Something changed in the way organizations from our territories come to us. The usual invitation used to be to take part in an activity, lend a space, refer a group. Lately the invitation is a different one: let’s build this together. A project designed by four, six, eight hands. A shared budget. Shared accountability. Everyone’s brand on the materials.
That is a good thing. It means the network has matured, and that there is enough trust to share what is hardest to share: money, responsibility and credit.
It is also considerably harder than it looks.
About two years ago we took part, as an ecosystem-building organization, in the training programme of the Coalizão pelo Impacto, the Brazilian national initiative that supports organizations providing direct assistance to impact entrepreneurs and businesses, with ICE as its executive secretariat. The module on Partnerships and Ecosystem Management, led by Carol Andrade, co-founder and senior advisor at Social Good Brasil, put names to a good many things we were already doing without naming them, and showed clearly where we still take a beating.
We are going back to that material now, precisely because the subject has returned to the table with the invitations that keep arriving. Putting that content together with what we have learned in practice over the years, I decided to organize it all in one place. Not as a manual, because we have no authority for that, but as an honest record from people who have already got these ten fronts wrong and right, and would rather the next organization got them less wrong.
Partnership, network, ecosystem: why the word matters
The training offered a simple distinction that saves a lot of frustration:
Not every partnership operates within a structured ecosystem. But every structured ecosystem operates through partnership.
A partnership is an agreement between two or more organizations to support one another: to complement a capability, a resource or a reach. An ecosystem is another scale altogether. As Silvio Meira defines it, an ecosystem is “the complex of institutions, organizations and agents in a given environment and their interrelations within a certain unit of space and time”. Operating within an ecosystem is far more complex than operating through partnership, and confusing the two is the origin of half the problems.
The confusion is expensive in practice. A project designed as a bilateral partnership but executed within an arrangement of six territorial organizations will discover halfway through that it has no governance for what is actually happening. By then it is too late: the money has arrived, the schedule has started, and the conversation about who decides what becomes a negotiation under pressure.
Shared governance, as we understand it here, is the set of agreements that answers very concrete questions before they show up as emergencies:
- Who decides what, and how far does each organization’s autonomy reach?
- Who signs their name when someone has to sign their name?
- What happens when the budget has to change?
- Who looks after what was built once the funding ends?
Put that way it sounds like bureaucracy. In practice it is the opposite: it is what allows a project to survive its first surprise without anyone having to choose between the relationship and the result.
There are ten fronts. Here is each one.
1. Governance begins on ten pages already written, not on a blank one
The first mistake is the most seductive: sitting down with partner organizations and designing governance from scratch, as though the project were everyone’s first commitment in life. It is not.
Every organization entering a shared arrangement arrives with bylaws, internal rules, a procurement policy, approval thresholds, deliberative bodies, legal constraints and its own way of deciding. Ignoring that in the name of the project’s agility is not efficiency. It is agreeing to something the partner has no way of delivering.
So the work begins with a tedious and necessary mapping exercise: how does each institution decide? What are its deliberative bodies? How far does the autonomy of whoever sits at the project table extend? What commitments is it barred from making by its own bylaws?
Where policies converge, it is easy: formalize the agreement and move on. Where they diverge, and they diverge more often than people imagine, the divergence has to be opened up explicitly and reconciled right there, at the design stage.
A divergence that does not become an agreement at the design stage becomes a conflict during execution, and then it costs ten times as much. At the design stage the conversation is about design. During execution it is about frozen money, late delivery and a worn-out relationship.
Respecting each organization’s internal policy does not mean being held hostage to anyone’s particularities. It means putting the differences on the table while there is still room to design around them.
2. A role without a name is nobody’s role
I have written this about integrity and it holds just as well here: a function belonging to everyone tends to become a function belonging to no one.
Each organization’s role has to be defined from the design stage and recorded in a document someone can consult two years later: minutes, a memorandum of understanding, a written agreement. An arrangement that is not written down is a memory, and different people never remember the same thing.
