- A letter of intent stage is a filter the funder runs, not a document the applicant perfects. Letter of intent, letter of inquiry and concept note all describe the same thing: a short first submission that decides whether a full application is worth anyone's time.
- It is not an eligibility check. Eligibility is binary and belongs in the form's logic. An LOI answers a question no rule can, which is whether this is the kind of project the programme wants to fund.
- Five questions are usually enough, and the one that filters hardest is why this call. An applicant sending the same LOI to twelve funders cannot answer it well.
- The stage only pays if the decline is fast. An LOI that takes six weeks to answer has added to the applicant's wait without removing any of their work.
A letter of intent stage is a short first submission that lets a funder tell an applicant whether a full proposal is worth writing. The applicant sends one or two pages. The funder replies with an invitation or a decline. The point of the stage is to move the cost of a declined application off the applicant and out of the review panel.
Almost everything published about letters of intent is written for the people submitting them. This is the other side of the desk: how to design the stage, what to ask, and how to decide.
Letter of intent, letter of inquiry, concept note
Three names for the same stage. Letter of intent is the most common term in corporate and foundation grantmaking, letter of inquiry appears more often in North American philanthropy, and concept note is standard in research and international development.
The differences are conventional rather than substantive. All three describe a short, non-binding first submission that a funder screens for fit before opening a full application. If a call uses one term and applicants ask about another, they are asking about the same thing.
What matters more than the name is whether the stage has a decision attached to it. A letter of intent that everyone is invited to expand into a full proposal is not a filter. It is an extra form.
What the stage is actually for
An LOI stage answers one question that no eligibility rule can: is this the kind of project this programme wants to fund.
Eligibility is binary and mechanical. An organisation is a registered non-profit or it is not. It operates in the funded region or it does not. Those conditions belong in the form's logic, applied on submission, and they need no human at all.
Fit is different. A request can satisfy every published condition and still be wrong for the call: the right kind of organisation proposing the wrong kind of work, or work the programme funded three years ago and has since moved on from. Judging that takes a person, and it takes a few minutes when the submission is two pages rather than forty.
That ratio is the entire economic case for the stage.
When an LOI stage is worth running
It pays when three conditions hold together.
The full application is expensive to write. If a complete proposal takes an applicant a day, the stage adds friction without removing much. If it takes weeks, as research and capital proposals often do, the stage saves real work.
You decline most of what you receive. A programme funding one application in three gains little. A programme funding one in twenty is asking nineteen organisations to do avoidable work.
You can answer quickly. This is the condition most often missed. An LOI that takes six weeks to answer has added a stage to the applicant's wait without removing any of their effort. If the panel cannot turn LOIs around in about two weeks, the stage makes the process worse rather than better.
It is not worth running for a closed programme where applicants are invited directly, or where the pool is small enough that reading every full application is manageable.
The five questions an LOI should ask
Long LOI forms defeat the purpose. Five questions, answerable in about fifteen minutes, are usually enough to decide.
- Who is asking. Organisation, legal status, country of operation, and one line on what it does.
- What they propose to do. A paragraph. Not a methodology, not a workplan.
- Who it is for and at what scale. The population reached and roughly how many.
- Approximate amount requested, and total project cost. A range is fine. The purpose is to see whether the request sits inside your funding band at all.
- Why this call. One paragraph connecting the project to the published priorities. This question does most of the filtering, because it is the one an applicant sending the same LOI to twelve funders cannot answer well.
Everything else can wait for the full application form. Asking for a detailed budget or a complete logic model at LOI stage recreates the cost the stage exists to avoid.
An LOI is not an eligibility check
Conflating the two is the most common design error, and it is expensive in both directions.
Running eligibility criteria as part of the LOI review means a person is reading submissions that a rule should have declined at the point of entry. Running fit as an automated rule means the programme rejects on keywords and loses proposals it would have wanted.
The clean sequence is eligibility first, automatically, before the LOI form even accepts a submission. Then the LOI, read by a person, for fit. Our guide to automated grant application screening sets out the five stages this one belongs to.
How to screen LOIs without rebuilding the full review
The risk at this stage is over-engineering. A weighted scoring rubric with six criteria and defined anchors is the right instrument for full proposals. Applied to a two-page LOI it produces false precision, because there is not enough on the page to separate a 6 from a 7.
Three practices keep it proportionate.
Decide, do not score. Invite, decline, or hold. Three outcomes, one reviewer, with a second opinion only on the holds.
Write one sentence of reasoning per decision. Not for the applicant, for the record. Six months later, when someone asks why a well-known organisation was declined before anyone read a proposal, that sentence is the answer.
Read a batch, not a queue. Fit is comparative. Reviewers assessing LOIs one at a time as they arrive apply a drifting standard. Reviewers reading forty in a sitting apply a consistent one.
Contact us or request a demo to stop wasting time on spreadsheets— and start managing grants with speed and clarity.
What to tell the applicants you decline
An LOI decline is the cheapest good experience a funder can give, and most funders waste it.
The applicant has invested fifteen minutes and is being told no before spending three weeks. That is a favour, and it reads as one if the message names which of the published priorities the project did not match. It reads as a brush-off if it says the programme received many strong applications.
Two lines are enough: a named reason, and whether they are welcome to apply to a future round. Funders who do this get better-targeted LOIs from the same organisations the following year, which is the compounding return on the stage.
Where the LOI stage sits
The stage belongs to pre-award, between the published call for proposals and the full application. It changes the shape of a round rather than adding to it: one large review becomes one fast filter followed by one smaller review.
For the mechanics of the call that precedes it, see our guide to launching a call for applications. For what happens to the invited proposals afterwards, the reviewer panel and its rubric take over, and our note on reducing reviewer time covers the same problem from the panel's side.
How Optimy runs a two-stage application
Optimy applies eligibility rules automatically on submission, with conditions set per applicant location, then places a short questionnaire before the full form so ineligible and out-of-scope requests are filtered in minutes rather than weeks. Invited applicants are opened up to the full form. Declined applicants get an answer immediately rather than at the end of the round.
Both stages sit on the same record, so the LOI, the decision, the reasoning and the full proposal stay together for the whole life of the request. See the grant management software page, or the research grant pages if your rounds run on concept notes. To walk through your own two-stage process, book a walkthrough.
