The price of a fix often has little to do with the size of the problem. Much more likely, it depends on the quality of the diagnosis.
This runs against corporate instinct. A serious problem should require serious response, and a cheap answer to an expensive-looking problem feels irresponsible.
So, organizations reach for the proportionate fix. Complaints about speed? Let’s increase the speed. Complaints about service? Let’s buy more service. The budget now matches the apparent weight of the problem, and everyone leaves the room feeling the matter has been taken seriously.
Two old stories stand against this practice. In both, a real problem produced a stream of real complaints. In both, the obvious fix was the expensive one. And in both, the eventual answer cost close to nothing — not because someone got lucky, but because somebody finally figured out what was wrong.
The airport that made the walk longer
Some years ago, executives at a Houston airport had a problem. Passengers complained about long waits at baggage claim.
The response was the obvious one: put more baggage handlers on the shift, and it worked exactly as intended: the average wait time fell to eight minutes, comfortably within industry norms. But the complaints kept coming.
Finally, someone looked at the problem carefully. It took a passenger about one minute to walk from the arrival gate to baggage claim, and another seven minutes to wait at the carousel.
The fix was to move the arrival gates away from the main terminal and route bags to the furthest carousel. The walking time increased to roughly six minutes. (The bags did not arrive one second sooner.) Complaints dropped to near zero.
Notice what the money bought and what it didn’t. Extra handlers were a real improvement that changed nothing people cared about. Rerouting was a purely scheduling decision.
The building that hung mirrors
Tenants in a large office building complained that the elevator service was slow. A specialist consulting firm that was brought in established that the average wait really was long — the complaint was accurate, not imagined. The firm then worked through the options: add elevators, replace the existing ones with faster models, install computer controls to use them better. Each was examined. None was workable. The engineers concluded the problem could not be solved.
A young psychologist in the building’s personnel department heard about the problem and saw something different. The engineers heard a complaint about time. He heard a complaint about standing in a crowd with nothing to do. His suggestion was to place mirrors in the elevator lobbies, so that people waiting could look at themselves — and at each other.
As mirrors went up, the complaints stopped. Someone even congratulated management on the improved service.
What the two have in common
What unites these cases is that both named a cause rather than experience.
“The baggage claim is slow.” “The elevators are too slow.” Both complaints sound like descriptions of a fact. Neither is. Each is an ordinary person’s best guess at why they feel bad. The feeling is real. The explanation attached to it is a guess, and it arrives sounding like evidence.
And in both cases, the organization bought the guess wholesale. Houston hired more handlers. The building considered new elevators. Nobody was careless; both responses were competent, reasonable, and aimed squarely at the stated cause.
What happened next? Houston hit eight minutes, and the complaints kept coming. The elevator engineers exhausted every mechanical option and declared the thing unsolvable.
That moment is worth reflecting upon, because it is the most common ending of all. Not an expensive failure, but a confident conclusion that nothing can be done. Most misdiagnosed problems don’t just blow up. They get quietly reclassified as facts of life, and the organizations learn to live with them.
What broke the deadlock in both stories was the same move: somebody asked what the person complaining was actually going through. Not whether the complaint was justified or not — it was — but what the experience consisted of. Seven minutes of standing. A crowded lobby with nothing to look at.
Once you have that, the fix isn’t clever. It’s obvious, and usually cheap, because you are no longer buying a solution to a problem nobody has.
This is the part that should be encouraging and usually isn’t. Diagnosis is the cheapest thing an organization can do and the thing it does least.
So, the next question is an obvious one. If correct diagnosis is so much cheaper, why is it so rare?
A while back, I argued that there are two ways to put AI to work. You can treat it as an oracle — an answer machine you query and obey — or as an interlocutor, a thinking partner that sharpens your judgment instead of replacing it. I still think the distinction stands. But a recent essay by Ethan Mollick, Choosing to Stay Human, forced me to revisit it — because Mollick pointed to a troubling cost of the oracle mode that I had not fully appreciated back then, and because the evidence of this cost that has piled up since is hard to ignore.
So let me start where I left off.
The wrong question, answered beautifully
The most common way organizations use AI is as an oracle. You ask, it answers, and the answer arrives fast, fluent, and confident. This is genuinely useful — until you remember that AI has no way of knowing whether you asked the right question. If your problem is poorly defined, AI will not tell you so. It will produce a polished, sophisticated, and occasionally brilliant response to the wrong question, and it will do it at a speed and price that makes the output feel authoritative. Garbage in, gospel out.
That is the first danger, and it’s the one I’ve written about before: AI doesn’t move you toward a good decision; it moves you faster in whatever direction you’re already pointed. Point it wrong, and it becomes an efficient engine for going wrong. For organizations that already struggle to define their problems clearly — which applies to most small businesses and nonprofits, stretched thin and rarely afforded the luxury of a strategy team — this isn’t a small risk. It is the risk.
The subtler cost: cognitive surrender
Here is what Mollick’s essay made me see more clearly. Even when you ask the right question, the oracle mode exacts a second, quieter tax. Mollick’s colleagues at Wharton have a name for it: “cognitive surrender” — the documented tendency of people to stop thinking about a problem and simply defer to the machine, even when the machine is wrong.
The evidence is accumulating from every direction. A study from researchers at Carnegie, Oxford, MIT, and UCLA found that just ten minutes of AI-assisted problem solving measurably reduced people’s ability to work through problems on their own afterward. Once the AI was gone, they skipped more and solved less — across both math and reading. The crucial detail: the damage was concentrated among those who used AI to get direct answers. People who asked only for hints or clarification showed no meaningful drop-off. That is the oracle and the interlocutor, separated inside a single experiment.
The same issue appears elsewhere. A study in Nature found that collaborating with generative AI boosted immediate performance, but that the gain vanished once people returned to working alone, and that the hand-off left them less motivated and more bored. In software development, where AI tools became standard fast, novices posted real productivity gains while quietly compromising their own grasp of the fundamentals: conceptual understanding, code reading, and debugging. AI-enhanced output, it turns out, is not a shortcut to competence.
Mollick offers the cleanest illustration of all, drawn from two studies by overlapping research teams. In one, about a thousand high-school students in Turkey used plain ChatGPT to study math; they did their homework better, felt they were learning more, and then underperformed their AI-free classmates on the test. In the other, an AI tutor that delivered a personalized sequence of problems rather than answers, across schools in Taipei, produced gains equivalent to months of extra schooling. Same technology, opposite results. The entire difference was whether the AI did the thinking or provoked it.
What the chess players reveal
The study that ties all of this together comes from Wharton and INSEAD, and I’m a little embarrassed it took Mollick’s essay to send me back to it. Researchers ran a large experiment with around three hundred chess players, giving them two different kinds of AI help. The first was an action signal: here is the best move. The second was an attention signal: this is a critical moment — slow down and think.