Three definitions cannot be missing:
Who manages the funds. One organization will handle the finances, meet tax obligations, report to the funder and to the other partners, and keep the accounting records. The choice weighs technical capacity, administrative structure, prior experience and reputation, not proximity or order of arrival.
How transparency works in practice. It is not enough for the managing organization to be competent. Periodic financial reports with dates, financial documentation open for the others to check, accountability meetings with an agenda, information in a format everyone can read. Transparency that depends on asking is not transparency. It is customer service.
Who bears financial and legal responsibility. This is the most delicate one, and the one most people put off. If a funder raises a question, or an administrative proceeding or a lawsuit appears, someone will have to answer. That organization needs to know exactly what it is taking on. And some organizations simply cannot take it on, because of their own bylaws, which is information rather than a problem, as long as it comes up early.
Alongside that: a professional dedicated to risk, contracts and legal safeguards, budgeted from the start. It can be the managing organization’s in-house counsel, external advisors hired for the project, or both. What cannot happen is for the cost of legal advice, insurance and contingency to appear as a surprise mid-execution. Governance that did not budget for that cost is already exposed to a risk it never needed to run.
3. The budget also has to budget for changing the budget
A project with several organizations is financially complex by nature. Facing that head-on at the design stage is cheaper than discovering it later.
It starts with time. Every organization needs real time to analyse the project’s impact on its own operation and build a budget that reflects the true cost, indirect cost included. Administrative expenses, staff costs, infrastructure, utilities, accounting. It is precisely the indirect cost that tends to be underestimated, and it is the one that erodes an organization from the inside while the project appears to be going well.
Then, specification. Every activity with an identified responsible organization, an allocated budget, a schedule and a result indicator. An activity with no owner is the main source of conflict in a shared project, because when it falls behind, everyone has an explanation and no one has a responsibility.
And then comes the triple constraint that project management has been teaching for decades: time, scope and resources do not move for free. There is no widening the scope without touching the deadline or the money. There is no shortening the deadline without touching the scope or the money. Anyone promising all three at once is merely postponing a conversation, and the postponed conversation always comes back worse.
That is why the financial agreement has to say, before it needs to:
- How is a budget change approved? A board with a representative from every organization, or does each one reallocate its own budget lines autonomously? Who ratifies it? Within what timeframe?
- Whose are the financial returns? If the project has a fixed value and the invested balance yields returns, do those returns stay with each organization in proportion to its share, are they split equally, or are they reinvested in the project? And what legal instrument supports that choice? The answer generally runs through a contract amendment justifying, activity by activity, the allocation of those investment proceeds, and that is a conversation involving a tax specialist, not just a manager.
- What happens if funds fall short? Cut the scope, extend the deadline, or raise more? That decision cannot be improvised on the day the problem shows up. It needs criteria agreed in advance.
A budget that did not budget for changing the budget was born expired.
4. A shared project is not a project without an owner
There is a mistaken reading of “shared management” that causes real trouble: the idea that dividing governance means diluting leadership. It is the opposite. The more organizations at the table, the better the coordinator has to be.
Leadership of a shared arrangement needs three things at once: real experience managing complex projects, the ability to mediate conflict without taking sides, and enough stubbornness to keep the objective in focus.
But the skill that decides the game is communication. Anyone who works in project management knows that most of the time is conversation, and in a shared project that proportion is even higher, because leadership talks to the team, to the leadership of every partner organization, to the funder, to the actors in the territory and to the community. Five audiences, five languages, one single coherence.
This has to be structured, not improvised: a defined formal channel, regular meetings with an agenda and a record, progress reports, minutes of decisions and, most importantly, transparency about what is going wrong as well. Bad news that arrives late becomes a crisis. Bad news that arrives early is still a management problem.
