Starting an SGO: Who Is Actually Going to Run It?

Every organization that looks seriously at the federal scholarship tax credit ends up in the same meeting. Somebody has read enough to know the credit is real, believes their families would use it, and asks what it would take to start a Scholarship Granting Organization.

The answer they get is almost always about formation — incorporation, the IRS, state registration, seating a board. That is a real project, and we have written the whole sequence down. But formation is a project with an end date, and it is the easy half. The half nobody costs out is the one that begins the day after: somebody has to run the thing, every week, indefinitely.

This post is about that half. It is also about a distinction that opens up an option most organizations never consider, because they collapse two separate questions into one.

The Education Freedom Tax Credit (EFTC), the Federal Scholarship Tax Credit (FSTC), the Educational Choice for Children Act (ECCA), and Section 25F are four names for the same federal program — a dollar-for-dollar tax credit of up to $1,700 per year for donations to scholarship granting organizations, effective January 1, 2027.

The Two Questions People Ask As One

"Should we start our own SGO?" is not one question. It is two, and they have different answers.

Whose SGO is it? Who holds the 501(c)(3), sits on the board, writes the eligibility rules, and has their name on the receipt a donor files with their tax return.

Who does the operating work? Who receipts the gifts, screens the applications, verifies the incomes, prepares the docket, moves the money, keeps the books, and files the reports.

Almost everyone answers these together — "it's ours, so we run it" or "we can't run it, so it can't be ours" — and in doing so eliminates the middle option before it is ever on the table. There are three combinations that actually exist:

• Your SGO, you run it. You own the entity and staff the operation.
• Your SGO, somebody else runs it. You own the entity; the operating work is done by a team you hire rather than employ.
• Somebody else's SGO, they run it. You do not own an SGO at all. A school joins one that already exists.

Which of those is right for you depends on facts about your organization, not on which one sounds most serious. We compare all three in detail, including where the answer is "not us," at Start or Join an SGO. What follows is the part that decides it: what the job actually is.

The Job Nobody Writes Down

Here is the operating work, stated plainly. Not the compliance framework in the abstract — the things a person has to do, and when.

On every gift. A qualified contribution has to be cash — card, ACH, check, wire, or currency. Appreciated stock does not qualify. A donor-advised fund grant does not earn the credit, because the credit runs to an individual taxpayer. Each gift is designated to a state at the moment it is made, tracked against that donor's $1,700 annual cap ($3,400 filing jointly), and receipted with the unique donor number the program uses instead of a Social Security number. If the receipt is wrong, the donor's credit is at risk, and you will hear about it in April.

On every applicant. Household income has to be verified against 300% of the area median — the area where that family actually lives, not a national figure, which means the same income qualifies in one county and does not in the next one over. Somebody collects the documents, reads them, applies the household-size adjustment, and writes down why the determination came out the way it did.

On every award. Decisions must be made at arm's length by an independent committee, against a written policy, with conflicts screened — and committee members' own families are disqualified from receiving scholarships from that SGO. No gift may be earmarked to a school or a student, so a donor's preference is an input the committee may weigh and can never be an instruction it follows. Awards have to reach ten or more students who do not all attend the same school. Somebody assembles that docket, staffs the meeting, and keeps the minutes.

On every dollar out. Tuition goes to a school; other qualified expenses go out on a restricted instrument or come back as a receipt-verified reimbursement. Each release is reconciled, dual-controlled, and drawn from the same state's segregated account the gift landed in. Money designated to one state never funds a student in another.

Every month. The 90/10 test runs per state account. At least 90% of qualified contributions must reach students, and the operating side is a withdrawal cap, not an expense budget — a distinction that surprises nearly everyone the first time, and which we unpack in The 90/10 Rule Is a Withdrawal Cap, Not an Expense Rule. Processing fees come out of the same side of the line, which is why card fees are a compliance question and not just a finance one.

Every year. State annual reports, in every state you are listed in. An audit from year one. Records retention. A compliance calendar with real deadlines on it.