The action signals did work: players made better moves. But the help came with a hidden bill: in the moves that followed, performance dropped. The players had gone passive; they had stopped engaging, and when the AI fell silent, they struggled. The attention signals produced the opposite effect: immediate gains were more modest, but players thought harder, stayed in the game, and played better afterward. The AI had functioned as a complement to their judgment rather than a substitute for it.
Read that again with our two words in mind. The action signal is the oracle. The attention signal is the interlocutor. The experiment doesn’t merely support the distinction — it measures the cost of choosing wrong. And it points to something more useful than “AI good” or “AI bad”: the right kind of support depends on who is deciding, what they’re deciding, and how fast they need to decide.
When a single decision is high-stakes and speed is everything — aviation alerts, real-time fraud — give me the answer. When decisions unfold over time and expertise compounds, protect the thinking.
Which brings me back to small organizations
For a nonprofit or a small business, the answer to “who is deciding, what, and how fast” is almost always the second case. These are not high-frequency, single-shot decision environments. They’re places where judgment compounds — where the same handful of people will face the same class of problem again next quarter, and where there is no second analyst to catch the error. Which is exactly why the oracle mode is the wrong default for them, and why the surrender temptation bites hardest precisely where it does the most harm. When you’re stretched thin, the machine that simply hands you an answer is almost irresistible.
But this is also where the real opportunity lives — and it’s bigger than avoiding harm. Used as an interlocutor, inside a disciplined process rather than as a replacement for one, AI becomes the thinking partner these organizations have never been able to afford. The loop is simple to state: define the problem, surface and stress-test your assumptions, reframe, and only then generate and weigh options.
AI earns its place at each step not by answering but by pushing back. The equalizing power of AI for small organizations isn’t that it hands them the same answers as the big players. It’s that it gives them the ability to ask better questions — and that ability, embedded in a process, is a genuine change in what a small team can do.
And the interlocutor can do something the oracle can’t
It would be easy to read all of this as a counsel of caution: use AI gingerly, keep your hands on the wheel. But the most exciting part is what becomes possible when a sharply defined problem meets a model you’re genuinely engaging with.
Consider the combinatorial geometry problem that the mathematician Paul Erdős posed in 1946. For eighty years, it resisted the tools of combinatorial geometry. Last month, OpenAI reported that a general-purpose reasoning model — not one trained for the task — found a solution by reaching into an entirely different field, algebraic number theory. It questioned an assumption the human researchers had quietly shared and produced an answer no one expected.
Notice two things. First, that result was available only because the problem was defined with absolute precision, which is the whole argument of this piece, running in reverse. Second, questioning a shared assumption and proposing an unexpected framing is not what an oracle does. It is the defining move of an interlocutor. The model didn’t just retrieve a solution; it challenged the way the problem had been seen. That is the capability worth reaching for — and you can only reach it if you’ve done the work of defining the problem and stayed engaged enough to recognize a good reframe when it arrives.
Choosing to stay human, as an organization
Mollick’s essay is titled Choosing to Stay Human, and his point is that the choice is ours: intentional use, rather than reflexive dependence or reflexive avoidance. He notes, rightly, that the defaults are being set right now — by the companies designing AI to be frictionless, by employers deciding what “using AI well” means, by everyone teaching the slippery thing we call AI literacy. Once a generation builds its habits, those defaults will be hard to reverse.
For a small organization, the stakes of that choice are concentrated and immediate. What you choose to hand to the oracle, and what you choose to keep as your own thinking, will shape the institution you become. The discipline I’m describing — staying in the problem, even with a capable machine at hand — is not nostalgia. It’s the difference between an organization that gets answers and one that gets better at asking.
Protect that, and AI stops being a way to move faster in the wrong direction. It becomes the partner you could never afford.
Earlier this spring, I wrote about the structuralpressures bearing down on the nonprofit sector. The numbers are not comfortable. According to the State of Nonprofits 2026 report, a national survey of 380 nonprofit leaders published by the Center for Effective Philanthropy, nearly three-quarters of nonprofit CEOs report that demand for their services has increased since January 2025. Almost 60% say it has become harder to secure foundation grants. More than a third have seen reduced government funding. About 30% have cut staff.
These are not outlier findings. They describe a sector caught in a widening gap between what communities need and what organizations can provide. And the tools available to close that gap are becoming less reliable.
One of those tools is the foundation grant. And it is precisely here that I want to slow down, because I think the difficulty many nonprofits are experiencing with grant funding is not quite what it appears to be.
A Fictional Case — and a Real One
What follows is a diagnostic exercise built around a fictional organization: Harbor Bridge Family Services, a human-service nonprofit with an 18-person staff, an annual budget of approximately $2.8 million, and a 25-year record of serving low-income families, immigrants, and older adults in its community.
Harbor Bridge is not real. But the situation it represents is.
I constructed the case to reflect conditions that appear with increasing regularity in the sector: rising service demand, a development function under strain, a grant success rate in visible decline, and leadership already working beyond capacity. The State of Nonprofits 2026 data suggest that the narrative I created is not a composite of edge cases. It is damn close to reality.
My purpose in building the case was not to demonstrate that I can describe nonprofit problems accurately. It was to test whether a structured diagnosis process — the kind of rigorous, assumption-interrogating problem definition I have been writing about recently — could produce genuine insight when applied to a sector-specific situation. Not whether the process could just name the problem, but whether it could reframe it.
It could. And the reframe matters.
The Pattern Harbor Bridge Tried to Escape
Harbor Bridge’s data tells us that it has a problem with the grant application process:
Year
Applications
Awards
Total Requested
Total Awarded
2022
14
8
$920,000
$610,000
2023
16
7
$1,100,000
$575,000
2024
19
5
$1,350,000
$420,000
H1 2025
11
2
$760,000
$145,000
More applications each year. Fewer awards. Smaller amounts. Harbor Bridge was, as its staff described it, running harder and falling further behind.
The organization had not been passive in response. It had tried five distinct interventions: submitted more applications, brought in an outside grant writer, added more client stories to proposals, built an internal impact dashboard, and launched a monthly donor campaign. Each response was reasonable. None of it helped much.
That pattern — multiple intelligent attempts, no sustained improvement — is one of the clearest diagnostic signals suggesting that the stated problem may not be the real one. The organization has been solving something accurately described but not accurately defined.
The Trap Inside the Data
The structured diagnosis, conducted using RCFinder — an AI-assisted tool that establishes a neutral problem statement before any analysis begins, then generates and deepens a structured hypothesis set — identified two primary root causes. They were not independent. They formed a loop.