5. A value that does not become a decision criterion is decoration
Here we are quite uncompromising. The agreement on values between organizations has to go beyond the cliché. “Commitment to social transformation” is not an agreement. It is a sentence.
An agreement looks like this:
- The project will affect people. Will there be affirmative-action places? For which groups, at what percentage, defined how, advertised where, filled by what criterion?
- The project commits to accessibility. What does that mean as a budget line? Materials in accessible formats, venues with guaranteed access, plain language, Brazilian Sign Language (Libras) interpretation, audio description. Each of those has a cost and a responsible party.
- How is it measured? If we say the team will be diverse, we need to be able to show the team’s composition. If we say the results benefit different groups equitably, we need disaggregated data.
An agreed value has to appear in three places: in the decision criterion, in the evaluation indicator and in the budget line. If it does not appear in all three, it is decoration.
And there is the question almost nobody asks at the design stage: what if new people join halfway through? A new partner arrives with what level of participation in decisions? Do they enter the existing governance or an advisory arrangement? Settling that in advance avoids two bad outcomes: unbalancing governance that was already working, or breeding resentment among those who were there from the start.
6. What cannot be violated has to be written down before it is tested
Clear decision-making, at both levels: in the project and in each institution.
The distinction that helps most is between strategic and operational decisions.
Strategic is collective: the project’s objective, allocation of resources across fronts, admission of a new partner, significant budget changes, revisions to the theory of change.
Operational is delegated: execution of planned activities, day-to-day management, hiring suppliers within already approved thresholds.
Without that distinction, one of two extremes occurs: either everything becomes an assembly and the project stalls, or nothing becomes an assembly and someone decides alone what they should not have.
And alongside it: which transparency documents have to be kept current, at what frequency and available to whom. Financial reports, meeting minutes, decision records, progress reports. It is not good practice. It is what sustains trust. Information shared as a matter of routine produces a secure partner; information shared on demand produces a suspicious one. And a suspicious partner turns a small divergence into a large conflict.
7. A project of many that communicates as a project of one loses the many
Integrated communication is where good intentions stumble most, because it is where credit becomes visible.
It has to be settled from the start: will the project have its own visual identity? How does it relate to the organizations’ brands? How is each one’s contribution credited, in print, in a post, in a report, at an event? Who is responsible for the project’s communication, and with what authority?
And communication has to serve audiences that want different things: the community wants to know what changes in its life; the funder wants to know what was delivered and how; the institutional partner wants to see its role acknowledged; the press wants what is new. Different strategies, a message coherent across them, and the project’s collective nature visible in all of them.
When communication erases collective authorship, it is not only being unfair. It is dismantling the trust that holds the arrangement together.
8. Funding ends, domains expire, sites go down. The learning should not go down with them
This is the lesson that hurts most, because the loss is silent and final.
The project ends. The special site is switched off. The domain expires. The methodology that took two years to build becomes a PDF lost in the drive of someone who has already left the organization. The database disappears. And the next initiative, three years later, starts from zero, repeating a mistake that had already been solved.
It can be avoided with a simple definition made at the design stage: which organization is responsible for keeping the archive and keeping it online after the project ends: with what resources, for how long, under what licence.
That includes maintaining the site or page, preserving the materials, publishing methodologies and tools in accessible and reusable formats, and documenting the learning, including the parts that did not work.
For Casa Hacker this is non-negotiable. In every project we join, we fight to keep the knowledge generated available. That is why we released our accessibility toolbar under a free licence, and it is why our policies and our governance are published in full at casahacker.org/movimento. Knowledge that dies with the project is a public resource that turned into an expense.
9. A conflict of interest is not the problem. An undeclared conflict of interest is
Shared governance rests on good management practice, compliance and integrity, and that has to be agreed rather than assumed.
The central commitment is a single one: every decision, financial, strategic or operational, is guided by the impact the project set out to generate in the territory and in people’s lives. Conflicts will arise between one organization’s particular interest and the project’s collective interest. That is normal and it is not a sign of bad faith. What has to be agreed from the outset is which one prevails.
