Raising money for a small business: the five ways, ranked by how often they work

Start with the arithmetic that decides everything else. A business either can be worth a hundred times what someone puts into it, or it can't. Both are fine; they are simply funded by different money. A mobile bike repair in Denver with one van is a good business — it can pay a founder well, it can employ people, it can be sold one day for a multiple of its profit. It is not a venture business, and an investor whose fund needs a thirty-fold return cannot fund it, however much they like it or you. That is not a judgement on the business. It is a judgement on the shape of the money.

Getting that wrong is the expensive mistake, because it costs a season. Three months of pitching, of warm intros that go nowhere, of rewriting a deck — while the thing that would have worked, which is selling more and spending less, sat untouched. The Federal Reserve's 2025 Small Business Credit Survey found that 60% of employer firms applied for financing in the year before the survey, and that when they did they applied to banks and online lenders — a loan, a line of credit, an advance. Equity barely features, because for most firms it isn't on offer.

So the honest ranking below isn't pessimism. It's the order in which these things work for a business with revenue under a million dollars, and following it means you get to the money that will actually say yes before you run out of months.

Ranked for a small business under about $1m of revenue — a service business, a shop, a trade, a small software company with customers. The order changes for a deep-tech company with a patent, which is why grants sit second rather than fourth. Read the last column first.

Two things this table hides. The first is that the routes stack: revenue makes a loan approvable, a grant makes an angel interested, and a small loan repaid on time makes the next one bigger. The second is that time to money is time you spend, not time you wait — a six-month angel process is six months of your weeks, not six months of leaving something in the post.

Money from a customer is the only funding that arrives without a decision from someone else, costs no ownership, requires no filing, and doubles as proof the business works. It is also the one founders skip past, because raising feels like progress and selling feels like work. Here is the trade in numbers, using the business these pages follow.

Northline needs $2,143 a month more to stop losing money. That is about a dozen more jobs at $186, or a 10% price rise plus ten more jobs, or a smaller cost base and fewer of either. Any of those can start on Monday. The alternative — raising the equivalent of about fourteen months of that gap — takes three to six months of the founder's attention, and at the end of it Northline still has to do the selling anyway, only now with an investor watching.

Three revenue moves that behave like funding, in the order they pay off:

• Raise prices on new work. Every dollar is margin with nothing to deliver against it, it takes an afternoon, and it compounds on every booking after. A 10% rise on a $60 tune-up is $6 that costs nothing.
• Take money earlier. Deposits on booking, payment on completion rather than net-30, a small retainer instead of invoicing in arrears. This doesn't change annual revenue by a cent and can transform the cash position — which is the thing you were about to borrow against.
• Sell more to the customers you already have. A second service, a maintenance plan, a subscription. The acquisition cost is zero and they already trust you. Most small businesses have more revenue sitting in their existing list than in any campaign they could run.

The honest limit of this section: some businesses genuinely cannot sell first. A medical device needs a trial; a chip needs a fab run; a drug needs years. If yours is one of those, skip to grants — that's exactly what SBIR exists for. If yours isn't, and you're reading a fundraising page instead of a pricing page, the gap in the table above is the more useful place to spend the week. The runway calculator shows what each lever does to the months you have left.

Grants: real, specific, and not for opening a shop

The single most useful sentence about small business grants is the SBA's own: “SBA does not provide grants for starting and expanding a business.” Every site promising free government money to open a café is selling you a list or a lead. Federal grant money for businesses is overwhelmingly research money, and it comes through two programmes.

SBIR and STTR are the real ones — non-dilutive, no equity taken, awarded by eleven participating federal agencies for research with a technical risk and a commercial future. As of April 2026 agencies can award up to $323,090 for Phase I and $2,153,927 for Phase II without seeking SBA approval; larger awards need a waiver. To qualify, a company must have no more than 500 employees including affiliates, be more than 50% owned and controlled by US citizens or permanent residents (or by other small businesses that are), have a place of business in the United States, and do the work here. STTR adds a partner: a US non-profit research institution, with the small business performing at least 40% of the work and the institution at least 30%. Open topics are searchable on SBIR.gov.

Everything else federal lives on Grants.gov, and Grants.gov is mostly not for you. It lists what federal agencies fund, and the applicants are largely state governments, universities, tribes and non-profits. It tells people looking for personal or start-up money, in as many words, that it is probably not the site for them.