Funding pressure and rising service demand were pushing Harbor Bridge to submit more applications. More applications consumed the time and relational bandwidth needed to qualify funders carefully, cultivate relationships, and position the organization as a genuine mission partner rather than a grant applicant. That reduced capacity produced lower-fit applications. Lower-fit applications produced lower win rates and smaller, more restricted awards. Smaller awards intensified the funding pressure, which pushed the organization to submit still more applications.
Every response Harbor Bridge had tried was operating entirely within this loop. Bringing in a grant writer improved proposal quality but did not address funder fit or relationship depth. Submitting more applications increased volume — and made the capacity problem worse. The loop was self-reinforcing, and every attempt to escape it through the obvious responses was, in effect, tightening it.
The core finding: Harbor Bridge did not have a grant-writing problem. It had a funding strategy problem — one that had been progressively hidden behind a grant-writing problem for three years.
Why This Matters Beyond Harbor Bridge
I want to be precise about what this diagnostic finding does and does not claim.
It does not claim that the external environment is irrelevant. The State of Nonprofits 2026 data is real, and the tightening of foundation grant markets is a genuine structural shift. Harbor Bridge was not imagining this difficulty.
What the diagnosis reveals, however, is that the organization’s response to external pressure can create a secondary problem that compounds the original one — and that the secondary problem, once established, can become the dominant driver of poor outcomes. Harbor Bridge’s declining win rate was not caused only by a tighter grant market. It was being actively worsened by a volume-driven response that was consuming the relational capacity on which grant-making depends.
This distinction matters because it changes what the right response looks like.
If the problem is purely external — funder priorities have shifted, the market is harder — then the appropriate response is largely strategic repositioning: find different funders, develop alternative revenue streams, adjust program offerings. These are real and necessary conversations.
But if the problem is partly internal — a self-reinforcing loop that is destroying the relational infrastructure that grants require — then the most urgent response is to stop the loop before anything else.
And stopping the loop requires first seeing it clearly.
What Diagnosis Made Possible
Once the loop was visible, the path forward became considerably more tractable. A second tool, RCSolver, generated a set of recommendations organized by implementation readiness.
The immediate priority was not to write better grants. It was to stop the capacity drain. That meant implementing a Go/No-Go qualification gate before any new application was begun — a simple filter based on funder fit, relationship status, and strategic alignment. It meant protecting existing funder relationships above all else, because renewals are won on relationship depth, not proposal quality. And it meant building a systematic process for learning from recent declines, to understand what differentiated funded applications from unfunded ones.
Over the following weeks and months, the work would shift toward restructuring the pipeline: prioritizing renewals and warm relationships over cold outreach, repositioning the organization as a funder partner rather than a grant applicant, and gradually rebuilding the relational infrastructure that had been consumed by volume-driven grant-seeking.
None of these recommendations required Harbor Bridge to become a fundamentally different organization. They required it to stop doing one thing — treating grant volume as a response to grant pressure — and start doing something else: treating funder relationships as the asset that makes grants winnable in the first place.
The Diagnostic Question Worth Asking
Earlier in this series, I wrote about the double bind facing resource-constrained nonprofits: the organizations that can least afford to solve the wrong problem are also the least equipped to invest in defining the right one. The Harbor Bridge case is, among other things, an illustration of what that bind looks like from the inside.
Harbor Bridge’s leadership knew something was wrong. They had tried to fix it, repeatedly, with reasonable tools. What they lacked was not effort, intelligence, or commitment, but a structured way to step outside the problem and ask the question that the pressure of the situation made nearly impossible to ask: Are we solving the right problem?
That question — asked before more resources are committed to the existing approach — is worth more than any number of improved proposals.
If your grant success rate is declining despite increased effort, the instinct to work harder is understandable. But before you hire the grant writer, submit more applications, or redesign your impact metrics, it is worth pausing long enough to ask: is this a grant-writing problem? Or is it something else, wearing a grant-writing problem face?
The answer will determine whether the next round of effort moves you forward — or tightens the loop.
A few years ago, a large Midwest-based paint and coatings manufacturer asked me to help them crowdsource a redesign of a new industrial pump. The new pump worked beautifully in the lab, but clogged constantly when used outdoors. The engineers were convinced they had a mechanical design problem. They wanted to put a challenge out to a crowd of external solvers to get a better pump.
Before we started to design a crowdsourcing campaign, I asked the engineers one question: were the indoor and outdoor testing conditions identical?
They weren’t. Lab testing took place in summer, with indoor temperatures often in the mid-seventies. Outdoor testing happened in autumn, with temperatures rarely exceeding sixty degrees. The paint loaded into the pump was becoming viscous with even a small drop in temperature — choking the device. The engineers fixed the problem by adjusting the paint formulation. No redesign. No crowdsourcing. No new pump.
What they were about to spend was tens of thousands of dollars on service fees and prize money, at least six months running the campaign, and additional months (and money) testing and validating new designs. The actual cost to fix the problem? Essentially nothing.
The engineers had not made an engineering mistake. They had made a diagnostic one. And it is the most common kind.
The Step Nobody Examines
Here is the move that almost derailed this project — and that derails far more projects than most organizations ever realize.
The engineers observed a failure: the pump clogged outdoors but not indoors. That observation was accurate. Their next step, however, was not an investigation. It was a classification. They decided, immediately and confidently, that they were dealing with a mechanical design problem.
That leap — from “we observe a failure” to “we know what type of failure this is” — is the most dangerous step in problem solving. It is dangerous not because it is always wrong, but because it feels indistinguishable from reasoning. It presents itself as a logical inference when it is actually an assumption.
Once that classification is made, everything that follows becomes coherent. If you have a mechanical design problem, you need better mechanical design. You bring in external solvers. You evaluate proposals. You test prototypes. Every subsequent step is rational, given the premise. The premise itself is never revisited, because it was never recognized as a premise in the first place.
This is how organizations spend months and significant resources solving the wrong problem with great discipline and intelligence.
Why Confidence Makes It Worse
There is a particular irony in the pump story. The engineers were not careless. They were experienced professionals who had designed, built, and tested the pump themselves. That expertise is precisely what made the misdiagnosis so easy to make and so hard to question.
When you know a domain deeply, pattern recognition accelerates. You have seen pump failures before. You know what mechanical problems look like. The clogging fit a familiar category, and the familiar category triggered a familiar response. Confidence was not a character flaw here — it was the natural output of genuine competence applied one step too early.
This is worth reflecting on, because it means the problem is not solved by hiring smarter engineers or more experienced consultants. It is solved by building in a deliberate pause between observation and classification — a moment where the question is not “what should we do about this?” but “what kind of problem is this, and how do we know?”