And integrity here is quite practical. If a partner organization has a commercial relationship with a supplier the project might hire, that gets declared. If someone on the team has a personal interest in a decision, that gets declared. If there is a family tie between whoever decides and whoever is hired, that gets declared.
Declaring prevents nothing. Declaring moves the decision out of the realm of suspicion and into the realm of the record, and that difference is exactly what protects the honest person.
10. Every extra chair in governance is another meeting in everyone’s calendar
The last lesson is the one that took us longest to understand, and perhaps the most useful: governance has to be proportional to the resources available and to the maturity of the people at the table.
There is a generous impulse to invite everyone into governance. It almost always costs dearly. Each additional participant multiplies communication channels, increases the number of meetings, lengthens decision time and raises coordination costs. In a project with tight resources, that cost eats precisely the part that should have become impact.
Being reasonable about numbers is not being exclusionary. It is acknowledging real conditions. Lean governance with prepared, present people usually delivers more than broad governance with people who have no time to devote to it. If the intention is to widen participation, other formats exist: thematic committees, advisory boards, regular listening sessions with the community. Not every voice needs to sit in the deliberative body to be heard. What it needs is a real place.
Maturity matters too. A partner with prior experience of shared arrangements handles a more flexible structure and greater autonomy well. A partner doing this for the first time needs a more detailed structure and closer support. Applying the same level of detail to both is getting it wrong twice.
And the other side of the coin: beware of definitions that are too rigid at the start of a long project.
The most common example is contracting a supplier or provider under closed exclusivity from the design stage. It looks like organization; it becomes a shackle. If in two years the strategy evolves, the theory of change is revised, or a new partner arrives with a capability nobody had, the obligation to keep the original arrangement limits adaptation, and the price of that is impact that never happened.
A supplier locked in by exclusivity on a three-year project is a bet that nothing will change in three years. I have never seen that bet pay off.
The balance is this: definitions clear enough to guide action, open enough to allow adjustment when reality changes. Write the criterion, not the supplier’s name. Write the decision process, not the decision.
What theory already knew before we did
It is worth saying honestly: we invented nothing. We crossed, taking our knocks and in our own territory, a path that very good people had already mapped.
Elinor Ostrom won the Nobel Prize in Economics in 2009 by showing that communities can manage shared resources without privatizing or nationalizing them, provided they build clear boundaries about who is inside, rules that fit local conditions, participation by those affected, monitoring, graduated sanctions, low-cost conflict resolution and governance nested across several levels. Published in 1990, and it is item by item what we were working out on our own. The resources of a territorial project are exactly of that kind: knowledge, trust, social capital, shared infrastructure.
John Kania and Mark Kramer named the problem in 2011: isolated impact, each organization seeking its own solution to a question that is interconnected. The alternative requires a common agenda, shared measurement, mutually reinforcing activities, continuous communication and a backbone structure dedicated to coordination. That last one is what most people try to economize on, and it is the one that yields least.
The rest of the framework explains why. Granovetter showed that it is weak ties, the connection between different groups, that make information and opportunity circulate; Putnam linked social capital to the performance of local institutions; and Bourdieu reminded us that access to privileged networks is one of the mechanisms by which elites reproduce advantage, which makes building networks in the periphery a confrontation with inequality rather than networking. Donella Meadows placed rules and information flows among the most powerful leverage points in a system, which is exactly where governance lives. And Sen, Sachs, Favareto, Singer, Fraser and Boaventura de Sousa Santos sustain the part that is not technical: development is the expansion of freedoms, it depends on institutional arrangements able to pierce sectoral fragmentation, and justice is not only redistribution: it is also recognition and representation.
