Kitchen Ventilation Bids: Clarifications, Alternates and Exceptions

Most kitchen ventilation bids are not lost on price. They are lost on structure: a question asked too late, an alternate priced so it subsidises the base bid, an exception buried in a general note that the buyer only reads after the contract is signed. Clarifications, alternates and exceptions are the three documents that decide how a tender is actually read – and each one obeys a different rule. This piece sets out the discipline for all three, written for the buyer and the bidder at the same time, because both sides lose when the structure is sloppy.

Three documents, three different jobs

The confusion starts because suppliers treat all three as one thing: text sent to the buyer before or with the price. They are not interchangeable, and a tender that mixes them up produces a contract nobody can price a variation against.

The test that separates them is simple. A clarification asks about the buyer’s documents. An alternate changes the bidder’s own offer. An exception reduces the buyer’s documents. If a document claiming to be one of the three is actually doing the job of another, the pricing that follows will be wrong in a way nobody notices until a variation is disputed.

Clarification discipline: what to ask, and when

A good clarification question changes a quantity or a method. A bad one is a leading question designed to shift risk, and buyers read it as exactly that. The questions that earn written answers in kitchen ventilation tenders are factual, and they cluster around the interfaces nobody owns by default:

1. Which party runs the duct route through the ceiling void, and is a route reserved in the reflected ceiling plan?
2. What is the clear height at the hood position, measured, not nominal?
3. Where does make-up air come from, and who provides the balancing?
4. Who supplies the electrical supply point – breaker size, cable route, isolation switch?
5. Is fire suppression integrated into the hood by the supplier or fitted separately by the sprinkler contractor?
6. Does grease drain into a gutter in the hood or to an external drain line, and who builds the connection?
7. What is the acoustic limit at the nearest occupied boundary, and who verifies it?

Each of those questions changes either the airflow calculation, the equipment count, or a work package boundary. That is what justifies the paper. Questions that only express preference – “do you accept delivery in two batches” – belong in the offer, not in the clarification round.

Three procedural rules make the round work. First, ask before the deadline stated in the tender; buyers close the question window seven to ten days before submission for a reason, because an answer that arrives after the pricing is locked is an answer nobody can use. Second, one subject per question – a question that bundles three topics gets a single vague answer covering none of them. Third, never ask about a competitor’s bid; the addendum that answers it discloses your strategy to everyone holding the tender.

The addendum register

Addenda are the only instrument that changes a tender after issue, and they arrive while pricing is already under way. Without a register, the last addendum issued three days before submission is the one nobody priced. The register is a table with six columns, and it should be kept from the day the tender lands:

Two habits close the loop. Every addendum must trigger a re-check that the earlier ones still stand – buyers occasionally rescind an instruction in a later addendum and rely on nobody noticing. And the bid submission should carry an addendum acknowledgement sheet listing every addendum received, signed. Where that sheet is mandatory, a bidder who omits an addendum has bid on superseded documents, and the price comparison against other bidders is invalid.

The rule worth writing into the purchase file, on either side: if a point was raised in a clarification and never answered in an addendum, it is not agreed. A verbal answer at a site visit or a reply to an email outside the question window has no contract value. Both parties are better served by restating the point in a clarification and forcing the addendum.

Alternates: same outcome, different route

An alternate is not a cheaper version of the same equipment. It is a different technical route to the performance the specification demands – a different separation method, a different duct material, a different control package, a different service level. The buyer keeps the option because the specification writer could not predict every route, and the bidder prices each route on its own merits.

The integrity rule comes first: if the bid is only compliant when an alternate is accepted, it is not an alternate, it is a requirement. The base bid must stand alone – complete, compliant, operational – with every alternate priced as a genuinely separable package. A base bid that omits balancing valves because “the alternate package includes them” is a contingent bid, and most evaluation rules treat it as non-compliant rather than cheap.