That pause is uncomfortable. It can feel like doubt, hesitation, or a lack of decisiveness. In reality, it is quite the opposite: it is the discipline to avoid committing resources to a direction before the direction has been validated.
This Is Not a Manufacturing Problem
The pump story is industrial and concrete, which makes it easy to follow. But the diagnostic trap it illustrates is not specific to manufacturing, or to engineering, or to any particular domain.
Consider an organization that has tried to fix a recurring operational problem three times, with three different interventions, none of which have helped. The instinct after the third failure is usually to look harder for a better version of the same type of solution — a different vendor, a different process, a different team. The question that is rarely asked is whether the problem type was correctly identified in the first place.
Consider a leadership team preparing to invest in AI tools because competitors are moving quickly and the pressure to act is real. The stated problem is often something like: “We need to be more efficient” or “We need to automate more of our workflow.”
Those are directions, not diagnoses. The classification — this is an automation problem, this is a technology problem — has already been made, usually before anyone has examined what is causing the inefficiency or where automation would genuinely create value versus simply accelerating a flawed process.
In both cases, the leap from symptom to assumed cause type has already happened. By the time the solution is being designed, the diagnostic window has effectively closed.
Intervention Is Simpler Than You Think
The question that stopped the crowdsourcing campaign was not sophisticated. It did not require advanced analytical tools or weeks of investigation. It required only a willingness to treat the problem classification as provisional rather than settled.
“Were the testing conditions identical?” is a question about assumptions. It surfaces a factor, temperature, that had been invisible not because it was hidden, but because it had never been looked for. The engineers were looking at the pump. Nobody was looking at the conditions around the pump.
Structured problem diagnosis does not replace domain expertise; it works alongside it. The goal is not to slow down good engineers or second-guess experienced teams. The goal is to insert a brief, rigorous examination between the moment of observation and the moment of classification — to ask, before resources are committed: “What are we assuming about the problem, and what would it take to test that assumption?”
That question, asked early enough, is worth more than almost any solution.
Before You Fund the Fix
The pump story ended well and quickly. One question, asked before the campaign launched, redirected months of effort and significant cost. Not every story ends that cleanly — sometimes the diagnostic question comes later, after some of the wrong path has already been traveled. But the principle holds regardless of timing.
The most expensive mistake an organization can make is not choosing the wrong solution. It is choosing the right solution to the wrong problem. Because the right solution to the wrong problem can be executed brilliantly, on time and on budget, and still fail to change anything that matters.
Before you fund the fix, redesign the system, or automate the workflow — make sure you’re solving the right problem.
That is not analysis paralysis. It is the precondition for genuine progress.
Three articles in, the picture is not a comfortable one. A sector under compounding pressure — rising demand, shrinking capacity, exhausted leadership. Three structural problems that make the crisis so difficult to escape: the double bind that tightens with every resource constraint, the mission trap that makes honest diagnosis feel dangerous, and the nature of the work itself, which resists clean solutions almost by definition. And an apparent technological rescue — AI — that, without the right foundation, risks making things faster without making them better.
If that were the whole story, there would be little reason to continue writing. But it isn’t.
There is a discipline that addresses precisely these conditions. It is not new, it is not complicated in its core logic, and it does not require resources that small nonprofits don’t have. What it requires is a different way of thinking about problems before attempting to solve them. That discipline is structured problem-solving, and this article is about why it fits the nonprofit context specifically — and what it looks like at its core.
The Fundamental Shift
Most organizations, when confronted with a problem, move quickly toward solutions. This is understandable. It feels like action. It looks like leadership. In a resource-scarce environment, where every day of inaction has a cost, the pressure to do something is almost physical in its intensity.
But moving fast toward a solution before the problem is correctly defined is not action. It is the efficient pursuit of the wrong objective. And in a sector where resources are already stretched to the limit, a well-executed initiative aimed at the wrong target is not just ineffective — it is harmful. The resources it consumes are resources that could have addressed the real problem. The time it takes is the time the people who depend on these organizations don’t have.
Structured problem-solving begins with a different question. Not “what should we do?” but “what is actually happening, and why?” It treats the presenting problem — the symptom that triggered the conversation — as a starting point for investigation, not a destination for solutions. It insists on distinguishing between what is known and what is assumed. And it recognizes that the assumptions most worth questioning are precisely the ones that feel most certain, because those are the ones that have been unquestioned the longest.
Organizations treat assumptions as facts, especially when those assumptions are long-held, widely shared, or endorsed by senior leadership. The longer an assumption goes unquestioned, the more solid it appears — and the more dangerous it becomes.
What the Process Actually Does
At its core, structured problem-solving moves through a sequence that any nonprofit, regardless of size or budget, can follow. It begins with problem definition — not the problem as it has always been described, but the problem as it actually is, examined freshly, with the willingness to discover that reality differs from the inherited narrative. This step alone is where most organizational problem-solving fails. Not because people are careless, but because the inherited narrative feels like the truth.
It then moves to root cause analysis — the disciplined pursuit of why the problem persists, rather than what to try next. Root cause analysis doesn’t ask what we should do differently. It asks why this keeps happening. The answer is rarely comfortable. It often points to something structural, something long-established, something that implicates decisions made years ago by people who are no longer around. But it is the only answer that leads to something genuinely new.
From there, the process moves to the development of options — deliberately in plural. A process that produces only one solution hasn’t explored the solution space. It has merely confirmed a prior preference. Real options come with real trade-offs, and presenting those trade-offs honestly is what allows an organization to make a genuine decision rather than ratify an assumption.
Finally, the process moves to implementation — but implementation grounded in clear problem definition and chosen from a real set of alternatives, not grabbed at in the heat of operational pressure.
Why This Fits Nonprofits
The objection is predictable: this takes time and capacity that nonprofits don’t have. It’s a fair concern, and it deserves a direct answer.
Structured problem-solving, properly adapted for small organizations, does not require a strategy department or a consulting budget. It requires protected time — not much, but some — and the intellectual honesty to ask hard questions before committing resources. Many nonprofits already do pieces of this instinctively. What they rarely do is put the pieces together in a sequence, with enough discipline to ensure that the definition of the problem is genuinely interrogated before the solution is chosen.
The adaptation for nonprofits is not about simplifying the process until it loses its rigor. It is about making the rigor accessible — building it into the rhythms of how an organization thinks about its work, rather than treating it as a special project requiring special resources.
And this is precisely where AI, used correctly, becomes genuinely valuable: not as a replacement for thinking, but as a tool that makes the thinking faster, more thorough, and more affordable than it has ever been before.
