Beyond Growth
What We Measure and What We Miss
What Evaluation Inherits
Programs are evaluated on whether they expanded, not on whether they can sustain what they built. Logic models show the path from activities to outcomes but not whether that path is sustainable at the pace being asked. Outputs are treated as equivalent regardless of what they cost to produce, the way GDP treats weapons manufacturing and healthcare as equal contributions to economic growth. Time frames are built around grant cycles that encode assumptions about how quickly change should happen, assumptions borrowed from growth economics rather than from what communities actually need.
The care work that makes programs function stays invisible until it is gone. The community organizer who shows up early and stays late. The peer navigator who carries knowledge about which families need what and when. The program coordinator who holds relationships together through transitions and crises. This work is treated as context, as the background against which the real program operates. When these people leave, the program discovers that what looked like context was actually infrastructure. But evaluation frameworks do not measure infrastructure. They measure outputs.
These assumptions come from somewhere. They are borrowed from an economic system that measures activity rather than quality, growth rather than stability, transactions rather than relationships. GDP counts everything that involves money changing hands and treats it all as progress. A car accident increases GDP through medical bills and repairs. Cutting down a forest increases GDP through timber sales. The system cannot distinguish between activity that depletes and activity that sustains because it only measures whether something happened, not whether what happened was good.
Evaluation has inherited this logic without examining whether it serves what we claim to care about. We measure reach without asking if reach is sustainable. We count outputs without asking what they cost to produce. We track growth without asking if growth is making things better or just making them bigger. Every time we design an evaluation that asks whether a program grew without asking whether it can sustain itself, that measures activities without measuring the care work that makes those activities possible, that treats expansion as success and stability as stagnation, we reproduce assumptions we might not hold if we examined them directly.
The frameworks we use determine what programs optimize for. If the framework measures reach, programs will optimize for reaching more people. If it measures speed of expansion, programs will optimize for growing quickly. If it only counts what can be quantified, programs will focus on what produces numbers and treat everything else as secondary. This means changing the framework is not a theoretical exercise. It changes what programs actually do.
So what changes when we change the question? Instead of asking how many people did you reach, we ask how deep are the relationships. Instead of asking how quickly did you expand, we ask can you sustain what you built. Instead of asking what changed, we ask who decided what change looks like. Instead of asking what is the return on investment, we ask who benefits and who bears the cost. Instead of asking did you hit your targets, we ask what did it cost to hit them.
These are not abstract theoretical questions. They are practical questions about what programs need to know to sustain themselves. Growth metrics tell you whether you are expanding. They do not tell you whether you can maintain what you built. Which raises the question: what does evaluation systematically leave out when it focuses only on growth?
What Gets Left Out
Even when evaluators try to measure relationships, through network mapping, linkage indicators, relationship quality scales, we are measuring structures and outcomes, not the labor that creates and sustains them. Network mapping shows who is connected to whom. It can visualize density, identify key connectors, reveal patterns of communication. What it cannot show is the work of maintaining those connections. The phone calls that happen at odd hours. The checking in that looks casual but is deliberate. The remembering of details that signal care. The labor is invisible in the map.
We can count frequency of contact, number of touchpoints, hours spent in meetings, but we cannot easily count the on-call nature of being the person people trust. The program coordinator is not only working during program hours. She is carrying the knowledge of the community, holding the trust that makes everything else possible. She knows when something is off before anyone says it. She is present even when she is not physically there. This presence is not measurable in discrete units.
The same gap appears when we track activities. We can document that a staff member conducted 15 home visits, facilitated 8 support groups, attended 12 community meetings. What we cannot capture is the multitasking that defines care work. The home visit where you are also noticing how the kids look, whether there is food in the house, if the energy has changed since last time. The support group where you are facilitating while also tracking who is quiet today, who seems more stressed, what the group needs that is not in the curriculum. The multitasking is not supplementary. It is how the work actually happens.