And then the part that never appears in any framework: a parent calls in March wanting to understand why she was denied. A donor's card fails on December 30 and the credit is a calendar-year credit. A school changes its bank details and somebody has to notice before the wire goes. A family's documents arrive as photographs of a phone screen. This is not exotic work. It is just work, and it does not stop.

Three Things That Make It Harder Than It Looks

It is continuous, not seasonal. Development offices are built around seasons — an annual fund, a gala, a year-end push. Scholarship administration has no off-season. Gifts arrive in July. Applications close, and then appeals open. The 90/10 test has to hold on every day of the year, not on December 31.

The role has no veterans. The credit takes effect January 1, 2027. Nobody on earth has five years of experience administering it, because the statute is younger than that. An organization deciding to staff its own operation is not choosing between a good hire and a great one. It is choosing between an unproven hire and no hire — in the first year of a program, which is exactly when mistakes are cheapest to make and most expensive to explain.

Failure here is quiet. Nothing about a half-staffed SGO looks like a crisis. Receipts go out a few weeks late. One income verification gets taken on faith because the family is obviously eligible. A release nudges an account past the cap and nobody runs the number until quarter end. A state report gets assembled from a spreadsheet that was never reconciled against the bank. None of that triggers an alarm. All of it surfaces in an examination, at once, eighteen months later.

Option One: Own It and Run It

You form the nonprofit, seat the board and the scholarship committee, and operate the program with your own staff on software built for it.

What it buys you. Your board writes the eligibility criteria — who qualifies, what a priority is, how large an award is. Your brand is on the giving pages, the receipts, the family application and the tax documents. You choose which states to operate in and when to expand. And the administrative share of every gift is your organization's revenue rather than a vendor's.

What it costs you. Months before the first gift. A board that genuinely meets and a committee independent enough to survive the disqualified-person rules — which, in a small and tightly connected community, is harder to seat than it sounds. And a person whose job includes the 90/10 report and the state filing, every year, forever.

Who it is for. An organization that represents several schools, already has development staff or a foundation, has a real view about who should receive a scholarship, and can name the person who will own compliance on Monday morning.

Option Two: Own It, and Have Someone Else Run It

This is the option the collapsed question hides. You form the nonprofit and seat the board exactly as above. The entity is yours, the policy is yours, the brand is yours, and your committee decides every award. What changes is who does the desk work.

That is what ClearPath Managed is: gift processing and receipts, application intake, income verification, docket preparation, disbursement, per-state 90/10 bookkeeping, state reports, and an audit package maintained continuously rather than assembled in a panic — done by a team that does this for other programs too, with your organization's name on everything a donor or a family sees.

The line that does not move. We prepare the docket. Your committee decides. Section 25F requires awards to be made at arm's length by the granting organization itself, so no service provider can vote on a scholarship, and any arrangement that implies otherwise is one an auditor will unwind. Governance is not the part you are outsourcing. The desk work is.

What it actually costs. The same thing running it yourself costs, pointed at a different payroll. The managed fee comes out of the operating allowance — the same share that would otherwise pay your administrator, your audit prep and your systems. You are not paying extra. You are paying somebody else.

Which is exactly why the arithmetic flips with scale. On $2 million of annual giving, the operating allowance is up to $200,000, and that can fund a real internal team; the case for keeping it in-house gets strong. On $200,000 of giving it is up to $20,000, which does not fund one competent full-time person anywhere in the country — and pretending otherwise is how programs end up out of compliance. (Remember that the allowance is a ceiling on withdrawals, not a budget you are handed, and the platform and processing fees live under the same ceiling.)

What you give up. The administrative share stops being institutional revenue and becomes a cost. And your team does not build the tacit feel for the program that comes from doing the work — an organization whose staff have never processed a disbursement knows its own operation less well. Both are real trade-offs. Neither is permanent: the entity, the donors and the history are yours the whole time, so bringing operations in-house later is a change of who logs in, not a migration.

Who it is for. An organization with the donors, the mission and a functioning board, that is not going to build a back office to get a scholarship program — and that would rather be an owner than an employer.