An organization that knows what problem it is solving, and why, and has chosen its approach from a real set of options — that organization can use AI to remarkable effect. One that hasn’t done that work first will use AI to produce sophisticated answers to wrong questions instead.
Discipline comes first. Everything else follows from it.
What Comes Next
This series will spend the next several articles making this concrete. Not in the abstract, but in the specific context of the nonprofit world — its cultural dynamics, its resource constraints, its relationship with problems that are messes rather than puzzles. There are aspects of this process that require adaptation for mission-driven organizations, and those adaptations are worth examining carefully.
The goal is not to import a corporate framework into a world where it doesn’t fit. The goal is to offer something that has been largely unavailable to this sector: a rigorous, honest, practical approach to the most important question any organization can ask. Not what should we do next — but what is actually going on, and what would genuinely help.
The people this sector serves deserve organizations that ask that question. And ask it well.
This is the fourth and last article in a series on nonprofits, problem-solving, and what it takes to help organizations become more effective at achieving their missions.
There is a moment in many conversations about the nonprofit crisis when someone says, “But what about AI?”
The question is reasonable. Technology is real, the capabilities are growing fast, and the appeal to resource-stretched organizations is obvious. If AI can write grant proposals, summarize research, handle administrative tasks, and analyze program data — all at a fraction of traditional cost — then surely it offers some relief to a sector that is under pressure.
It does. And it doesn’t. The distinction matters enormously, and it is worth being precise about it.
The State of Nonprofits 2026 report surveyed 380 nonprofit leaders across the country and found that one in five is currently considering using AI tools to reduce operational costs. That is not surprising. When funding is tightening, staffing is strained, and burnout is accelerating — the same report found that nearly half of nonprofit CEOs describe their own burnout as a serious personal concern, up from under a third just a year ago — any tool that promises efficiency is going to attract attention.
The question is not whether nonprofits should use AI. Many already do, and most of them will. The question is what they are using it for, and whether that use is making them more effective or simply more efficient at what they were already doing.
Those are not the same thing.
The Efficiency Trap
AI is extraordinarily good at a specific category of tasks: processing, generating, and organizing explicit knowledge. It can draft documents, surface patterns in data, synthesize research, answer questions, and accelerate workflows. Applied to the right problems, it is genuinely transformative. Applied to the wrong ones, it produces something more insidious than failure. It produces the appearance of progress.
Consider two scenarios that are playing out in the sector right now. A nonprofit uses AI to generate more grant proposals — faster, better written, more professionally formatted than anything its small staff could produce alone. The proposals go out. Some are funded. The organization grows. But no one has paused to ask whether its programs are addressing the root causes of the problem it exists to solve, or whether years of accumulated assumptions about who needs help and why have quietly hardened into unexamined certainties. The AI has accelerated the operation. It has not improved the diagnosis.
Or consider a community organization that uses AI to automate and scale its client outreach — more contacts, faster response times, broader reach. The metrics look good. But the people being reached most efficiently are not necessarily the people most in need of services. The algorithm optimizes for what it can measure. What it cannot measure — the neighbor who doesn’t respond to digital outreach, the family that doesn’t know help exists, the community that distrusts institutions — becomes invisible precisely because the tool is working so well.
In both cases, the AI works. The process it is applied to doesn’t. This is what automating dysfunction looks like. It is not dramatic. It doesn’t announce itself. It simply makes organizations faster and more productive at pursuing objectives that were never rigorously defined in the first place.
If you haven’t defined your problem correctly, AI will efficiently generate sophisticated answers to the wrong question.
The Knowledge Gap
There is a second, less obvious reason why AI underdelivers in the nonprofit context, and it has to do with the nature of the knowledge these organizations possess.
AI operates on explicit knowledge — information that has been articulated, documented, and made available in a form the system can process. What it cannot access, by definition, is tacit knowledge: the hard-won expertise that lives in the heads of experienced staff, accumulated through years of frontline work, refined through thousands of interactions with clients and communities. This knowledge — who shows up and why, what language lands and what doesn’t, which interventions work in this neighborhood with this population under these conditions — is often the most valuable thing a nonprofit has. It is also almost entirely uncodified.
Most nonprofits don’t have a solid body of documented institutional knowledge. High turnover, constant operational pressure, and minimal administrative capacity mean that expertise walks out the door with every staff departure and has to be rebuilt from scratch. AI cannot compensate for this gap — it can only work with what has been made explicit. And in many cases, what has been made explicit is the surface of the organization’s knowledge, not its depth.
This gap — between the ocean of AI-accessible information now available to nonprofits and the interpretive infrastructure required to use it well — is precisely where problem-solving failures live.
What AI Actually Offers
None of this is an argument against using AI. It is an argument for using it correctly, which means using it within a structured process rather than as a substitute for one.
The real value of AI for nonprofits is not in generating answers. It is in improving the quality of questions. When deployed inside a rigorous problem-solving process — one that begins with honest diagnosis, interrogates assumptions, and defines the problem before designing solutions — AI becomes something these organizations have rarely been able to afford: a thinking partner.
It can help surface assumptions that have gone unexamined. It can challenge problem framing. It can map potential root causes and stress-test proposed solutions before resources are committed. It can do in hours what would otherwise take weeks, if the process directing it is sound.
Without that process, it does something else entirely. It accelerates. It scales. It makes organizations look modern while leaving the underlying dysfunction untouched.
Looking modern but performing no better is not progress. It is a more expensive version of standing still.
The sector is under enormous pressure, and the temptation to reach for powerful tools is completely understandable. But the organizations that will navigate this moment most effectively will not be the ones that adopt AI fastest. They will be the ones that define their problems most clearly — and then use every tool available, including AI, in the service of that clarity.
That is what the next part of this series is about.
This is the third article in a series on nonprofits, problem-solving, and what it takes to help organizations become more effective at achieving their missions.
The previous article drew on a Massachusetts white paper describing a workforce crisis in the human services sector: vacancy rates four to six times the state average, wages running $17,000 below the median, capacity shrinking as demand rises. A legitimate question follows: Is Massachusetts an outlier? Is this a regional story?
The answer, according to a separate national survey published this spring by the Center for Effective Philanthropy, is no. The State of Nonprofits 2026 report surveyed 380 nonprofit leaders across the United States and found a sector under compounding pressure on every front simultaneously. Nearly three-quarters of nonprofit CEOs report that demand for their services has increased since January 2025. Almost 60% say it has become harder to secure foundation grants. More than a third have seen reduced funding from federal, state, or local government sources. About 30% have reduced their staff size, most by more than 10%.
