-level work but do not address ongoing operational accountability.

The accountability architecture has three components:

• Outcome ownership: One named individual — not a team, not a function, not a committee — owns a cross-functional outcome and is evaluated on it. This person does not have to manage all the people involved in delivering it, but they are the one answering for the result. In practice, this is often a General Manager, a Chief of Staff, or a senior director with a remit that deliberately crosses functional lines.
• Contributing accountability: Each function with a role in delivering the outcome has a specific, measurable contribution defined. Not “support the initiative” — a concrete deliverable, timeline, and quality standard. This contribution appears in that function head’s performance review.
• Escalation clarity: When the outcome owner and a contributing function head disagree, the path to resolution is written down before the disagreement happens. Typically this means the outcome owner can escalate to a defined executive sponsor within forty-eight hours, with a defined timeframe for resolution.

The practical work of building an accountability architecture starts by inventorying the five to ten outcomes in your organization that most frequently get stuck at functional boundaries. For most mid-market B2B companies, these include customer onboarding, product launches, enterprise sales cycles, technology implementations, and annual planning. For each outcome, map which functions touch it, where handoffs happen, and where decisions currently get stuck. That map tells you where accountability gaps exist.

Decision rights: removing escalation bottlenecks at the source

The most reliable indicator of a cross-functional operating model that is not working is executive calendar share. When executives are spending more than thirty percent of their time resolving decisions that their direct reports could not agree on, the organization has a decision rights problem, not a talent problem.

Effective decision rights frameworks for cross-functional contexts distinguish between four types of decisions:

The common mistake is classifying too many decisions as collaborative or executive when they are actually consultative. In our experience, approximately sixty to seventy percent of decisions that currently reach the executive level in mid-market organizations belong in the consultative category — one function actually bears the consequence, but the norm of seeking consensus before acting has created a false requirement for agreement. Training function heads to distinguish between “I need your input” and “I need your agreement” resolves a significant portion of escalation volume.

The other structural fix is eliminating the standing invitation to escalate. When any function head can bring any cross-functional disagreement to the CEO’s staff meeting, the incentive to resolve disagreements below that level is weak. Replacing open-ended escalation with a structured process — written brief, defined timeframe for executive review, explicit consequence for failing to attempt resolution at the working level — changes the incentive structure without removing the safety valve.

Shared OKRs: building collaborative intent into the planning process

Objectives and Key Results have become the planning framework of choice for most growth-stage and mid-market organizations, and for good reason — they create clarity, measurability, and cadence. They also, when implemented poorly, entrench silos by giving every function a set of objectives that are wholly owned and independently measured.

The modification that creates cross-functional alignment is the introduction of shared objectives — OKRs where two or more function heads are co-owners of a single objective, with key results that reflect each function’s contribution to a common outcome. This is structurally different from coordination between adjacent objectives. When two function heads share an objective, they share credit and share accountability for the result. The incentive to collaborate is built into the performance system.

Shared OKRs work when:

• The objective names the shared outcome explicitly. “Improve enterprise customer time-to-value” is a shared objective. “Sales closes deals faster” and “Operations implements faster” are adjacent, unshared objectives that will not produce the same collaborative behavior.
• Key results are differentiated by function. Each co-owning function has key results that reflect their specific contribution. Sales might own “reduce average contract-to-kickoff time from 14 to 7 days.” Operations might own “reduce average implementation time from 45 to 30 days.” Both roll up to the shared objective.
• Review cadence is joint. The quarterly OKR review for a shared objective includes both function heads in the same conversation, with the outcome owner facilitating. Separate reporting eliminates the collaborative signal.

The planning process itself — not just the OKR document — needs to change. Most mid-market organizations run functional planning in parallel and then attempt to align across functions at the end. The result is objectives that were designed in isolation and retrofitted with cross-functional language. The alternative is to run a cross-functional planning workshop before functional planning begins, where the three to five most critical shared outcomes for the year are identified and owned jointly, and functional planning is then built to support those shared outcomes rather than alongside them.

Rituals: building cross-functional relationships at scale

The word “ritual” is used deliberately here rather than “meeting” or “process.” A ritual is a structured, recurring interaction with a defined social and relational function beyond its task output. Organizations that sustain cross-functional collaboration over time do so not because their people happen to like each other, but because they have designed recurring interactions that build familiarity, surface interdependencies early, and create the relational capital that makes informal coordination possible between formal meetings.