Option Three: Join an SGO That Already Runs

The third option is not owning an SGO at all. A school joins one that already exists and is already certified, as a partner school: a branded giving page and QR code, donors able to name your school as their preferred school, and one recurring job — confirming that a student is actually enrolled, which is the one thing no SGO can verify from the outside. Every obligation stays with the SGO.

For a single campus this is usually not the lesser option; it is the only workable one. An SGO's awards have to reach ten or more students who do not all attend the same school. The bar is not ten schools — but an entity formed to fund one campus cannot lawfully award only to that campus, and no committee running that close to the line will read as arm's length. We laid out the structural argument at why one school should join rather than form. If you are choosing between SGOs to join, twelve questions worth asking will tell you more than any brochure.

Status, plainly: our own ClearPath Partner Schools program is not open yet. We are standing the SGO up ahead of the January 1, 2027 start of the credit, and schools can join the early-access list at no cost and no commitment. Forming your own SGO — with your staff or with ours — is something you can begin today.

Four Questions That Settle It

Skip the feature comparison. Answer these in order.

1. Does your organization have a view about who should get a scholarship that a general-purpose SGO could not implement? "Low-income families in our county" is implementable by anybody. "Students across our diocese, weighted by parish participation and assessed by our own aid office" is not. A specific answer means the SGO needs to be yours.

2. What will you realistically raise in year one? Not the ambition — the bottoms-up number from donors you can name. The fixed costs of forming and auditing an SGO land very differently against $2 million than against $80,000. Our state-by-state market math is a reasonable sanity check on the ceiling.

3. Can you name the person who owns compliance on Monday morning? Not a consultant for the launch. A person on staff whose job description includes the 90/10 report and the state filing, every year.

4. Do you need the administrative share to be revenue, or can it be a cost? If the program's economics only work when the allowance funds your own staff, you need to run it. If it can be an expense like any other, that constraint disappears.

Now notice how they resolve. Questions 1 and 2 decide whether the SGO should be yours. Questions 3 and 4 decide who runs it. Answer them in that order and the path falls out on its own:

• Yes to 1 and 2, yes to 3 and 4: form it and run it.
• Yes to 1 and 2, no to 3 or 4: form it and have it operated for you.
• No to 1 and 2: join one, and put your energy into raising money instead of administering it.

The Mistake That Costs the Most

It is not picking the wrong option. It is answering question 3 first and letting it decide question 1.

Organizations do this constantly. Somebody says "we don't have anyone who could run this," everyone nods, and the SGO conversation ends there — even when the answers to questions 1 and 2 were both an emphatic yes, and the organization had exactly the donor base and the mission specificity that justify owning one. A staffing constraint gets treated as a verdict on ownership, and a program that should have existed does not.

The reverse mistake is rarer but more expensive: forming an entity, discovering the program raises $60,000, and carrying a board, an audit and a compliance function against it indefinitely.

Both come from the same error, which is treating "should this be ours" and "who does the work" as one question with one answer.

What to Do Before January

The credit's start date does not move, and formation takes four to six months from decision to first qualified contribution. From here, the fast end of that range gets you live for the first giving year. The slow end does not.

That timing has a practical consequence worth being blunt about: if you are leaning toward owning an SGO, the formation work is identical whether you or somebody else ends up operating it. So form the entity now and settle the staffing question in parallel. The months are the constraint; the org chart is not.

• Check your state. You can only accept qualified contributions in a state that has opted in. Current status is on our state tracker, sourced against the IRS participating-state list.
• Answer the four questions above with your board, in that order, and write the answers down.
• If the SGO should be yours, start formation now — ClearPath Launch does the filing work — and decide the staffing question while the paperwork runs.
• If it should be yours but you will not staff it, ClearPath Managed is the detail: what we do, what never leaves your board, and how the handover back to your own team works if you want it later.
• If you are one school, get on the partner-school early-access list and spend the next four months building the donor list you will use either way.

The comparison across all three, dimension by dimension and with an explicit "probably not you if" for each, lives at Start or Join an SGO. The shorter version of the argument is in Three Ways In.

Whichever one you pick, the destination is the same: a family that can afford the school that fits their child. The only question this post is asking is who is going to do the work between here and there — and whether you have honestly answered it, or quietly assumed it away.