JV Success Starts with Governance
A joint venture is built with two bosses by design. It only works when governance is as practical and enforceable as the shareholder agreement.
Joint ventures are everywhere in the Gulf economy. Sovereign-backed groups partner with international operators to bring capability into the region. Multinationals partner with local groups to access markets, licenses, and relationships. Family conglomerates partner with technology providers to build new ventures. The JV is often the right structure. It is also, structurally, the hardest one to run.
Here is a useful way to think about it. An acquisition is an integration with a deadline: painful, but it ends, one owner, one direction, one operating model. A joint venture is an integration that never closes. Two owners, each with its own strategy, its own reporting requirements, its own risk appetite, and its own definition of success, permanently. Years spent inside post-merger integrations at GE teach exactly what breaks when two organizations must share processes, systems, numbers, and decision rights. In an acquisition, those breakages are transition problems. In a JV, the same breakages are the permanent operating condition. That is why the disciplines that rescue integrations, explicit decision rights, one version of the numbers, engineered interfaces, relentless cadence, are not optional in a JV. They are the design.
The shareholder agreement resolves the legal questions. It rarely resolves the operating ones.
Opposite Objectives: The Fault Line Underneath
Before governance, there is a more fundamental question that too many JV negotiations rush past: do the partners actually want the same thing from this venture?
I lived this when working for a scale-up, working on a joint venture project with one of the giants of the optical industry. The deal ultimately fell apart, and the core disagreement was not valuation or control. It was that the two sides had opposite objectives for what the venture should achieve. No shareholder agreement, however well drafted, can reconcile partners who are building the same company for different reasons. Walking away was the right outcome, and it taught a lesson that no successful deal teaches: objective alignment is not a preamble to the negotiation, it is the negotiation.
The same fault line runs through many of the most important ventures in this region: partnerships between Western companies and local groups managing strategic assets, water, energy, utilities, infrastructure. The international partner is there for returns, market access, and contract economics. The local partner is there for security of supply, sovereignty over a strategic asset, and the transfer of capability to national teams. These objectives are not wrong, and they are not secret. But they are structurally different, and they pull the venture in different directions on every operational question: pricing, investment pace, localization, technology transfer, dividend policy.
The mature response is not to pretend alignment in the MOU and discover the divergence in year two. It is to name the divergence at the start and engineer the governance around it: which objectives the venture serves, in what order, measured how, and what each partner explicitly trades away. JVs do not need identical objectives to work. They need honestly declared ones, priced into the design.
The Gap Between the Agreement and the Operating Reality
Most JV problems live in the gap between what the shareholder agreement says and what the organization actually does every day.
The agreement says decisions above a threshold go to the board. It does not say how a pricing exception gets approved on a Tuesday when the board meets quarterly. The agreement defines reserved matters. It does not define which parent’s procurement policy applies, whose ERP the venture runs on, whose HR grades are used, or which parent’s brand standards win when they conflict. The agreement allocates board seats. It does not stop both parents from sending informal instructions directly into the management team.
Each of these small gaps produces the same effect: the JV management team spends its energy managing upward, toward two shareholders, instead of outward, toward customers and performance. In the worst cases the venture becomes a diplomatic institution, staffed with capable people whose real job is arbitrating between parents.
The Symptoms of Governance Drift
JV underperformance rarely announces itself as a governance problem. It shows up as operational symptoms that look like management weakness.
Decisions take months. Not because the analysis is hard, but because every significant decision needs an informal pre-alignment with both parents before it can even be tabled. Managers learn that the safest move is to not decide.
The venture runs on two of everything. Two reporting formats, because each parent wants its own. Two sets of policies, applied inconsistently. Sometimes two seconded management layers, each loyal to its parent. Overhead grows while accountability shrinks.
Secondees face split loyalty. People seconded from the parents carry their home company’s interests, career incentives, and reporting lines into the venture. That is not a character flaw, it is the system working as designed. Unless the JV builds its own identity, its own scorecard, and its own incentives, the leadership team is a coalition rather than a team.
Performance visibility is negotiated. When numbers make one parent look bad, the numbers get debated instead of the performance. A JV without one trusted, independent version of its figures will spend board meetings arguing about measurement instead of deciding about direction.
Deadlock becomes culture. Formal deadlock provisions are rarely triggered. Instead, the venture develops soft deadlock: contested topics simply stop being raised. The agenda shrinks to what both parents can tolerate, which is rarely what the business needs.