Alternates must be priced independently. If alternate A is priced high and alternate B low so that the buyer is steered toward B regardless of technical merit, that is cross-subsidy, and the buyer who catches it discounts the whole submission. Each alternate carries its own delta against the base bid, its own delivery time, and its own effect on the warranty and the maintenance schedule – because an alternate that changes the equipment changes the spare parts list too.

Buyers evaluate in their own numbering, not the bidder’s. A submission that renumbers the alternates, merges two of them, or prices one “included at no charge” has broken the comparison grid, and inclusion at no charge is the classic flag: nothing is free, so the cost has moved somewhere – usually into the base bid or into the maintenance contract the buyer has not seen yet.

The pricing structure the buyer should demand

Structure is what makes two bids comparable and a post-award variation priceable. Five lines, each with a defined purpose and a known trap:

Unit rates deserve the emphasis they rarely get. The most expensive number in a kitchen ventilation contract is often not the equipment price but the first variation order, priced by a supplier who quoted no rates. A one-page rate schedule – per metre of duct, per extra hood position, per hour of commissioning, per site visit – costs the bidder nothing to produce at tender stage and removes the only negotiating position a buyer has after award.

Options follow the same logic but in the other direction. An option priced as a bundle (“12-month extended warranty and spare kit”) cannot be compared against another bidder’s split pricing, and cannot be dropped if the budget moves. Every option should carry its own standalone price so the buyer can assemble the estate package line by line.

Substitution requests: the five-part structure

A substitution request asks the buyer to accept a different product than the one named in the specification. Most are rejected – not because the product is worse, but because the request is incomplete. A request that survives review has five parts:

1. The clause reference it replaces, quoted, so the reviewer knows exactly what is being displaced.
2. The proposed substitution, with manufacturer, model designation and the technical data sheet.
3. A side-by-side comparison against the specified item: performance, dimensions, electrical rating, materials, service access, connection points.
4. Evidence: test reports, certification documents, and references from comparable installations.
5. An equivalence statement signed by a qualified engineer – not a sales office – stating that the substitution meets or exceeds the specified performance in the parameters the specification lists.

The rejections cluster around four causes, and they are worth knowing before submitting: a dimensional mismatch that breaks the reflected ceiling plan; a service access difference that changes the maintenance routine; spare parts availability that the buyer cannot verify in its market; and a certification gap against the standard the specification cites. A substitution request that answers those four questions inside the comparison table is reviewed on merit. One that says “equivalent to the specified product” without data is rejected on receipt.

Buyers should publish what their reviewers need and bidders should supply it the first time. The most common cycle wasted on both sides is three rounds of “please provide the technical data sheet”, which is a process failure, not a product failure.

Exceptions: naming what you will not do

An exception list is honest only when it is priced honestly. The discipline is mechanical: every exception names the clause it qualifies, states its position, and quantifies the cost and schedule impact of removing it. Without the “cost to remove” figure, the exception is not a risk statement – it is a discount hidden in plain sight, because the buyer cannot see what compliance would have cost.

Three formats fail review predictably. Conditional language – “subject to site conditions”, “subject to final measurements” – transfers unbounded risk to the buyer and reads as a bid the bidder has not yet decided to stand behind. General notes – “all dimensions to be confirmed on site” – qualify nothing specific and therefore disclaim everything. And the single word “noted” next to a clause number tells the buyer the bidder has read the clause but refuses to say what it will do about it.

Buyers hold the countering lever: cap the exceptions. A tender that allows unlimited exceptions rewards the bidder who writes the longest disclaimer. A tender that states “submissions carrying more than five exceptions, or any exception to the acceptance criteria, will be rejected” forces every bidder to make a real choice between pricing the risk into the offer and accepting the clause. The five-exception cap is arbitrary but functional – it is small enough to force selection and large enough to survive a genuine site constraint.

