Forbidden NCOER Language: The “Potential” Rule Every Rater Breaks
“Has potential to be a future First Sergeant.” It reads like a compliment. It’s actually a rater writing in someone else’s lane, and the rule that says so is one sentence in AR 623-3, paragraph 1-40c. Here’s the rule, the phrases that break it, and how to rewrite them so they hold up.
If I had a dollar for every draft bullet I’ve kicked back for talking about the future instead of the rating period, I could retire early. This is the single most common doctrinal error in rater bullets, and it’s the easiest to fix, because the regulation draws the line in plain language.
// 01The rule, sourced
For OERs and NCOERs, rater comments are specific to performance-based assessments. Intermediate raters (OER only) may comment on both performance and potential when writing assessments. Potential comments are reserved for senior raters on both OERs and NCOERs. AR 623-3, ¶1-40c
Read it twice, because everything else in this post follows from it. Rater comments (the Part IV section bullets you sweat over) are performance-based. Performance is what the NCO did during the rating period and what came of it. Potential (what the NCO could do, will do, should someday be) is reserved for senior raters. Not shared. Reserved.
// 02The phrases that get rater bullets bounced
Any construction that points at the future instead of the rating period is out of the rater’s lane. The usual offenders:
- “has potential to…”
- “will become…”
- “capable of…”
- “shows promise…”
- “future First Sergeant” / “future leader of…”
- “ready for” a job they haven’t done yet
- “destined for…”
- “promotable based on potential”
Notice what these all have in common: none of them state a fact from the rating period. Every one of them is a prediction. Predictions are assessments of potential, and ¶1-40c already told you whose job that is.
One clarification before the rewrites, because it trips people up: recommendation language tied to past performance is a different animal. “Promote ahead of peers” sitting next to a measurable accomplishment in the same bullet is a recommendation grounded in demonstrated performance. “Has potential to excel at the next grade” is speculation about the future. The first is defensible; the second is the senior rater’s sentence coming out of the rater’s pen.
// 03Why the lane split exists
The Evaluation Reporting System splits the assessment on purpose. The rater is the person closest to the NCO’s day-to-day performance: they saw the rating period happen, so they document what happened. The senior rater sits a level up with a wider view across more NCOs and more years, so the system gives them the forward-looking call: potential, promotion timing, where this NCO fits against the population they’ve rated across a career.
When a rater writes potential language, two bad things happen at once. The bullet spends words on speculation instead of evidence, and boards weigh evidence. And it collides with the senior rater’s lane: if the SR assesses potential differently than the rater’s freelancing implied, the report argues with itself. Stay in lane and the report reads as one coherent document: the rater proves what happened; the senior rater says what it means for the future.
// 04Before and after: three rewrites
The fix is always the same move: find the demonstrated performance hiding behind the prediction, and write that instead. If the NCO “has potential to be a First Sergeant,” ask: what did they actually do that made you believe that? Write that.
Wrong: potential language in a rater bullet
o has potential to be a future First Sergeant based on demonstrated leadership ability
A prediction, not a performance fact. “Demonstrated leadership ability” is doing all the work and citing none of the evidence.
Right: the performance behind the prediction
o served as acting First Sergeant for 45 days; led 112 Soldiers through a no-notice EDRE with zero late deployers
Past tense, rating-period fact, quantified. A board reads this and reaches the “future 1SG” conclusion on its own, which is exactly how the system is designed to work.
Wrong: the future-job prediction
o will become an outstanding Platoon Sergeant; shows great promise as a senior NCO
Two forbidden constructions in one line (“will become,” “shows promise”), zero facts from the rating period.
Right: the job they already did
o served as acting Platoon Sergeant for two quarters; sustained 100% MEDPROS and zero safety incidents across a 40-Soldier platoon
If they’re ready for the job, odds are they already performed some of it. Document the fill-in time and its measurable outcomes. That IS the case for the next grade, made legally.
Wrong: capability speculation
o capable of leading at the next level; ready for increased responsibility
Nothing here happened. Both clauses describe a hypothetical future NCO, not the one who just finished a rating period.
Right: what the leading actually looked like
o led a 9-Soldier team through 4 field exercises; met every timeline with zero safety incidents over 240 days
Concrete scope, concrete count, concrete duration. Every word is checkable against the rating period.
// 05The senior rater exception
None of this means potential language is banned from the NCOER. It means it has an owner. ¶1-40c reserves potential comments for senior raters, so in the senior rater’s portion of the report, forward-looking language is not just allowed, it’s the assignment. “Promote immediately,” “select for SLC ahead of peers,” a “future First Sergeant” call: that is the senior rater doing their doctrinal job. The error this post exists to kill is that language migrating down into the rater’s Part IV bullets.
If you’re now wondering how to make a performance-only bullet still read strong enough to earn the box check you’re after, that’s a tier question. Start with the four NCOER tiers in plain English, and then the Exceeded vs Far Exceeded discriminator.
// 06The test to run on every draft
Before a bullet leaves your desk, ask one question of it: did every clause already happen? Past tense, inside the rating period, checkable. If any part of the bullet describes what the NCO might do, could do, or will someday be, it’s out of lane. Find the performance behind the prediction and write that. It’s always there, and it’s always stronger.
BulletForge enforces this rule automatically: the generator is hard-blocked from producing potential language in rater section bullets, per AR 623-3 ¶1-40c, and it grounds every bullet in the facts you actually give it.
Performance-based, past tense, by the paragraph.
Open BulletForge