The rituals that consistently produce cross-functional relationship value in mid-market organizations are:

• Cross-functional standup (weekly, 30 minutes, function leads): Not a status report — a structured exchange of blockers and upcoming decisions that cross functional lines. The agenda is fixed: what each function is working on that will affect another function in the next two weeks, and what they need from other functions to move forward. The discipline is in keeping it to thirty minutes and following up offline.
• Shared problem-solving session (monthly, 90 minutes, rotating topic): One cross-functional challenge is brought to the group with a brief written context document distributed forty-eight hours in advance. The session is structured to produce a decision or a clear next step, not a discussion. Function heads rotate ownership of bringing the challenge.
• Cross-functional working team (project-based, time-bound): For specific cross-functional outcomes — a product launch, a system implementation, a new market entry — a working team with representation from each function is formed with a defined mandate, a named lead, a weekly cadence, and a sunset date. The team is dissolved when the outcome is achieved or the mandate expires. This prevents the proliferation of standing committees that outlive their purpose.
• Retrospective (quarterly, 60 minutes, cross-functional): A structured look at what cross-functional work went well, what created friction, and what one change the group will make in the next quarter. This is where the operating model itself is held accountable and adjusted.

The mistake organizations make with rituals is adding them without removing others. Cross-functional rituals consume time and will be abandoned if the overall meeting load is already high. Before introducing new cross-functional rituals, audit existing meetings for ones that can be shortened, reduced in frequency, or eliminated. The net time cost of a well-designed cross-functional ritual program should be close to zero for most function heads.

Implementation sequence: what to do first

Organizations attempting to implement all of the above simultaneously will not succeed. The right sequence matters. In typical mid-market deployments, the highest-leverage starting point is decision rights — specifically, cataloguing the ten to fifteen cross-functional decisions that most frequently get stuck and assigning clear decision authority for each. This produces visible, fast results (decisions start moving) and builds the credibility for the larger changes that follow.

The accountability architecture comes next, because it requires naming outcome owners and building contribution accountability into performance systems — changes that need to be tied to the next performance management cycle to be credible. Shared OKRs follow, aligned to the next annual planning cycle. Rituals are introduced after the structural foundations are in place, because rituals without accountability and decision clarity become talk shops that produce frustration rather than collaboration.

Frequently asked questions

We have tried cross-functional teams before and they always revert to silos. What are we doing wrong?

In most cases, the answer is that the cross-functional structure was added on top of functional accountability without changing the underlying incentive system. When a function head’s performance review is based entirely on functional metrics and a cross-functional team asks them to prioritize shared outcomes, the rational choice is to deprioritize the cross-functional work when functional demands increase — which they always do. If the cross-functional team’s outcomes are not reflected in individual performance accountability, the team will not sustain collaborative behavior under pressure. The fix is not better facilitation or stronger mandates from the top. It is changing what people are measured on.

How do we handle the politics of naming a single outcome owner when multiple function heads feel they have equal claim?

This is a real and common obstacle, and there is no clean way around the discomfort. The CEO or COO needs to make the call on outcome ownership when function heads cannot reach agreement. The practical move is to name the outcome owner before the accountability architecture is built, not after — ownership decisions are harder to make once the stakes are visible. It also helps to separate outcome ownership from resource authority: the outcome owner does not need to manage the contributing functions to be accountable for the result. That distinction reduces the perceived status stakes of the decision.

Our executive team is the bottleneck. How do we get them to push decision-making down?

The most effective approach is to show executives the cost of the current pattern concretely. Map the decisions that escalated to the executive level in the last quarter, calculate the average time from escalation to resolution, and multiply by the cost of delay for each decision type. For most mid-market organizations, this calculation produces a number that is large enough to motivate change. The follow-on step is to give executives a clear alternative: a decision rights framework where they are explicitly reserved for irreversible, high-stakes decisions, with a defined process for everything else. Executives rarely resist decentralization when they understand what they are being freed to focus on.

How do shared OKRs work in practice when functions have very different planning cycles?

This is a legitimate structural tension, particularly between functions like finance (annual planning dominant) and product or technology (quarterly or six-week cycles). The solution is to set shared objectives at the annual level — where alignment is feasible across planning cycles — and differentiate the cadence of key results by function. A shared objective like “reduce time-to-decision for enterprise customers” can have a finance key result measured annually and a product key result measured quarterly. The shared objective creates the collaborative intent; the differentiated key result cadence respects operational reality.

What is a realistic timeline to see results from these changes?

Decision rights changes produce visible results within sixty to ninety days if the framework is applied to real decisions rather than documented and filed. Accountability architecture changes take one performance cycle — typically six to twelve months — before the behavioral impact is clear, because people respond to what they are measured on at review time, not at announcement time. Shared OKRs show impact after the first full OKR cycle with genuine shared accountability. Rituals build relational capital gradually; most organizations report meaningful improvement in cross-functional relationship quality after two to three quarters of consistent practice. Expecting transformation in thirty days is the most common way organizations abandon changes that would have worked.

Cross-Functional Collaboration: The Operating Model That Removes Silos Without Reorganizing

Most senior operations and strategy leaders know their organization has a cross-functional collaboration problem but underestimate how much of it is structural rather than cultural. This post offers a concrete operating model — accountability architecture, decision rights, shared OKRs, and purposeful rituals — that addresses the structural roots of siloed behavior without the disruption and cost of reorganization.