You Approved the License. You Didn’t Approve the Work

It is June. For most organizations operating on a calendar fiscal year, that means one thing: the gap has arrived.

Not the gap between what was planned and what was possible. That one has always been there. The gap between what the board approved in January and what the organization actually needed to make it work — that one is visible now, in staff frustration, in systems that aren’t being used, in the quiet recognition that the platform that was supposed to change things has changed very little.

The technology didn’t fail. The budget did. And not the technology budget specifically — the structural logic that produced it.

Technology is just where the failure becomes visible. It comes with invoices, licenses, and expiration dates — hard evidence of what was purchased and what wasn’t funded. But the same logic that underfunds software also underfunds training, supervision, financial systems, HR infrastructure, and the organizational capacity required to sustain complex work over time. Technology didn’t create this problem. It just made it harder to look away from.

This is the fourth piece in the Foundry Fundraising series on what’s actually broken in nonprofit technology. The first three addressed nonprofit standards (“The Standards Problem”), nonprofit decisions (“The Decision Problem”), and vendor accountability (“The Vendor Problem”). This one is addressed to the funders and boards who set the structural conditions before any of that plays out.

You are not bystanders in this story. You are architects of it.

The Overhead Myth Is Costing You the Outcomes You Say You Want

The nonprofit sector has been managed toward a fiction for decades.

The fiction: administrative and operational spending is waste. The metric: overhead percentage. The outcome: organizations that have systematically starved the infrastructure required to do their work — including, consistently, technology.

Funders who restrict grant dollars to program-only spending are not being fiscally responsible. They are transferring their discomfort with operational investment directly onto the organizations they fund — and then measuring those organizations on outcomes produced by infrastructure they refused to support.

This is not a fringe critique — and the people who made it mainstream have since tried to disappear from the conversation.

The Overhead Myth campaign was organized by GuideStar, Charity Navigator, and the Better Business Bureau’s Wise Giving Alliance. They sent an open letter to the donors of America in 2013 formally repudiating overhead percentage as a meaningful measure of nonprofit effectiveness. The research, they acknowledged, was unambiguous: organizations that underinvest in infrastructure produce worse outcomes, not better ones.

What they did not do — and have not done in the decade since — is reckon with their role in building the thing they were dismantling. GuideStar spent years as the infrastructure of overhead-ratio culture. Charity Navigator built its brand on star ratings that made overhead the metric that mattered. The BBB’s Wise Giving Alliance set the standards that donors cited when they refused to fund operations. Together, they industrialized a framework that systematically starved nonprofits of the capacity to do their work — and no budget line absorbed that pressure more consistently than technology. Technology looks like overhead. It doesn’t serve clients directly. It doesn’t appear in program reports. It is precisely the kind of investment that overhead-ratio culture trains funders to cut and donors to distrust.

But technology wasn’t the only casualty. The same funding logic systematically underfunded training, supervision and coaching, financial infrastructure, HR capacity, and the middle management layer required to translate strategy into execution. Technology is the most legible case because it comes with a price tag and a failure mode you can point to. The rest of the damage is harder to locate — until the staff turnover numbers arrive, or the program quality drops, or the organization discovers it has been running complex work on spreadsheets and collective memory for a decade.

The campaign was real. The accountability was not. A letter to donors does not undo a decade of infrastructure. Candid — GuideStar’s rebranded successor — has moved on. Charity Navigator has updated its methodology. The organizations they helped condition have not recovered the funding they lost to overhead anxiety, and the funders who internalized those norms are still writing program-only grants.

They created the monster. They issued a press release. Then they moved on — leaving behind a funding culture that still treats operational investment as waste, and organizations that cannot get a grant to train staff on the platform the previous grant purchased.