And then there is the human toll at the leadership level, which the report documents with unusual candor. The proportion of nonprofit CEOs who describe their own burnout as “very much a concern” jumped from 29% in 2025 to 46% in 2026 — in a single year. One CEO, quoted in the report, describes the situation with a precision that no statistic can match: “When funding revenue is insufficient or insecure, the staff feels it every day. They are the ones who must look into the eyes of someone seeking help and tell them that we cannot provide services for them. It is demoralizing.”
Two reports, two methodologies, two geographies. The same picture. Rising need, shrinking capacity, exhausted people doing essential work with inadequate tools.
This is the landscape. The question this series is asking is not whether the crisis is real — it clearly is — but why it persists so stubbornly, and what it would take to address it. That requires looking beneath the surface at the structural problems that make the crisis so difficult to escape. There are three of them, and they are worth examining carefully.
The Double Bind
The first is a trap with no obvious exit.
Nonprofits operate with severely limited resources — not as a temporary condition, but as the permanent architecture of their existence. Every dollar is committed. Every hour of staff time is taken. There is no slack in the system: no reserve capacity set aside for thinking, no innovation budget, no strategy team.
When an organization is running this lean, the consequences of misallocating resources are not merely inconvenient. They can be fatal. A small nonprofit that spends a year implementing the wrong solution — one aimed at a symptom rather than a root cause, or designed for a problem that was never correctly defined — may not recover. The resources spent on the wrong answer are resources that cannot be spent on the right one.
This is already painful. What makes it a bind is the other side of it. The very resource constraints that make it so critical to get things right also make it nearly impossible to invest in getting things right. Structured problem-solving — the kind of rigorous diagnostic work that distinguishes root causes from symptoms, that interrogates assumptions before committing resources — takes time, attention, and often outside expertise. Large organizations have strategy departments for this. Nonprofits, particularly small ones, don’t.
The organizations that can least afford to solve the wrong problem are also the organizations least equipped to define the right one. That is the bind. And the tighter the resource constraints become — as the State of Nonprofits 2026 data suggest they are — the tighter the bind becomes.
The Mission Trap
The second structural problem is more delicate, and it requires careful framing.
In a corporation, questioning the definition of a problem is uncomfortable, but acceptable in principle. It is a strategic conversation.
In a nonprofit, the same act can feel entirely different. The organization’s identity, its fundraising narrative, its board’s commitments, and its staff’s emotional investment are often bound up in a particular understanding of the problem it exists to solve. Questioning that understanding doesn’t feel like a strategy. It feels like a challenge to the mission itself.
This is not irrationality. It is, in fact, a completely rational response to the situation. When your ability to raise funds, retain staff, and maintain community trust depends on a coherent and stable story about what you do and why, destabilizing that story carries real costs. Leaders who protect that narrative are not being defensive — they are managing genuine risk.
But here is the quiet difficulty at the center of this: the very protection of the mission can become the thing that puts the mission at risk. When the problem definition goes unquestioned for long enough, it stops feeling like a definition and starts feeling like a fact. Assumptions harden into certainties. The question “are we solving the right problem?” becomes not just uncomfortable, but unaskable. And organizations continue to deliver services that may be well-executed, compassionately delivered, and aimed, very precisely, at the wrong target.
Puzzles and Messes
The third structural problem adds a final layer of difficulty, and it has to do with the nature of the problems themselves.
There is a useful distinction in problem-solving between what I’d call puzzles and messes. A puzzle is a problem that is well-defined, bounded, and solvable with the right method. A mess is something else entirely — a tangle of interacting conditions, shifting variables, and human behaviors that resist clean definition and rarely yield to any single intervention. Messes don’t get solved. At best, they get managed, improved, or navigated.
The problems nonprofits work on are almost always messes. Poverty, addiction, family dysfunction, mental illness — these are not puzzles. They don’t have solutions in the way that engineering problems have solutions. They require ongoing, adaptive responses that learn and adjust over time. They require honest diagnosis, not just determined effort.
The cruel irony is this: messes demand more rigorous problem-definition than puzzles do, precisely because they are so resistant to quick answers. The temptation is always to move fast, try something, and adjust later. But with limited resources and high stakes, moving fast toward the wrong destination is worse than moving slowly toward the right one.
Taken together, these three problems — the bind, the trap, and the nature of the work itself — explain a great deal about why capable, dedicated people working in well-intentioned organizations so often find themselves running hard and falling behind. It is not a failure of commitment. It is a failure of diagnosis. And diagnosis, it turns out, is precisely what the current moment makes most difficult.
In the next article, we will look at the obvious modern answer to all of this — and why, applied without care, it makes things worse rather than better.
This is the second article in a series on nonprofits, problem-solving, and what it takes to help organizations become more effective at achieving their missions.
In April 2026, the Providers’ Council, a Massachusetts human services membership association, and the UMass Donahue Institute published a white paper. It is titled Stretched to Capacity: The Workforce Crisis in Human Services. The title is not a metaphor. It is a diagnosis.
The report focuses on Massachusetts, but its findings describe a condition that is national in character. The numbers are worth reflecting upon for a moment before we move on to anything else.
As of January 2026, one in six client-facing positions in Massachusetts human services organizations was unfilled. Clinicians with independent licenses — the people most needed for mental health, substance abuse, and behavioral care — carried a 22% full-time vacancy rate. Direct support professionals, the frontline workers who assist people with disabilities in their daily lives, had a vacancy rate of 18% to 19%. The case and social workers were running at 18% full-time and 23% part-time vacancy.
To put those numbers in context: the statewide job-opening rate across all Massachusetts industries was 3.3% at the end of 2025. Human services organizations are operating with vacancy rates four to six times the broader economy — not as a temporary disruption, but as a persistent structural condition.
The wages explain part of it. The median income of a human services worker in Massachusetts runs $17,000 below the state median. And these are not simple jobs. They require emotional resilience, clinical training in many cases, and a tolerance for conditions that exhaust people quickly. The sector is asking a great deal and paying considerably less than comparable work elsewhere. State agencies — the Departments of Developmental Services, Mental Health, Children and Families — offer higher wages and defined career ladders. Community-based nonprofits cannot compete on those terms — and they don’t; they lose people instead.
And then there is the immigration dimension, which the Stretched to Capacity report documents with unusual candor. Foreign-born workers make up 24% of the human services workforce overall, and a significantly higher share in direct support roles: 35% of home health and personal care aides.
In 2025 alone, three Massachusetts providers collectively lost 205 staff members due to changes in work authorization. One-third of surveyed providers reported being greatly or somewhat affected by federal immigration and work-authorization policy changes. The consequences run in two directions simultaneously: organizations lose staff, and clients — many of them immigrants themselves — withdraw from services out of fear. The sector’s capacity contracts on both ends.