In the Coalizão training, what tied all of this together came from Jeremy Heimans and Henry Timms: old power is the possession and control of information and resources; new power is possession and control shared among networked people. They are not mutually exclusive: they are spectrums. Networked governance, transparency, a movement mindset, hacker culture: the list describes, point by point, what Casa Hacker has been trying to be since 2018.
All of it converges on a sentence we could have saved ten years by learning: a complex social problem is not solved alone, nor from outside, nor from above.
What comes next for us
Three things are in our queue for the coming months.
Turning this into an instrument, not a text. We are documenting these lessons and turning part of them into working instruments in-house: a checklist for mapping internal policies, a matrix of roles and approval thresholds, and the list of financial questions that need answers before any signature. It is work in progress, and what we learn using it comes back here.
Bringing these checks into the system. Cockpit, our own management and integrity technology, already runs daily compliance checks over the execution of our programmes. The next front is extending that to shared arrangements: an activity with no responsible organization, an expired transparency document, a budget change with no record of approval. A rule that lives in the process works; a rule that depends on someone remembering does not.
Publishing the archive, always. We are going to formalize, as a standard clause in our partnerships, the definition of who keeps the archive and for how long, with an open licence as our declared preference. It is easier to agree on that in the first meeting than in the last.
Why this has everything to do with our mission
What I would say to anyone starting a shared arrangement now is that governance is not the boring part that slows the project down. It is the part that gets the project to the end in one piece, with the relationship preserved, the money accounted for and the learning kept.
And that almost everything that goes wrong later was already going wrong at the design stage, only silently.
In our manifesto we commit to treating technology as a right and as a language of power, and not as the privilege of a few. Shared governance is that same commitment applied to the way we work: decision-making is a language of power too, and when it stays concentrated in a single organization, the territory remains the object of a project instead of its subject.
The organizations of our territories are not coming to us to execute an idea that arrived finished. They are coming to build together. That changes what we need to know how to do, and this text is our attempt to do it properly.
When Casa Hacker is born, born with it is the certainty that peripheries are not a place of waiting — they are a place of beginning. And a place of beginning is governed together.
References
ACATE. Ponte para a inovação. Florianópolis: ACATE, 2021.
ANDRADE, Carol. Módulo 4 — Parcerias e gestão de ecossistemas: aula 1. Formação de organizações dinamizadoras da Coalizão pelo Impacto. Florianópolis: Social Good Brasil, [2023]. 35 slides.
BARABÁSI, Albert-László. Linked: the new science of networks. Cambridge: Perseus, 2002.
BESSANT, John; TIDD, Joe. Inovação e empreendedorismo. Porto Alegre: Bookman, 2009.
BOURDIEU, Pierre. The forms of capital. In: RICHARDSON, John G. (org.). Handbook of theory and research for the sociology of education. New York: Greenwood, 1986. p. 241-258.
CASTELLS, Manuel. A sociedade em rede. A era da informação: economia, sociedade e cultura, v. 1. São Paulo: Paz e Terra, 1999.
COLEMAN, James S. Foundations of social theory. Cambridge: Belknap Press of Harvard University Press, 1990.
CORAGGIO, José Luis. Economía social y solidaria: el trabajo antes que el capital. Quito: Abya-Yala, 2011.
FAVARETO, Arilson. Paradigmas do desenvolvimento rural em questão. São Paulo: Iglu; Fapesp, 2007.
FORRESTER, Jay W. Industrial dynamics. Cambridge: MIT Press, 1961.
FRASER, Nancy. Justice interruptus: critical reflections on the “postsocialist” condition. New York: Routledge, 1997.
FRASER, Nancy. Scales of justice: reimagining political space in a globalizing world. New York: Columbia University Press, 2008.
FUNNELL, Sue C.; ROGERS, Patricia J. Purposeful program theory: effective use of theories of change and logic models. San Francisco: Jossey-Bass, 2011.
