A feasibility phase is supposed to answer a question: does this work well enough to be worth building? You cannot answer it unless you decided, in advance and in writing, what "well enough" means. A proposal that never states a threshold is hard for a panel to score and, later, nearly impossible to declare successful, because success becomes a matter of opinion at exactly the moment you need it to be a matter of record.
What a real success criterion looks like
It is a number, a method, and a condition. Weak: "Phase I will be considered successful if the prototype demonstrates the feasibility of the approach." That is circular. Strong: "Phase I succeeds if the prototype sustains at least 85 percent of rated output after 1,000 thermal cycles between the specified limits, measured on three units against the reference method in Task 4."
Note what the strong version commits to: a threshold, a sample count, a test method and an environment. It also implicitly commits to a failure. If the units come back at 60 percent, you said in advance that this is not feasible as designed, and that is a legitimate, reportable outcome of a research project.
Choosing a number you can defend
The threshold has to come from somewhere outside your own preference, or a reviewer will read it as arbitrary. The usual defensible sources are the performance of the incumbent technology you are trying to beat, a published requirement or standard in your field, a customer's stated minimum, or the point at which the economics of your commercialization plan start to work. State which one you used in a single sentence: "The 85 percent figure is the retention level below which the module would need annual replacement, which the target customer has said is unacceptable."
Set it honestly. A threshold set low enough to be certain of clearing looks like what it is, and reviewers who work in the field will know that your bar is trivially easy. A threshold set implausibly high invites the panel to conclude you will fail. Aim for the level a knowledgeable skeptic would agree actually settles the question.
Why it matters more at the next stage
A Phase II proposal is largely an argument that Phase I answered its question. If Phase I stated a threshold and you met it, that argument is one paragraph and a figure. If Phase I stated nothing, you are left asserting that things went well, and a fresh panel has no way to check. The transition between the phases is covered in Phase I versus Phase II, and the criteria you write here become the backbone of what you report to the agency, as described in reporting and milestones.
Success criteria pair naturally with the go/no-go decision points in your work plan: each criterion is checked at a specific task, and each check has a defined consequence. That structure is also what makes a proposal legible to the panel, as covered in how panels score. If you have never set thresholds like these before, the exercise resembles a feasibility study, done with engineering rather than market numbers.
Projects House helps founders set thresholds a reviewer will accept and a test rig can actually measure. Start through the contact form.