Knowledge transfer presents a similar problem. We can document it through training manuals and onboarding processes, but we cannot transfer the embodied knowledge of knowing who needs what when. The community organizer who has been there for five years carries information that no manual captures. She knows which families will not ask for help until it is desperate. She knows which young people need the program to feel like their idea, not something imposed. She knows when to follow protocol and when protocol will break what matters. When she leaves, that knowledge leaves. No amount of documentation fully replaces it.
Feminist economists have been pointing this out for decades. Care work is systematically devalued not because it is unmeasurable but because measuring it in the ways institutions demand, as discrete activities, as time units, as countable outputs, fundamentally misrepresents what the work is. Arlie Hochschild wrote about emotional labor in the 1980s. Nancy Folbre showed how time-use surveys miss the simultaneous nature of care work. Marilyn Waring demonstrated that GDP’s exclusion of unpaid work was not an oversight but a structural choice about what counts as productive.
The same structural choice shapes evaluation. When we build logic models, we list activities and outputs. We do not list the holding of trust, the carrying of knowledge, the being available when needed. These do not fit in the boxes. So they become context, background, the soft stuff that supports the real work. But they are the real work. The activities in the logic model depend entirely on this foundation. Without it, the program is just a series of transactions with no underlying relationship to hold them.
This is not a problem of finding better metrics. A program coordinator does not just manage relationships. She is the relationship. She carries the knowledge that makes the program work. She holds the trust that allows participants to be honest about what they need. She knows the community in ways that cannot be extracted and handed to her replacement in a transition document. This knowing is not an output. It is not an outcome. It is the infrastructure everything else rests on.
When evaluation treats this as context rather than as core work, we have already accepted the capitalist frame that what matters is what scales, what is replicable, what can be standardized. We have already decided that the labor of care is less important than the outputs it produces. We measure the outputs and assume the infrastructure will sustain itself. Then we are surprised when staff burn out, when programs lose their effectiveness after key people leave, when communities stop trusting programs that keep cycling through new staff who do not know them.
The care work problem is not that we have failed to find the right metrics. The problem is that some kinds of work resist quantification without being reduced to something they are not. Counting the hours of relationship maintenance still treats relationships as discrete activities rather than ongoing presence. Measuring emotional labor in units still misses that emotional labor is not a task you complete but a way of being present to the work.
This does not mean care work should stay invisible. It means we might need different ways of making it visible. Description instead of counting. Narrative instead of metrics. Naming the work for what it is instead of translating it into units that fit existing frameworks. Recognizing that visibility does not always require quantification, that some things are made visible by being described honestly rather than measured. Which raises the practical question: given these limits, what can evaluators actually do?
What We Could Change Tomorrow
Individual evaluators won’t overthrow capitalism through better indicators. Funders want proof of impact measured in reach and scale. Institutional resistance to alternative metrics is real and structural. These are not small obstacles. They are the conditions we work within.
But alternative frameworks already exist that challenge growth-based measurement. The Genuine Progress Indicator adjusts GDP by accounting for costs that economic growth produces: pollution, resource depletion, inequality, the value of unpaid work. Bhutan’s Gross National Happiness measures collective well-being across nine domains rather than just economic output. Degrowth-aligned metrics measure reduction as success when reduction means sustainability. Kate Raworth’s Doughnut Economics proposes staying within ecological and social boundaries rather than pursuing endless growth.
These are not fringe experiments. New Zealand launched a Wellbeing Budget. Amsterdam and Copenhagen use the Doughnut framework for urban planning. They are governments and municipalities acknowledging that GDP and growth metrics do not tell them what they need to know. The frameworks exist. What has been missing is not the conceptual tools but the institutional will to use them.
There is a deeper tension here, though, that Marilyn Waring identified decades ago when she critiqued GDP for excluding women’s unpaid work. She argued this work should be counted, that making it visible in economic metrics would force recognition of its value. But she also warned against attaching price tags to care and ecological services because doing so subjects them to market logic, as if their value depends on what they would cost if purchased rather than on their inherent necessity for human life.
The same tension exists in evaluation. We can develop better metrics that capture care work and relationship maintenance. We can count the hours spent building trust, measure the depth of community connections, track the emotional labor of staying present in difficult work. But the act of counting can reduce care to units, relationships to transactions, presence to billable hours. Some things might need to be described rather than counted, named rather than quantified, valued through narrative rather than metrics.