Most nonprofits operate with minimal technology investment, often under 3 percent of their budgets — a fraction of what for-profit organizations spend, according to a 2025 Chronicle of Philanthropy survey. NTEN’s funder research finds that only one in five funders provide dedicated technology funding at all, and the organizations that do receive it report the budget rarely covers training. Total cost of ownership runs two to three times the advertised price — a gap almost entirely composed of implementation, migration, training, and governance costs that funders routinely decline to support.

The license gets funded. The work does not.

What Board Approval Actually Approves

Inside the organization, the failure has a different author: the board that approves the platform without understanding what the platform actually costs to implement.

This is not malice. It is a failure of governance literacy — and specifically, a failure to understand what the board’s role in a technology decision actually is.

It is not to select the platform. It is not to evaluate features, debate vendors, or weigh in on implementation timelines. Boards that go that deep into operational decisions create a different problem: paralysis, second-guessing, and the slow suffocation of staff authority. That failure is real, and it is common.

But the alternative to micromanagement is not rubber-stamping. A board that approves a license fee and calls it a technology decision has confused signing off on a number with exercising judgment about an investment. Those are not the same thing.

The board’s role is narrower than micromanagement and larger than approval. It is to ask: what does this actually cost, in total? What are the conditions required for this to succeed? Who owns it? How will we know if it’s working? Those are not operational questions. They are governance questions — and they are questions most boards never ask, because no one has told them they’re supposed to.

The license fee is the smallest number the organization will pay. It does not include data migration, configuration, training, governance structure, or the ongoing management costs that compound for years. A board that approves the invoice without interrogating the full investment hasn’t made a technology decision. It has signed an installment plan on a system the organization is not prepared to use.

Eighteen months later, when the CRM sits underutilized, staff have built workarounds, and the development director is making the case for a different platform, the board will ask what went wrong. The answer is: this meeting. This approval. The questions that weren’t asked.

Magical Thinking Has a Board Resolution

There is a pattern so common in this sector it has become invisible.

A technology need is identified — usually urgently, usually by someone with a specific problem they need solved. A platform is selected, often under time pressure and without adequate evaluation, because the urgency doesn’t leave room for rigor. A budget line is created for the license fee. The board approves it. The ED announces it. Everyone exhales.

And then the real work begins — and immediately starts not happening.

Implementation takes longer than projected. Data is messier than anticipated. The staff champion who drove the selection has seventeen other priorities and no dedicated time for configuration. Training gets scheduled, then rescheduled, then quietly dropped from the calendar because there is always something more urgent. The vendor’s onboarding materials assume a level of organizational readiness that doesn’t exist. The governance structure that was supposed to own the system never gets formally built, so informal workarounds fill the gap.

Six months in, the person who championed the platform has moved to a different role or left the organization entirely. The system is technically live. No one is using it the way it was designed to be used.

This is not a technology failure. It is not a staff failure. It is the predictable consequence of an organization that treated the license fee as the investment and everything else as execution details — as though the hard part was the purchase, and the rest would follow.

It won’t. It never does.

The magical thinking is not that organizations believe technology will solve their problems. It is that they believe it will solve their problems without deliberate investment in the conditions required to make it work. Planning. Training. Governance. Time. None of those appear in the vendor’s sales deck. All of them are required.

You cannot approve the destination and refuse to fund the road.

The Mid-Year Reckoning

June is when the math becomes visible.

Organizations that entered the year with optimistic technology plans are now six months in. The variance report tells the story: staff time absorbed by implementation that wasn’t budgeted, training that hasn’t happened because the budget didn’t include it, a system that’s running but not producing what was promised.

This is the moment when organizations make the wrong decision for the second time.

The wrong decision: conclude that the technology failed, and begin the conversation about replacement.

The right decision: recognize that the technology was never given the conditions to succeed, and make a deliberate choice about whether to invest those conditions now or exit deliberately.

That distinction matters. A reactive technology replacement — entered into because the original investment was underfunded — will produce the same outcome, because the structural conditions haven’t changed. The new platform will be approved at license-fee cost. The real work will be underfunded again. The cycle will repeat, with a new name in the contract and the same gap in the budget.