State and local money is where a small business's odds actually improve, because the pool is smaller and the eligibility is narrower. Two things to know. First, the eligibility box is a gate, not a preference: Colorado's Advanced Industries Early-Stage Capital and Retention Grant awards up to $250,000 (up to $500,000 across multiple industries) — and requires you to be in advanced manufacturing, aerospace, bioscience, electronics, energy, infrastructure engineering or tech, headquartered in Colorado or with half your staff there, past proof-of-concept, and able to put up $2 of non-state money for every $1 of grant. Northline's van fails at the first clause. Second, a great deal of state small-business capital is now the Treasury's State Small Business Credit Initiative — a nearly $10 billion programme whose money reaches businesses only through state, territory and tribal programmes: loan participations, loan guarantees, collateral support, capital access and venture funds. You apply to your state, never to Treasury.

SBA-backed loans: the SBA guarantees, a bank lends

The SBA does not lend to you. It guarantees part of a loan a bank or credit union makes, which is what lets that lender say yes to a business it would otherwise refuse. Three programmes matter to a small business, and their limits are these:

One 2026 change worth knowing: from 4 July 2026 a borrower can hold up to $10 million in combined SBA-backed financing across the 7(a) and 504 programmes, double the previous cumulative cap. The per-programme maximums above did not change.

What the table doesn't say, and every borrower discovers: a small SBA loan will almost always require a personal guarantee, which puts your own assets behind the business's debt and quietly undoes part of the reason you formed an LLC. Read for that before anything else. And a loan payment is new burn from the month it starts — the calculator below shows what that does to your break-even.

If you don't know which lender to approach, the SBA's own Lender Match connects borrowers to participating lenders for free, and a certified community lender — a CDFI — is often the one that will look at a business the banks won't. Where to find both →

Friends and family: the fastest money and the most expensive to get wrong

It's fast because the decision is about you, not about the business. That is also the whole problem: the diligence that would have protected them didn't happen, and the person losing money is at Christmas dinner. Three rules, and they're not negotiable if you want to keep the relationship.

• Only take money the person can lose entirely without their life changing. Say that sentence to them, out loud, before they say yes. If they hesitate, that's the answer.
• Write it down, and pick one thing. A loan with a rate and a repayment date, or equity with a price, or a convertible note that becomes equity later. Not “we'll sort it out”. Ambiguity is what turns a bad outcome into a lost relationship.
• Understand it's a securities offering. Selling equity or a note to anyone, including your uncle, is regulated. Most small rounds are done under Regulation D, and the issuer files a Form D with the SEC within 15 calendar days of the first sale; there is no filing fee, and it's public on EDGAR. State “blue sky” notices may apply too. A lawyer for a couple of hours is the cheapest part of this.

One more thing that surprises people: whether your friend counts as an accredited investor changes the paperwork. The individual thresholds are income over $200,000 for each of the last two years ($300,000 with a spouse or spousal equivalent), or a net worth over $1 million excluding the primary residence — or holding a Series 7, 65 or 82 licence in good standing. Rounds with non-accredited investors carry heavier disclosure obligations, which is one reason a lawyer belongs in the room.

Angels: the route most small businesses should stop pursuing

An angel is an individual putting their own money into an early company, typically $25,000 to $500,000, in exchange for equity or a note that becomes equity. They are real, they are useful, and they are the wrong answer for most of the people reading this. Here is the maths from their side, which is the only side that matters when you're asking.

Angels invest across a portfolio knowing most of it returns nothing. That means every individual cheque has to be capable of returning the whole portfolio — ten, twenty, fifty times the money. A business that can grow to $2m of profit and be sold for four times that is an excellent life and a poor angel investment; the same cheque had to be able to make thirty times. So when an angel passes on a profitable, sensible, well-run local business, they are not saying it's a bad business. They are saying their money can't do its job inside it. Believing otherwise is what costs founders a season.

Who genuinely fits: software or a product that can be sold repeatedly without the founder present; a market measured in billions rather than in neighbourhoods; a reason this company gets there rather than an incumbent. Who doesn't: a bike repair van, a bakery, an agency, a plumbing firm, a shop — no matter how good. If you're the second kind and you want to grow, the money is a loan, and the loan is approvable because you have revenue.

If you are the first kind, the work is not finding names. It's finding the small number of investors whose stage, cheque size, sector and geography actually match yours, and reaching them through someone they already trust. Finding the ones that fit →

Same shape whether the money is a loan, a grant or equity, because all three are asking someone to take a risk on numbers they can check.