The question, then, is not just what should we measure instead. It is what are the limits of measurement itself, and when does quantification become another form of the extraction we are trying to critique. Working within these limits, here is what becomes possible.
Post-capitalist evaluation is not a distant goal that requires overthrowing the system before we can begin. It is a practice that starts with the next evaluation, the next set of questions, the next conversation with a funder or program staff. You cannot redesign the entire framework. But you can add questions to your data collection. When you ask participants about outcomes, you can also ask what it took to achieve them. When you interview staff about program activities, you can ask about the work that happens between the activities, the relationships that make everything else possible. When you conduct observations, you can notice not just what is being delivered but what is being sustained and what is being depleted. These questions do not require permission. They are within the scope of understanding how the program actually works.
You can name what the standard metrics miss. Every evaluation includes a limitations section. That section can acknowledge that the metrics chosen measure certain things and not others, that they capture reach but not sustainability, outputs but not the labor that produces them. You can note that relationship maintenance is happening but is difficult to quantify in ways that fit logic models. This is not methodological weakness. It is honesty about what the methods can and cannot show.
You can describe what matters that cannot be easily counted. Qualitative sections exist in most evaluation reports. Those sections can do more than illustrate what the quantitative data shows. They can capture the care work that makes the program possible. The ways community organizers hold knowledge and build trust. The emotional labor of staying present in difficult work. The on-call nature of being the person people rely on. These descriptions are evidence. They are data about how the program functions. They belong in findings, not just in background sections.
Connected to this, you can make care work visible by naming it as work. When the report discusses program implementation, it can identify relationship maintenance as a core activity, not as context. When it discusses staffing, it can name the carrying of trust and knowledge as infrastructure that the program depends on. When it discusses sustainability, it can note that care work is what holds the program together and that depleting it threatens viability. This does not require new metrics. It requires treating care work as central to how the program operates rather than as background that supports the real work.
You can write recommendations that acknowledge trade-offs. Not every recommendation needs to be about growth. Some can be about what the program needs to sustain what it has built. Recommendations can name that serving more people would require either more staff or acceptance that quality will decline. They can state that the program is operating at a sustainable size and that maintaining that sustainability is itself a success worth recognizing. This might not be what funders want to hear, but it might be what programs need to say.
You can have conversations with funders that complicate the narrative. When you present findings, you can note not just that targets were met but what it cost to meet them. You can point out when a program succeeded by staying focused rather than expanding reach. You can name when infrastructure, the people and relationships that hold everything together, is being maintained or depleted. These conversations do not require confrontation. They require honesty about what the data shows, including data about sustainability that might not have been requested but is relevant to whether the program can continue doing what it does.
None of these changes require permission from funders or approval from institutions. They are within the scope of what evaluators already do: asking questions, collecting data, analyzing findings, writing reports, having conversations about what the evidence shows. The difference is in what we choose to ask about, what we choose to name, what we choose to treat as evidence rather than as context.
The system shapes evaluation. Funders shape what gets measured. Institutions shape what counts as rigor. But evaluators still make choices within those constraints about what questions to add, what to name in reports, what to describe alongside what we count. Those choices accumulate. They create a record that programs are being asked to do things that are not sustainable. They make visible the work that holds everything together. They complicate the narrative that growth is always good and that scale is always the goal.
Post-capitalist evaluation is not a framework to adopt wholesale. It is a set of questions to ask, a commitment to naming what growth costs, a practice of making care work visible, an acknowledgment that sustainability matters alongside reach. It is what we can do tomorrow, in the next evaluation, with the tools and constraints we already have. The work is not to fix evaluation before we can do good work. The work is to do good work differently, to notice where our frameworks reproduce assumptions we might not hold, and to practice something else in the spaces where we have choice.
Let’s begin there.
Anthralytic explores evaluation, strategy, and the practice of seeing clearly. Subscribe for writing on Buddhist evaluation, post-capitalist metrics, and building systems that don't extract.