Traps that survive every evaluation

The same five patterns recur across tenders for kitchen ventilation, and each has a signature the evaluator learns to search for:

• The lowest non-compliant bid. Priced below every rival because it quietly omits balancing, make-up air interlocks, or the grease management package. The signature is a base bid unusually low against the quantity schedule.
• The contingent bid. “Pending confirmation of site measurements” – an offer the bidder has not committed to. Signature: conditional language in the cover letter.
• The steered comparison. Alternates renumbered or merged so the buyer’s evaluation grid no longer matches. Signature: the bidder’s alternate numbering differs from the tender’s.
• The unevidenced “or equal”. A substitution claiming equivalence with no test data. Signature: the word “equivalent” without an attachment.
• The disclaimer by volume. Twenty pages of general notes of which one page contains real exceptions. Signature: exception content in sections titled “standard commercial terms”.

None of these is sophisticated. That is the point: they survive because evaluators run out of time, not because they are hard to find. A checklist applied to every submission catches all five in under an hour.

What the buyer should write into the tender

The structure described above only holds if the tender demands it. Five requirements, none of which costs the buyer anything to include:

1. A clarification window closing seven to ten days before submission, with answers issued to all bidders as numbered addenda.
2. A mandatory addendum acknowledgement sheet, signed with the bid.
3. A fixed alternates schedule format: base bid first, buyer’s numbering, standalone deltas, delivery time and warranty effect per alternate.
4. An exception cap, with acceptance-criteria exceptions stated as automatic grounds for rejection.
5. A mandatory unit rate schedule for variations, one page, priced at tender.

Suppliers benefit from the same five. A tender that demands structure is a tender the supplier can price accurately, which means fewer re-quotes, fewer contested variations, and a lower cost of sale. The bidder who receives a well-structured tender and responds with a well-structured bid is not obeying bureaucracy – they are shortening the distance between offer and contract.

Ten checks before the bid goes out

1. Every clarification asked before the deadline, one subject per question, each changing quantity or method.
2. All addenda registered, re-priced, and cross-checked against each other.
3. Addendum acknowledgement sheet signed and attached.
4. Base bid complete and compliant with no alternate accepted.
5. Each alternate priced standalone, with its own delta, delivery time and warranty effect.
6. No alternate “included at no charge”; no renumbering of the buyer’s schedule.
7. Every substitution request carries the five parts, with an engineer-signed equivalence statement.
8. Each exception names its clause, states its position, and quantifies the cost to remove.
9. No conditional language, no disclaiming general notes, nothing left to “site confirmation”.
10. Unit rate schedule attached – per metre of duct, per hood position, per commissioning hour, per site visit.

The last check is the one that decides the first variation. A bidder who priced unit rates has committed to a baseline; a bidder who deferred them to “mutual agreement” has kept the right to name any number after award. Buyers who enforce the schedule at tender stage are enforcing their own leverage for later.

How the architecture shapes the bid structure

The equipment architecture a bid proposes determines how much of this structure it actually needs. A ductless system that recirculates air after multi-stage filtration carries no duct route, no rooftop penetration and no fire damper schedule, so the clarification round shrinks to a handful of questions – clear height at the hood, power supply point, and the odour performance target. Buyers comparing ductless range hood systems can usually close the question round in one pass, which is one reason ductless tenders move faster than the equivalents.

A ducted system imports the whole air path into the tender. Every ducted ESP purifier proposal raises the interface questions by number: duct route ownership, fan selection and its acoustic duty, damper and fire-suppression integration, and the balancing package. The alternates list grows accordingly – duct material, fan arrangement, control package – and the unit rate schedule matters more, because duct modifications are the most common variation in any fit-out.

Mobile and cart-based installations compress the structure from the other side. A proposal built around a ventless mobile catering cart has few fixed interfaces and no building works, so exceptions nearly vanish and the evaluation turns on the cart specification itself – power, filtration stages, service access, and the transport conditions the unit will actually see. The structural discipline shrinks but does not disappear: the base bid still must stand alone, and the unit rates still decide the first change.

The engineering conclusion is unchanged across all three architectures: a tender is a structure before it is a price. The bids that convert into clean contracts are the ones where clarifications ended in addenda, alternates priced honestly, exceptions quantified, and unit rates on the table before anyone signed. Everything after award is administration; everything before it is structure.