The set arrives as a pile of drawings, a spec, an RFP, and four addenda that arrived at 11 p.m. the night before. Your job is to turn that into a number that doesn't leave money on the table or a gap you have to eat later. The fastest way to lose on a bid package is not missing the big scope — it's missing the small assumptions that turn a won job into a break-even one.
This guide is the checklist of what to look for in a bid package, in the order that costs you the least, with the three questions every estimator should be able to answer about a set before pricing it: What's in? What's out? What are we assuming?
The three questions
1. What's in — the actual scope
Before anything else: does the set tell you, unambiguously, what you're being asked to build, install, or provide?
- The drawings tell a story the specs contradict. If the plan shows a system the spec text doesn't permit (or vice versa), that's a scope conflict, not a judgment call.
- The RFP says one thing, the drawings another. The invitation may promise a service the package doesn't actually spec, or omit one the drawings clearly call for.
- The 90% and 100% sets differ. Always check what changed between the issued-for-bid set and any earlier version you were given; the change is usually the part that bites.
2. What's out — exclusions and assumptions
Every bid has excluded items. The ones that sink you are the exclusions you assumed were included, or the includes you priced without writing down.
- Explicit exclusions. The set may carve out testing, mobilization beyond a radius, site cleanup, or owner-furnished items. Read the exclusion list like a contract — because that's what your bid will be held to.
- Ambiguous scope. Language like "as required," "per code," or "coordinate with others" passes responsibility to whoever is more specific — you, at cost.
- Who carries the interfaces. Slab cutting and patching, MEP rerouting, coordination with an existing crew — these are classically silent in the bid form and classically expensive. If the package is silent on who restores the slab, that's an RFI, not a guess.
3. What we're assuming — the things we have to write down
The assumptions you don't write down are how you lose. For each, either resolve it or put it in the scope memo:
- Ground conditions — how old is the geotech? Are the borings still current against structural revisions? What stratum are you assuming the design carries load in?
- Schedule and production — what production rate are you pricing? Does the schedule allow it? Is owner-imposed float already gone?
- Materials and lead times — what did you assume about casing, rebar, or long-lead items? A 4–6 week casing lead against a 90-day window is a procurement risk you price, not absorb.
The three gaps that sink bids
Across every trade, the gaps that turn winning bids into losing jobs cluster into three types:
The scope-gap RFI. A drawing shows a system; the spec doesn't permit it; the RFP is silent. Left unresolved, you price the drawing and eat the spec, or price the spec and eat the drawing. Either way it's a gap. Resolve it before you price.
The exclusions gap. You assumed an item was included because it's implied by the scope you bid. The contract reads it as excluded. The only defense is a scope memo that lists every exclusion before the GC signs you.
The ambiguity gap. "Coordinate," "per code," "as shown" — responsibility lands on whoever's most specific. The estimator who prices the ambiguity gets stuck with the interpretation at cost.
The Canadian Construction Association's go/no-go guidance is explicit that most of these answers come from a formal RFI or a simple phone call to the GC — you do not need to price everything to know whether a job fits.[1] CMAA's go/no-go frame (capacity, capital, risk, return) applies the same discipline at the package-review stage: screen the scope gaps before you commit hours to takeoff.[2]
What auto-flagging a bid package looks like
BidCertain reads the bid package — drawings, specs, and RFP — and produces a structured analysis with the scope breakdown, exclusions, risk flags, and auto-drafted RFIs, every claim cited to source. The point is not that it finds gaps you wouldn't; it's that it finds them in the order that costs you the least, before you've spent a week on takeoff for a set that should have been screened in an afternoon.
The Phoenix Convention Center sample (illustrative, in our library) shows a set where the gaps aren't in the drawings — they're in what the drawings and the geotech don't tell you: the pile table has empty "length below cut-off" cells, and the existing subsurface condition conflicts with what the structural set assumes.[3]
Those are exactly the two RFIs a sharp estimator sends before bid day: "Who restores the slab at each pile location?" (bid form silent, $30–60K swing) and "Are the 2019 borings still current after the structural revisions?" The output doesn't answer them; it surfaces them fast enough for you to send the RFI and get an answer while there's still time to price.
Run the check in the right order
1. Scope — what's in? Compare drawings, specs, RFP for conflicts before anything else.
2. Exclusions — what's out? Read exclusions and ambiguity like a contract; list your assumed-includes in the scope memo.
3. Assumptions — what are we assuming? Ground conditions, schedule/production, materials/lead time — each either resolved or written down.
4. Screen the gaps. Resolve the scope-gap, exclusions-gap, and ambiguity-gap RFIs before pricing.
If you're staring at a set too big to read and screen all four with the time you actually have, that's the moment a structured citation pass earns its keep — which is worth seeing on a real package before you buy a seat.
Sources
[1] CCA — "Bid Go or No Go" bulletin. RFI/phone-call-first discipline for bid screening.
[2] CMAA — "Navigating Construction Risks With the Go-No-Go Process." Capacity/capital/risk/return gates.
[3] Phoenix Convention Center sample — BidCertain sample library (illustrative).