GRANOVETTER, Mark S. The strength of weak ties. American Journal of Sociology, Chicago, v. 78, n. 6, p. 1360-1380, 1973.
HARDIN, Garrett. The tragedy of the commons. Science, Washington, v. 162, n. 3859, p. 1243-1248, 1968.
HEIMANS, Jeremy; TIMMS, Henry. O novo poder: como disseminar ideias, engajar pessoas e estar sempre um passo à frente em um mundo hiperconectado. Rio de Janeiro: Intrínseca, 2018.
KANIA, John; KRAMER, Mark. Collective impact. Stanford Social Innovation Review, Stanford, v. 9, n. 1, p. 36-41, 2011.
MANCE, Euclides André. A revolução das redes: a colaboração solidária como uma alternativa pós-capitalista à globalização atual. Petrópolis: Vozes, 2000.
MEADOWS, Donella H. Leverage points: places to intervene in a system. Hartland: The Sustainability Institute, 1999.
MEADOWS, Donella H. Thinking in systems: a primer. White River Junction: Chelsea Green, 2008.
MEIRA, Silvio. O que é estratégia? [S.l.]: Paradoxum, 2021.
MULGAN, Geoff et al. Social innovation: what it is, why it matters and how it can be accelerated. London: The Young Foundation, 2007.
MURRAY, Robin; CAULIER-GRICE, Julie; MULGAN, Geoff. The open book of social innovation. London: Nesta; The Young Foundation, 2010.
NESTA. Collective intelligence design playbook (beta). London: Nesta, 2019.
OSTROM, Elinor. Governing the commons: the evolution of institutions for collective action. Cambridge: Cambridge University Press, 1990.
PATTON, Michael Quinn. Utilization-focused evaluation. 4. ed. Thousand Oaks: Sage, 2008.
PATTON, Michael Quinn. Developmental evaluation: applying complexity concepts to enhance innovation and use. New York: Guilford Press, 2011.
PUTNAM, Robert D. Comunidade e democracia: a experiência da Itália moderna. Rio de Janeiro: Fundação Getulio Vargas, 1996.
PUTNAM, Robert D. Bowling alone: the collapse and revival of American community. New York: Simon & Schuster, 2000.
RASQUILHA, Luís. 30 tendências de negócios para 2030. MIT Sloan Management Review Brasil, São Paulo, 2021.
RAWLS, John. Uma teoria da justiça. São Paulo: Martins Fontes, 1997.
SACHS, Ignacy. Desenvolvimento: includente, sustentável, sustentado. Rio de Janeiro: Garamond, 2008.
SANTOS, Boaventura de Sousa. Para uma sociologia das ausências e uma sociologia das emergências. Revista Crítica de Ciências Sociais, Coimbra, n. 63, p. 237-280, 2002.
SANTOS, Boaventura de Sousa. A gramática do tempo: para uma nova cultura política. São Paulo: Cortez, 2006.
SEN, Amartya. Desenvolvimento como liberdade. São Paulo: Companhia das Letras, 2000.
SENGE, Peter M. A quinta disciplina: arte e prática da organização que aprende. São Paulo: Best Seller, 1990.
SINGER, Paul. Introdução à economia solidária. São Paulo: Fundação Perseu Abramo, 2002.
SOCIAL GOOD BRASIL. A era dos dados para o setor público. Florianópolis: Social Good Brasil, 2021.
WATTS, Duncan J. Six degrees: the science of a connected age. New York: W. W. Norton, 2003.
WATTS, Duncan J.; STROGATZ, Steven H. Collective dynamics of “small-world” networks. Nature, London, v. 393, p. 440-442, 1998.
WEISS, Carol H. Evaluation: methods for studying programs and policies. 2. ed. Upper Saddle River: Prentice Hall, 1998.
WESTLEY, Frances; ZIMMERMAN, Brenda; PATTON, Michael Quinn. Getting to maybe: how the world is changed. Toronto: Random House Canada, 2006.