This guide explains how to spot each type, why federal documentation uses different terminology than most SMBs do, and what the actual evaluation looks like behind the scenes.
The 30-second answer
If you only have 30 seconds before bid deadline, here's the rule:
Mandatory criteria use words like "must," "shall," and "will." They are pass/fail. Miss one, and your bid is non-compliant and gets no further evaluation, regardless of price or quality.
Desirable or rated criteria use words like "should," "preferred," "additional points awarded for." They are scored on a points scale. You can score zero on one and still win the contract if your overall score is competitive.
Almost every federal RFP uses both, in two distinct sections of the bid document. Mandatory criteria are evaluated first. Bids that fail any mandatory are eliminated. Surviving bids are then scored against the rated criteria.
What "mandatory" actually means in a federal RFP
Mandatory criteria are the minimum requirements a bidder must meet to be considered for the work at all. Treasury Board guidance is unambiguous on this: mandatory criteria are evaluated on a pass/fail basis, and bids that fail to meet them are given no further consideration.
Three things follow from that:
Mandatories cannot be partially met. If the requirement is "the proposed resource must hold a PMP certification," a candidate with PRINCE2 doesn't meet the mandatory, even if PRINCE2 is arguably equivalent. The evaluation is strictly objective. Either you have a current PMP certification on file, or you don't.
Mandatories cannot be relaxed after bid close. PSPC's evaluation template is explicit: mandatory requirements "must never be changed or relaxed in any way after bid closing." Even if a contracting authority later realizes a mandatory was unreasonably restrictive, they cannot waive it for compliant or non-compliant bidders. The fix has to come through a formal amendment before the bid closes.
Failing one mandatory ends the evaluation of your bid. Per PSPC's assessor guidance, if a bid fails the first mandatory criterion checked, evaluators do not need to assess the remaining mandatories or score the rated criteria. The bid is non-compliant and gone.
This stringency is by design. The rationale is that mandatories give every bidder a clear, identical bar to clear, and the government a defensible record if a losing bidder challenges the award through the Canadian International Trade Tribunal (CITT). Loose mandatory enforcement is one of the fastest ways to lose a CITT challenge.
What "desirable" or "point-rated" criteria actually mean
Desirable criteria, almost always called "point-rated" or "rated" in federal documentation, are how the government distinguishes between bidders who all clear the mandatory bar.
A typical rated criterion looks like this:
- 0 points: no demonstrated experience
- 5 points: 1-2 projects demonstrated
- 10 points: 3 or more projects demonstrated
The bidder is not disqualified for scoring 0 on R3. They simply collect zero points for that criterion and continue to be evaluated on the rest.
Most federal RFPs set a minimum technical score on the rated section, typically 70% of available points. Bids that fall below this threshold are eliminated even if they passed all mandatories. After the technical evaluation is complete, the surviving bids are evaluated on price, and the winner is selected using one of several formulas: lowest price among compliant bids, highest combined technical-and-price score, or a weighted blend (often 70% technical / 30% price for complex work, or 30% technical / 70% price for commoditized work).
The basis of selection is always disclosed in the RFP. Read it before you decide how much effort to put into rated criteria. If the procurement is 80% price, hammering optional rated criteria past the 70% threshold may cost more in bid effort than it returns in evaluation points.
Why the terminology trips people up
This is one of those areas where federal documentation and SMB shorthand never quite match.
What contractors usually say: "mandatory" and "desirable." What federal RFPs usually say: "mandatory" and "point-rated" (or "rated," or "technical scored"). What you'll occasionally see: "essential" vs. "non-essential," "must-haves" vs. "value-adds," or in older documents, "weighted criteria."
They all mean the same two things. But because the federal documentation rarely uses the word "desirable," SMBs reading their first RFP often miss that the rated section is where the desirables live. They look for a section called "Desirable Criteria," don't find one, and either skip the rated section or treat the entire RFP as one big pass/fail document.
The other common confusion: "shall" appearing in rated criteria. Some rated criteria are written with imperative language out of bad habit ("the bidder shall demonstrate experience in..."). The location of the criterion (Mandatory section vs. Rated section), not the verb, is what determines whether it's pass/fail. When in doubt, look at the table of contents. If the criterion sits under a heading like "Point-Rated Technical Criteria" or "R1, R2, R3...," it's rated, regardless of how the sentence is phrased.
How federal evaluators actually score a bid
Knowing how the inside of the evaluation looks changes how you write the outside.
For a typical RFP, an evaluation committee of at least three people independently reviews each bid. They follow a strict order:
- Mandatory check. Each evaluator works through the mandatory criteria one by one and marks pass or fail. If any evaluator marks fail on any mandatory, the committee discusses, and if the failure stands, the bid is eliminated.
- Rated scoring. Survivors get scored against each rated criterion individually. Evaluators score independently first, then meet to reconcile differences and agree on a final score per criterion.
- Minimum score check. Bids below the technical threshold (often 70%) are eliminated.
- Financial evaluation. Surviving bids' prices are normalized against the technical scores using whatever formula the RFP specified.
- Recommendation and award. The bid that wins under the basis of selection is recommended for contract award.
