Grace works your portfolio overnight so you walk in each morning to a ranked list of donors who need you today, every name backed by full relationship context.
150+ active relationships and no clear way to know who's at risk of going quiet this week.
Cultivations stall when a thread drops. By the time you notice, the moment has passed.
Hours every week pulling history, notes, and giving records together before a single call.
Grace is an AI teammate, not another dashboard. It works the parts of your job you'd hand off if you had the headcount.
Every morning, Grace ranks your portfolio by who needs attention - lapsing supporters, warm prospects, milestone moments - so you start the day knowing exactly who to call.
Ask Grace for a briefing before any meeting. You get giving history, notes from every team member, family connections, and program engagement - every claim cited.
Thank-yous, check-ins, and cultivation emails that reference real donor details. Not templates - drafts you'd be proud to send.
Every relationship, every note, every cultivation in progress stays in Grace's memory and transfers cleanly when team members change.
"I used to spend Sunday nights pulling reports for Monday calls. Now I open Grace and the work is already done."
A 30-minute walkthrough, tuned to your role and your CRM.
4-week free trial · No credit card · No commitment
How can a small development team keep up with donors? Not by working more donors. By lowering the time each donor costs. A one-person shop with eight hours a week for donor strategy can actively work about sixteen relationships. Cutting per-donor prep from thirty minutes to ten raises that to forty-eight, without adding an hour or a hire.
The Fundraising Effectiveness Project is a collaboration between the AFP Foundation for Philanthropy and GivingTuesday, and it publishes the only quarterly aggregate view of US giving. Its Q4 2025 report, released 21 April 2026, found this:
| Measure | 2025 vs 2024 |
|---|---|
| Total charitable dollars raised | +5.0% (±0.5%), the strongest growth in five years |
| Number of donors | -3.6% (±0.5%), the fifth consecutive annual decline |
| Overall donor retention | 43.1% to 43.3% |
| New donor retention | Essentially flat |
| Where the growth came from | Almost entirely Major and Supersize donors |
Read those rows together. The sector raised more money from fewer people for the fifth year running. Revenue is up because large gifts got larger, not because more people gave.
Woodrow Rosenbaum, chief data officer at GivingTuesday, put it plainly in the release: the sector is "serving fewer donors for the fifth consecutive year", with growth "concentrated among a shrinking number of large donors".
The prescription from AFP is the part that matters for a small shop. Ann Hale, executive vice president of the AFP Foundations for Philanthropy, said long-term sustainability depends on "engaging and retaining both mid-level and small donors at scale".
At scale. That is the instruction. Now hold it next to what one person can actually do.
Caseload is the number of donors one fundraiser can actively work, as opposed to the number sitting in the database with their name on it. The distinction is the entire problem. Most small shops report the second number and are quietly judged against the first.
The most useful formula in the field comes from David Lively of Northwestern University, published in Managing Major Gift Fundraisers: A Contrarian's Guide (CASE, 2017):
Caseload size = time available ÷ time required for strategy development per donor
Lively's own worked example: twenty hours a week for strategy at thirty minutes per donor gives a caseload of forty. Note what the formula excludes. This is planning time only. It does not include the calls, the meetings, the emails, or the writing.
Nobody publishes what that formula returns for a development director who is also running events, writing grants, managing the database and producing the newsletter. So here it is:
| Hours a week you genuinely have for donor strategy | At 30 minutes per donor | Donors you can actively work |
|---|---|---|
| 20 (a full-time major gift officer) | 40 | 40 |
| 8 | 16 | 16 |
| 4 | 8 | 8 |
| 2 | 4 | 4 |
Against a file of 1,200 donors, eight hours a week means you are actively working 1.3% of your own database. Everyone else gets whatever the mass email does for them.
That is not a discipline failure and it is not a prioritisation failure. It is arithmetic. And it is why "engage the broad base at scale" and "one person with eight hours" cannot both be true unless something in the equation changes.
Two kinds of advice dominate this question. Both are internally sound. Both assume something a small shop does not have.
| The advice | What it assumes | Why it breaks here |
|---|---|---|
| "Cut your portfolio to 40 to 150 donors" | You have a major gifts programme and someone else covers the rest of the file | There is no one else. Cutting the list does not reassign it, it abandons it |
| "Thank promptly, personalise, celebrate milestones" | The volume is small enough to do by hand | Sound practice, no capacity model. It does not survive contact with 1,200 records |
| "Keep good notes in your CRM so relationships survive turnover" | Somebody has time to write them, and later, to read them | Writing them is unpaid overhead. Reading 1,200 donors' notes before calls is impossible |
| "Lean on your board and your executive director" | Board members will make cultivation calls | Real and worth doing, but it adds coordination work to the person already at capacity |
The third row is the one worth sitting with. The sector agrees that institutional memory is the answer. Its delivery mechanism is a human typing into a notes field, and then a human reading it back. That is the assumption worth attacking, not the advice itself.
We wrote about the turnover half of this problem separately in the institutional memory crisis.
Three jobs consume a small development week, and only one of them is the job.
Preparation. Pulling giving history, past notes, event attendance and email threads together before a single conversation. This is the cost that sets the denominator in Lively's formula, and it is the one nobody measures. We put numbers on the data-entry portion of it in the ROI of 15 hours.