Designing Governance That Actually Governs
The good news is that JV governance is an engineering problem, and it can be engineered well. A handful of design choices separate ventures that perform from ventures that drift.
First, write the decision rights down, at operating altitude. Not just reserved matters, but the practical layer underneath: pricing authority, discount limits, hiring approvals, capex thresholds, procurement authority, system choices. A simple decision matrix, agreed by both parents and published to the management team, removes more friction than any team-building exercise. The test is that a middle manager can answer “who decides this” in ten seconds.
Second, give the venture one operating system, not two half-inherited ones. Choose whose processes, whose ERP, whose policies apply, or build the venture’s own, but decide explicitly. Every undecided inheritance becomes a permanent negotiation. The cheapest moment to decide is at setup. The second cheapest is now.
Third, build one version of the numbers, owned by the venture. The JV’s finance function must produce figures that both parents accept as the reference, mapped once into each parent’s reporting requirements. The moment each shareholder rebuilds its own view of the venture’s performance, trust erodes and every review becomes a reconciliation exercise.
Fourth, protect the management team’s mandate. The CEO of the venture should have a clear scorecard agreed by both parents, the authority that goes with it, and a single channel for shareholder direction, the board, not parallel phone calls from two head offices. Parents that bypass the board get short-term comfort and long-term dysfunction.
Fifth, install a real operating cadence between the parents. Quarterly boards are not enough in the first years. A monthly shareholder operating review, focused on decisions and blockers rather than presentations, keeps small misalignments from compounding into strategic ones. The cadence is also where partner trust is actually built, decision by decision.
None of these mechanisms is JV theory. Every one of them, the decision matrix, the single set of numbers, the protected mandate, the operating cadence, is standard equipment in post-merger integration, proven where the stakes were a failed acquisition rather than a strained partnership. The difference in a JV is only that they must be built to last, because there is no post-integration end state where the tension resolves itself.
The Partner Interface Is a Process, Not a Relationship
There is a comfortable belief that JV success depends on partner chemistry. Chemistry helps, but it does not scale and it does not survive management rotation. What survives is the interface: the defined points where the venture and its parents exchange decisions, services, data, and people.
Treat that interface like any other core process. Define what services each parent provides to the venture and at what cost. Define what data flows to each parent, in what format, at what frequency. Define how secondments work, how long, on whose payroll, with what return path. Define how disputes escalate before they reach deadlock provisions. Ventures with engineered interfaces survive personality changes on both sides. Ventures built on relationships alone are one reorganization away from drift.
The First Year Decides the Next Ten
Joint ventures have a formative period, roughly the first twelve to eighteen months, during which the operating habits form. Whatever is tolerated in that window becomes the culture: informal parent instructions, negotiated numbers, deferred system choices, decisions parked out of politeness. Habits formed early are extraordinarily expensive to change later, because by then they have beneficiaries.
That is why the setup sequence deserves executive attention, not just legal attention. In the first hundred days, the venture needs its decision matrix published, its finance function producing one set of numbers, its management scorecard signed by both parents, and its shareholder cadence running. None of this is exciting, and all of it is much cheaper to install while goodwill is at its peak and precedents have not yet hardened.
The formative period is also when the venture must earn the right to be autonomous. Parents delegate to ventures that demonstrate control: clean reporting, no surprises, commitments met. A JV management team that wants freedom should over-invest in transparency early, because trust extended is a function of visibility provided. The ventures that fight for autonomy while under-reporting get neither.
And when the venture inherits problems from the setup, an ambiguous scope, an unrealistic business plan, a services agreement priced for politics rather than economics, the first year is the moment to renegotiate, explicitly and factually. Every year that passes converts a setup error into a performance excuse, and performance excuses are corrosive: they give both parents a reason to disengage and the management team a reason to underdeliver.
The Test That Matters
There is a simple test of JV health that requires no consultant and no diagnostic. Take the last three significant decisions the venture needed. How long did each one take from being raised to being resolved? Who actually made it? And did anyone outside the boardroom notice the difference afterward?
If decisions are slow, made informally by the parents, and invisible in operations, the venture has a governance problem, whatever its P&L says this quarter. If decisions are fast, made at the level the matrix says, and visible in performance, the JV is doing the one thing joint ventures exist to do: combining two sets of strengths into one operating engine.
Joint ventures do not fail because two companies disagreed. They fail because nobody built the machinery for two companies to agree, week after week, at operating speed. That machinery is designable. The ventures that build it early capture the partnership premium. The ones that do not spend their best years negotiating with themselves.