The overall picture the report paints is not of a system straining under temporary pressure. It is of a system that has been operating past its structural limits for years and is now confronting the compounding effects of workforce losses, demographic change, pandemic-era burnout, and policy turbulence all at once. More service delivery sites are opening — providers are responding to rising community need — but employment is falling at the same time. The lines are moving in opposite directions, and the gap between them is filled with underserved need.
What Numbers Say
This is the context in which nonprofit organizations do their work. And it is worth pausing here to describe what most of those organizations look like in real life, because the public image of the nonprofit sector tends to be shaped by its most visible exceptions.
According to the National Council of Nonprofits, the largest network of nonprofits in North America, 88% of nonprofits in the United States operate on annual budgets of less than $500,000. 92% operate on less than $1 million. Only 3% to 5% have budgets exceeding $5 million.
The large, professionally staffed, nationally recognized organizations are real, but they are not representative. The sector is overwhelmingly composed of small and very small organizations, most of them deeply embedded in specific communities, most of them running on thin margins with minimal administrative infrastructure.
These are the organizations absorbing the workforce crisis described above. They don’t have HR departments to manage the vacancy problem. They don’t have strategy teams to think through the structural issues. They don’t have the budget to hire consultants. They have people — usually not enough of them — working hard to keep programs running for the populations that depend on them.
This is not a critique; it is a description. The people who work for and lead these organizations are, in most cases, doing remarkable work under genuinely difficult conditions. My point is that the resources available to the sector, especially at the small-organization level, are not commensurate with the problems it is being asked to solve. The gap between mission and capacity is not a management failure. It is a structural feature of the landscape.
The Question We Need to Answer
None of this would matter quite as much if the problems nonprofits were trying to solve were simple ones. But they aren’t. They involve human behavior, systemic poverty, mental illness, housing instability, addiction, and family breakdown — conditions that resist clean solutions and rarely yield to any single intervention. The sector is navigating extraordinarily complex territory with extraordinarily limited tools.
This gap — between the complexity of the problems and the resources available to address them — is where things get genuinely difficult. Not just operationally difficult, but intellectually difficult. When an organization is stretched to capacity just keeping services running, it has very little left over for the harder and more important work of asking whether those services are addressing the right problems in the right ways.
That question — whether resources, already scarce, are being aimed at the right targets — is what this series will explore. It is not an abstract question. It has real consequences for real people.
In the next article, we will look at three structural problems that make it so difficult for nonprofits to answer the above question well. The first is a bind that tightens the more you try to escape it.
This is the first article in a series on nonprofits, problem-solving, and what it takes to help organizations become more effective at achieving their missions.
There is a well-known framework for understanding how people respond to loss. Elisabeth Kübler-Ross identified five stages of grief: denial, anger, bargaining, depression, and acceptance. It was meant to describe how individuals cope with death and dying. It turns out it also describes, with uncomfortable precision, how societies respond to disruptive technology.
With artificial intelligence, we quickly moved through the denial stage. For a brief moment, the consensus was that AI was a clever parlor trick — impressive at chess, useless for anything that actually mattered. That window closed fast. Now we are firmly in the anger stage, and by all appearances, we intend to stay there.
The anger takes two forms, and it is worth distinguishing between them.
The first is trivial, though it consumes a disproportionate amount of energy. It is the sport of finding what AI cannot do. When an AI system miscounts the letter “r” in “strawberry,” or produces a garbled map of Europe, a certain kind of observer erupts in applause. These failures are real. They are also beside the point. No one cancelled the space program because early rockets exploded on the launchpad. No one concluded that surgery was a failed experiment because early anaesthesia killed patients. Every powerful technology is, in its early stages, also a failing one. The correct response to an imperfect tool is to improve it — or, in the meantime, to understand its limitations and work around them. It is not to declare the entire enterprise fraudulent.
The second form of anger is more legitimate and deserves a more serious response. The concerns about AI’s societal impact — large-scale displacement of workers, erosion of cybersecurity, the amplification of disinformation — are not invented. They are real risks, and anyone who dismisses them entirely is not paying attention. But here, too, perspective matters.
Consider what we chose to do with nuclear technology before we learned to use it for clean energy and medicine: we dropped two bombs on civilian populations, killing hundreds of thousands of people. Consider recombinant DNA technology, which gave us insulin, cancer therapies, and vaccines for diseases that once killed millions — but not before it was weaponized to produce more lethal biological agents. Consider the internet, which has done more to advance commerce, education, and human connection than almost any invention in history — and which was also, from its earliest days, a vector for fraud, hate speech, and the organized manipulation of public opinion.
In each case, society did not conclude that the technology was irredeemable. In each case, we developed — imperfectly, slowly, often reactively — the regulatory frameworks, cultural norms, and institutional safeguards needed to contain the worst and amplify the best. We will need to do the same with AI. That work is urgent, and it’s doable. But it is not an argument against AI. It is an argument for taking it seriously.
So why does AI attract a quality of suspicion that we did not, in hindsight, extend to nuclear fission or genetic engineering? Partly, perhaps, because this technology feels more intimate. It does not require a reactor or a laboratory. It sits on a laptop. It writes, reasons, and converses. It mimics — sometimes uncannily — what we have always thought of as distinctly human. That proximity is unsettling in ways that a uranium centrifuge is not.
But unsettling is not the same as dangerous, and dangerous is not the same as irredeemable.
The Kübler-Ross framework ends, eventually, in acceptance — not passive resignation, but the kind of clear-eyed reckoning that makes constructive action possible. We are not there yet with AI. But the path there does not run through a catalogue of its failures. It runs through an honest accounting of what it can already do, a rigorous effort to address its risks, and the intellectual honesty to tell the difference between a tool that is imperfect and one that is unwelcome.
Those are not the same thing. And until we stop confusing them, we are mostly just grieving.
This is the fifth and final post in the series “Problem First: AI-Assisted Problem Solving for Organizations That Can’t Afford to Get It Wrong.”
After I published the second post in this series, the one laying out a five-stage problem-solving process, a good friend of mine, Michael Fruhling, asked a question that stopped me cold: Who in the organization is actually responsible for running this process?
I didn’t have a good answer. And that was revealing, because it exposed a gap not just in my argument but in the way most of us think about problem solving. We describe the process. We explain why it matters. We demonstrate what goes wrong when it’s skipped. But we rarely address the most practical question of all: who does this work?
This post is my attempt to answer that question—honestly, including the parts I’m still thinking through.
The Ownership Problem
In large organizations, the question of who owns a structured process is familiar territory—and still contentious. Decades of debate about innovation management have produced no consensus on how to organize innovation teams: should they be embedded in business units or operate independently? Should they report to the CEO or to a functional leader? Should they be permanent or project-based?