Triage. Deciding who to call this week. In practice most small shops substitute a proxy for this: the largest gifts, the most recent gifts, or whoever happens to surface. None of those find the donor who is quietly on their way out. Knowing which donors are at risk of lapsing is a different job from ranking them by gift size.
Reconstruction. Rebuilding context that already existed but left with someone, or was never written down. This is the tax that turns a two-year tenure into a permanent restart.
Only the calls, the visits and the asks raise money. The other two are the overhead that decides how few of those you get to make.
This is tool-agnostic. It works on a spreadsheet. It exists because "prioritise better" is not an instruction.
Split the file by behaviour rather than by gift size. Gift size tells you what someone gave once. Recency tells you whether you still have a relationship.
| Group | Definition | What it actually is | Priority for a small caseload |
|---|---|---|---|
| Current | Gave in the last 12 months | An active relationship | Low. Already engaged |
| Closing | Gave 13 to 24 months ago | A relationship still open, quietly ending | Highest. This is the leak |
| Single-gift | Gave once, never again | The sector's largest and least worked group | Medium, in batches not one by one |
| Lapsed | Nothing beyond two years | Reacquisition, not retention | Low. Costs the most per dollar |
If you want the version of this that survives a staff change, the donor handoff checklist covers what has to be written down and when.
Look at Lively's formula again. It has two variables, and the entire category optimises only one of them.
Caseload size = time available ÷ time required per donor
Every piece of advice above works on the numerator. Get more hours, or accept fewer donors. Almost nobody works on the denominator, which is strange, because it is the one a small shop can actually move.
| If prep per donor falls to | 8 hours a week gets you |
|---|---|
| 30 minutes (typical, by hand) | 16 donors |
| 20 minutes | 24 donors |
| 10 minutes | 48 donors |
Same person. Same eight hours. Three times the caseload, because the cost per donor changed rather than the effort.
This is the layer Gratefully works on, and it is worth being precise about what that means. Grace does not replace your CRM and it is not a wealth screener. It reads what your organisation already holds, the CRM records, the inbox, the notes and the documents, and turns it into something you can ask a question of.
What that looks like in practice. The prep step in the triage system above is the one that collapses. Instead of assembling context before a call, you ask for a briefing and get giving history, notes from every team member, family connections and programme engagement, with every claim cited to where it came from. The ranking step in point four stops being a manual sort, because the Action Center surfaces who is drifting before the pattern is obvious. And the part of your file you had written off gets a second look through hidden revenue discovery, which is where the mid-level donor who has given $250 a year for six years and has never once been asked for more tends to surface.
That last one connects directly back to the FEP finding at the top of this section. If sector growth is coming from larger gifts while the donor base shrinks, the mid-level donors already in your file are the most valuable and most ignored asset you have.
Almost every development director I speak to apologises at some point in the conversation. For the donors they never called back. For the thank-you that went out three weeks late. For the plan they wrote in January and have not opened since. They describe it as a personal failing, and it lands as one. It is not. It is the arithmetic above, and it does not improve by caring more about it. That is the part worth saying plainly, because the people carrying this are usually the last to hear it.
None of the above is happening in a well-resourced sector.
The Center for Effective Philanthropy's State of Nonprofits 2025, surveying 583 nonprofit leaders, found that nearly two thirds reported difficulty filling staff vacancies in the past year, and that more than half said their biggest staffing challenges came from insufficient funding to recruit, retain and support people. Nearly all leaders expressed concern about their own burnout and reported it affecting their staff. Their own figures on the question, from the report:
| Level of difficulty filling staff vacancies | Share of leaders |
|---|---|
| No difficulty | 36% |
| A little difficulty | 28% |
| Some difficulty | 22% |
| A lot of difficulty | 14% |
Now put that next to the FEP numbers above. The organisations least able to hire are the ones being asked to engage a broad donor base at scale, in a year when the sector's growth came from a shrinking number of large gifts. That is the context for every capacity decision here, and it is why the answer has to be something other than working harder.
Using David Lively's formula, caseload equals time available divided by time required per donor. At thirty minutes of strategy time per donor, twenty hours a week supports about forty donors, eight hours supports about sixteen, and four hours supports about eight. The commonly quoted figures of 150 for major gifts and 300 to 1,000 for mid-level assume a full-time gift officer who does nothing else.
The Fundraising Effectiveness Project's Q4 2025 report put overall donor retention at 43.3%, up slightly from 43.1%. New donor retention remained essentially flat and is substantially lower. Converting a first gift into a second is described in that report as the sector's most consequential unsolved problem.
Retention, and specifically the donors who gave 13 to 24 months ago. They are the group where a relationship still exists and is closing. Acquisition replaces a lapsed donor at far higher cost, and FEP data shows donor counts have declined for five consecutive years, so the acquisition pool itself is shrinking.
Only if the constraint is knowing who has capacity. If the constraint is that nobody has time to work the list you already have, a screening file adds names to a queue that is already unworked. We covered how to tell the difference in our guide to DonorSearch alternatives.
By making the record independent of the person. That means context captured as work happens rather than written up afterwards, and retrievable by whoever arrives next. The donor handoff checklist sets out the minimum, and the institutional memory crisis covers why turnover costs retention.
It helps only where it lowers the time each donor costs. Tools that add a dashboard to check add work. Tools that remove the preparation step before a conversation change the arithmetic in Lively's formula, which is the only variable a small team can realistically move.