How to Find Major Gift Prospects in Your Own Database

Most of your future major donors have already given to you. The fastest way to find them is not a wealth screening of strangers but a ranking of the people in your own database, on four signals you can pull from any CRM: how recently they gave, how consistently, how large their biggest gift was relative to your size, and whether their giving is rising. This page sets out a points method that does that in an afternoon, gives you a size-appropriate definition of “major” from IRS data, and then compares the tools that will do the ranking for you.

How big is a major gift for an organization your size?

There is no standard figure, and the commonly quoted ones, $1,000, $10,000, $25,000, are meaningless without your budget beside them. The IRS, as it happens, draws a size-relative line of its own. To pass the public support test, a charity may count each donor’s giving only up to 2% of its total support over five years; anything above that is set aside on Schedule A. A donor whose giving crosses that line is, by the IRS’s own arithmetic, unusually large for that organization.

Computed from 119,498 charities that sit the test, here is that line in dollars for the median charity of each size, and the same thing as an annual gift.

For a charity spending $250,000 to $1m, one donor giving about $7,000 a year crosses the line, and 37.1% of charities that size have at least one donor who does. Under $250,000 the figure is about $2,300 a year; at $1m to $5m, about $29,000. That is a defensible starting definition of a major gift: a single donor’s giving at or near the point where the IRS starts to treat them as large. If your own threshold is far below the line for your size, you are calling ordinary gifts major; far above it, you are ignoring donors who already matter.

The figures come from the IRS Statistics of Income Form 990 extract; the method is the same as our donor concentration analysis, which shows how much of the sector’s support sits above that line.

The four signals, and why giving history beats wealth

Recency. A donor who gave this year is many times more likely to give again than one who last gave three years ago. Recency is the single strongest predictor in most donor files, and it costs nothing to measure.

Consistency. Five years of giving, at any level, is evidence of a relationship. A one off gift of the same size is evidence of an event. Major gifts come overwhelmingly from the first group.

Largest gift relative to your threshold. Someone who once gave a quarter of your major gift threshold has shown both capacity and willingness at a level most of your file never reaches. That first larger gift is the clearest signal in the data.

Direction. A donor whose annual total has risen is moving towards you. Two consecutive increases matter more than one large year.

These are the same ingredients as RFM analysis, adjusted for major gift work: consistency replaces raw frequency, because ten small gifts in a year say less about a major gift than five years of steady ones, and the size measure is set against your own threshold rather than the file’s average. What none of them measures is wealth, and that is deliberate. Capacity without inclination produces a list of rich strangers, which is the most common way a wealth screening ends up unused.

A points method you can run this week

Export every gift for the last seven years with the donor’s ID, the date and the amount. Every CRM can produce that file, whether it is Little Green Light, Bloomerang, Salesforce or Raiser’s Edge. Summarise it to one row per donor in a spreadsheet, then score each donor on the table below.

The maximum is 15. In most files, donors scoring 11 or more are a short list, often a few dozen even in a file of thousands, and they are your first major gift prospects. Scores of 8 to 10 are the next tier, and the natural population for a mid-level giving programme. Below that, keep them in the annual fund.

Three adjustments make the list better. Remove anyone you already speak to regularly about a major gift; they are in a portfolio, not a prospect list. Read every name in the top tier and strike the ones you know have moved, died or asked not to be contacted, because the export does not. And add the people the data misses: a board member’s close friend who has given little but asked to visit, or a long serving volunteer. The points are a starting order, not a verdict.

Turning the list into prospects

A ranked list is not a portfolio. Each name in the top tier needs a short written profile, a reason to make contact and an owner. The donor profile template covers what to record, and before any first meeting a one page donor briefing brings together what you know.

Then decide how many you can actually work. A full time gift officer can manage a portfolio in the low hundreds at most, and far fewer at the start; the evidence is in how many prospects a gift officer can manage. Most small organizations should start with the top fifteen to twenty-five and move each one through a defined cycle, which is what moves management is, with a free tracker to run it in.

Capacity research comes last, and only for the names you intend to approach. A quick look at public records, a foundation they run, a business they own, property, tells you whether to ask for $5,000 or $50,000. A wealth screening of the whole file before this point mostly produces work; the comparison of those tools, with prices, is in prospect research tools.

Tools that do the ranking for you

The spreadsheet method works, and its weakness is that it is a snapshot: the list is right on the day you build it and drifts afterwards. Several tools keep it current. Here is the honest range, with published prices where they exist.

The choice is mostly decided by the system you already run. If you are on Bloomerang, start with its own Predictive Donor Insights before buying anything else, because it reads the same data with no extra connection.

If you are on Little Green Light, which has no AI features of its own, or on Salesforce for Nonprofits, Gratefully is the most reachable way to keep the ranking current. It connects to those systems directly, plus Mailchimp, Constant Contact and HubSpot for email engagement, and produces a ranked list each morning with the reason each donor is on it, such as a lapsed pledge, a rising trajectory or a stewardship moment due. Its free plan covers one person with five weekly tasks and ten AI answers a month, which is enough to test whether it ranks your file the way you would; Essential is $79 a month billed annually for one person with unlimited answers. It is wrong for three situations: a CRM with only a year or two of patchy history, where it will rank confidently on thin data; anyone who needs capacity estimates, which it does not produce; and Raiser’s Edge NXT users, who can only connect by file import. How it compares with the enterprise option is in Gratefully vs Dataro.

Whatever does the ranking, check its top twenty against your own judgement before acting on it. A tool that puts the right people near the top on the donors you know has earned some trust on the ones you do not.

The mistakes that waste a prospect list

Starting with wealth. Screening the file first and sorting by capacity produces a list of people with money and no relationship to you, and gift officers sensibly ignore it.

Treating the list as permanent. Donors move, give elsewhere, retire. Re-score at least twice a year, or let a tool do it continuously.

No owner per name. A prospect nobody is responsible for is not a prospect. Every name in the top tier needs one person who will make the next contact and a date by which they will.

Ignoring the middle. The donors scoring 8 to 10 are where next year’s major donors come from. An organization that only works the top tier runs out of prospects in three years.

This is reference information, not legal or tax advice. Rules vary by state and change over time. For a decision that carries real consequences, check the current text at irs.gov or your state registry, and talk to a nonprofit attorney or CPA.