Problem-solving ownership faces all the same organizational tensions, and then some. Innovation teams, whatever their flaws, at least have a recognized mandate: generate new ideas, explore new markets, develop new products. A problem-solving team has a harder pitch: slow down, question your assumptions, and make sure you’re working on the right thing before you commit resources. That’s a valuable mandate, but not a popular one.
For large corporations with strategy departments and experienced leadership teams, the ownership question is at least conceivable, even if the answer varies. Someone can be assigned. A team can be formed. Budget can be allocated.
For SMEs and nonprofits, the question is more fundamental. These organizations don’t have spare capacity to dedicate to process ownership. The people who would run a structured problem-solving effort are the same people handling operations, fundraising, client services, and everything else. There is no one left over.
Three Ways to Organize the Work
I don’t think there is one correct answer to the ownership question. But I do think there are three viable models, each suited to different organizational realities.
Model 1: The external facilitator. An outside expert—a consultant, advisor, or specialist in structured problem solving—guides the organization through the process. The organization provides subject-matter expertise: the people who live with the problem and understand its technical, operational, and human dimensions. The facilitator provides process discipline: the structure, the sequencing, the right questions at the right moments, and the resistance to premature solutioning that insiders often can’t maintain.
This model works well when the problem is high-stakes or unfamiliar, when internal biases are strong, or when the organization has never been through a structured diagnostic process before. Its limitation is cost. Hiring an external facilitator is an investment, and for budget-constrained organizations—especially the small nonprofits operating on less than $500,000 a year—it may not be feasible for every problem that arises.
Model 2: The internal problem-solving lead. One person within the organization—not necessarily full-time, but clearly designated—takes ownership of the problem-solving process. This person doesn’t need to be the subject-matter expert. In fact, it’s often better if they aren’t, because their job is to maintain process integrity, not to have the answers. They convene the right people, ensure that each stage is completed before moving to the next, and serve as the organizational conscience that asks, “Are we sure we understand the problem?”
This model is more sustainable for organizations that face recurring problems and want to build internal capability. The challenge is that it requires someone with a specific and uncommon skill set: analytical rigor, the ability to push back on senior leaders, comfort with ambiguity, and the patience to slow a process down when everyone else wants to speed it up. Finding or developing that person is not trivial.
Model 3: Guided self-service. The organization works through the problem-solving process itself, using structured tools and frameworks, with bounded external support at key checkpoints. Think of it as a middle path: the organization does the thinking, but within a scaffolding that prevents the most common mistakes—weak problem framing, unexamined assumptions, premature convergence on solutions.
This model is particularly promising for SMEs and nonprofits, because it builds internal problem-solving capability while keeping costs manageable. The external support is focused where it matters most: at the intake stage (making sure the problem is framed correctly), at the midpoint (validating the direction of root-cause analysis), and at the end (reviewing solutions and helping translate analysis into action). Between checkpoints, the organization drives the process.
Tools for the Journey
The guided self-service model depends on having the right tools—not just AI in the abstract, but structured instruments designed specifically for the problem-solving process.
This is something I’ve been working on for some time. Over the course of my consulting practice, I’ve developed two AI-assisted tools that operationalize the five-stage process described in this series. They sit on the same problem-solving arc but serve different purposes.
The first, RCFinder, is a convergence engine. It guides users through the diagnostic stages of the process: framing the problem as it is experienced, clarifying assumptions and constraints, and generating multiple root-cause hypotheses. Its purpose is to create clarity before solutions are considered—to resist the Tylenol reflex by ensuring that the problem is properly understood before anyone reaches for a fix.
The second, RCSolver, is a divergence engine. It generates diverse, actionable solution options mapped to specific causes identified by the root-cause analysis. It surfaces trade-offs, risks, and constraints. It avoids the best-practice dumping that passes for solution generation in too many advisory contexts.
Each tool can be used independently. But together, they form a complete workflow—one guided journey from understanding to action. The tools don’t replace human judgment; they structure it. They ensure that the process moves through each stage with discipline, producing documented outputs—problem statements, assumption maps, root-cause analyses, solution portfolios—that become part of the organization’s institutional memory.
I offer these tools both through a done-for-you (DFU) consulting engagement, where I manage the process and involve the client at key checkpoints, and through a guided self-service (DIY) option, where organizations receive structured access along with onboarding and expert support at critical decision points. Both formats reflect the same conviction: the process matters more than the tool, and the tool is only as good as the discipline surrounding it.
What I Don’t Know Yet
I want to be candid about what remains unresolved.
The question of who owns problem solving inside an organization is, ultimately, a question about organizational culture. You can designate a lead, hire a facilitator, or deploy the best tools available—but if the culture rewards the Tylenol reflex, if the person who says “wait, are we sure about this?” is seen as an obstacle rather than an asset, the process will be undermined regardless of how well it’s designed.
I don’t have a formula for changing that culture. What I do believe is that the process itself, practiced consistently, can trigger shifting norms over time. When an organization works through a structured diagnostic process once and discovers that the problem it thought it had wasn’t the actual problem—as my paint company clients did, as the food manufacturer did—the experience creates a kind of organizational muscle memory. The next time a problem arises, there is at least a voice in the room that says, “Last time, we assumed we knew. We were wrong. Let’s check.”
That voice, over time, is how cultures change.
Where This Goes Next
This is the final post in the “Problem First” series, and I want to close by looking forward.
Over the course of five posts, I’ve built an argument that moves from the general to the specific: why organizations are bad at problem solving, what a rigorous process looks like, who needs it most, how AI fits in, and how the work gets done. The arc has been deliberately broad because the principles apply across sectors and organizational sizes.
But as regular readers have probably noticed, my center of gravity has been shifting. With each post, I’ve found myself drawn more deeply toward one particular community: nonprofits.
This isn’t accidental. The more I research the nonprofit sector—its scale, its constraints, its extraordinary commitment alongside its structural vulnerabilities—the more convinced I become that this is where structured problem solving can make the most profound difference. 88% of nonprofits in the United States operate on less than $500,000 a year. Many have been serving their communities for decades, led by deeply experienced professionals. They don’t lack dedication or expertise. They lack the resources and infrastructure to apply that expertise systematically to the problems that matter most.
That is the gap I want to help close. My next posts will focus specifically on the nonprofit world: the unique challenges these organizations face in defining and solving problems, the cultural dynamics that make honest diagnosis so difficult, and how AI-assisted tools can be adapted to serve mission-driven organizations on mission-driven budgets.
The series title has been “Problem First.” The principle doesn’t change. The audience gets more specific. And for these organizations—the ones doing the hardest work with the fewest resources—getting the problem right isn’t a competitive advantage. It’s a moral obligation.