The mid-year reckoning is not a technology problem. It is a governance problem arriving on schedule.

What Funders Need to Own

If you fund program outcomes, you fund the infrastructure that produces them.

There is no version of this where those are separable. A development team running on a CRM no one has been trained to use is not going to produce the donor retention numbers you’re measuring. A communications team managing campaigns on a platform they’ve never been properly onboarded into is not going to produce the community engagement metrics in your grant report. The outcomes you are funding require operational conditions you are actively disincentivizing.

Operational investment is program investment. The sector has known this for decades. The funding behavior hasn’t caught up.

Concretely: if you fund an organization to do work that requires technology, your grant should either include the full cost of that technology — implementation, training, governance, and ongoing management, not just the license — or it should not restrict operational spending so the organization can fund it elsewhere. A restricted grant that covers the platform but not the conditions required to use it is not a technology investment. It is a technology gesture. Anything less is funding outcomes you have made structurally difficult to achieve, and then measuring the organization against them.

What Boards Need to Own

Technology literacy is a governance responsibility.

Not technical expertise — no one is asking boards to evaluate software architecture or debate implementation methodologies. Literacy. The ability to ask the right questions before a vote, and to create the organizational conditions that make honest answers possible.

In practice, that means four things.

First: no technology investment gets approved without a total cost of ownership. License fee, implementation, data migration, training, governance, and ongoing management — all of it, projected over three years. If that number hasn’t been built, the vote doesn’t happen.

Second: every technology approval names an owner. Not “the technology committee will oversee this” or “IT will manage it.” A person, with the authority and the time allocated to do the job. Governance structures that exist only on paper fail on schedule.

Third: adoption gets built into the approval. A board that votes yes on a platform and checks back in eighteen months has abdicated its oversight responsibility. Approval should include a defined review point at ninety days — not to micromanage implementation, but to ask: are the conditions for success present? Is training happening? Is someone actually in charge?

Fourth: boards need to create the budget conditions that allow staff to bring the full ask. When staff present a technology request at license-fee cost, it is often not because that’s all it costs. It is because they have learned — correctly — what the organization is willing to approve. A board that signals it will only fund the invoice will receive proposals written around the invoice. That is not a staff problem. It is a board culture problem, and it belongs to the board to fix.

The board’s job is not to manage the technology. It is to ensure the organization has what it needs to make the technology work — and to hold leadership accountable when it doesn’t.

A board that shows up to vote and disappears until the next budget cycle hasn’t done governance. It has done paperwork.

The Design Failure Beneath the Execution Failure

This series has made the case, piece by piece, that most nonprofit technology problems are not technology problems.

Piece one: they are standards problems — the accountability organizations bring, or fail to bring, to vendor relationships.

Piece two: they are decision problems — the choices made before, during, and after adoption.

Piece three: they are vendor accountability problems — the structural exploitation of a sector that has been too patient for too long.

This piece makes the final argument: underneath all of it is a design failure.

Funders who have systematically disincentivized operational investment. Boards that approve technology at a fraction of what it costs to implement. A sector-wide pattern of underfunding the conditions required for success, and then measuring outcomes as though those conditions were present.

The technology didn’t fail.

The infrastructure required to make it work was never built — because the people with authority and resources to build it treated the license fee as the investment, and everything else as a detail.

Technology just made that decision legible. The same logic that underfunded the CRM also underfunded the people, training, financial systems, and organizational capacity surrounding it. Technology shows you the bill. Infrastructure underinvestment is the line item behind it.

Leadership is what changes.

The question is whether the people in the room with the authority to act on it are willing to stop treating the invoice as the investment.

What Doesn’t Change

The license fee is the smallest number you will pay.

If the budget doesn’t fund the conditions for success, the approval was theater.

Burnout is a design failure. So is this.