Two practical implications for the bidder:
Mandatories are checked literally. The evaluator is looking for a specific piece of evidence, in a specific place, in a specific format. "We've done similar work" doesn't satisfy a mandatory that requires "two project descriptions of at least 500 words each, demonstrating experience with X, including project name, dates, contract value, and client reference." Every element listed in the mandatory must be present. Missing the contract value, or providing 450 words instead of 500, has been enough to disqualify bids in past CITT decisions.
Rated criteria reward specificity. Generic claims score low. A claim like "extensive experience with federal change management" might earn 0 or 5 points. The same claim with three named projects, dates, contract values, and the specific change management methodology used, earns the full 10. Evaluators are scoring what's on the page, not what they assume your firm can do.
Skip the manual mandatory check.
Paste any CanadaBuys URL — BidFit pulls out every mandatory criterion, flags ambiguous wording, and tells you which documents you need to gather, in 30 seconds.
Why mandatories disqualify more SMB bids than anything else
Three patterns account for most SMB disqualifications on mandatories.
The "we basically meet it" trap. A mandatory requires a specific certification, security clearance, ISO standard, or years of experience. The bidder has something close, decides it's "essentially equivalent," and submits. The evaluator marks fail. The bid is gone. There is no "essentially equivalent" in mandatory evaluation.
The "buried in the proposal" trap. A mandatory requires a specific document or form to be submitted with the bid. The bidder submits the document inside the technical proposal, in an appendix, or attached separately. The mandatory specifies the document must be at a specific location in the bid package. Failing to put it in the right place fails the mandatory. This sounds petty, and it is, but it is also enforceable.
The "didn't read every mandatory" trap. A typical RFP has 8 to 25 mandatory criteria. SMBs reading their first or second RFP focus on the obvious ones (qualifications, experience, capacity) and miss the procedural ones (Canadian content certifications, integrity declarations, employment equity attestations, conflict of interest disclosures, former public servant declarations). Any missed certification is a failed mandatory.
The fix for all three is the same: build a mandatory checklist as the first step of every bid, line by line, and confirm each item before the bid is submitted. We've covered this in How to read a CanadaBuys tender notice (with annotated examples), which walks through where mandatories appear in the bid document and how to extract them.
A real example of a mandatory becoming a rated
The Office of the Procurement Ombudsman's 2023 review of PSPC documented a case that shows exactly how high the stakes are.
A Task and Solutions Professional Services (TSPS) RFP for a senior communications consultant included this mandatory:
Mid-solicitation, the contracting authority issued an amendment that deleted M2 entirely and replaced it with a rated criterion:
In one sentence, the requirement went from "no certification, no contract" to "no certification, lose 10 points but still in the running." Bidders who had ruled themselves out of bidding because they couldn't meet M2 had to be able to react to the amendment. Bidders who had already drafted their bids around M2 had to decide whether to revise.
That's why you read every amendment. A change to the mandatory section can completely flip whether you're allowed to bid, and the only way you'll know is by checking. We've written a separate guide on how to track tender amendments without missing a deadline if you want a full system for this.
How to read criteria correctly the first time
A reliable process for any new RFP:
- Open the bid document and find the evaluation section. It's usually Part 4 in PSPC's modernized solicitation format, or Section 4 in legacy templates.
- Identify the two subsections. One will be labeled Mandatory Criteria (sometimes Mandatory Requirements, M1/M2/M3...). The other will be labeled Point-Rated Criteria, Rated Criteria, or Technical Evaluation Criteria (R1/R2/R3...).
- Build a mandatory checklist. Copy every mandatory criterion verbatim into a working document. For each one, note: what evidence is required, what format it must be in, and where it must appear in the bid package.
- Map your evidence. For each mandatory, write down which document, certification, project description, or form proves you meet it. If you can't fill in this column, you have a gap.
- Triage the rated criteria. Sort by point value. The 10-point criteria deserve detailed responses; the 2-point criteria can be addressed briefly. Calculate the minimum score you need to clear the technical threshold and where your easy points are.
- Re-check after every amendment. Amendments can change mandatories into rated, rated into mandatories, or change the point allocation entirely.
If you do this on day one of a 25-day solicitation, you have 24 days to find or build evidence for any gaps you discovered. If you do it on day 23, you don't.
What to do when a criterion is ambiguous
Sometimes the mandatory is genuinely unclear. The classic example, also from the PSPC ombudsman's review: a mandatory required an "undergraduate degree in business, accounting or related field." A bidder proposed a resource with an undergraduate degree in engineering. Is engineering a "related field"? The honest answer is: it depends on the evaluator.
Two options when you encounter ambiguity:
Submit a clarification question to the contracting authority before the question deadline. This is the safest route. Every other bidder will see your question and the answer, but you will all have the same definitive interpretation.
Make the case in your bid. If the question deadline has passed, you can submit the bid with an explicit explanation of why your evidence meets the criterion. This is a gamble. The evaluator may agree, or may rule the criterion was clear and your interpretation was wrong. Document everything in case you need to file a CITT complaint.
What you cannot do is assume. The most expensive sentence in federal procurement is "they probably meant..." They may have meant something different.