How to Compare Construction Bids Without Getting It Wrong
Bid leveling is the discipline of correcting for that. It is the single highest return activity in construction procurement, and it is routinely skipped because it looks like administrative work rather than value creation.
Why the bottom line number is meaningless
A construction bid is a set of assumptions expressed as a price. When those assumptions differ, the prices are not comparable, no matter how carefully the drawings were prepared.
Source of variance | Typical impact on total |
Differing interpretation of scope | 10 to 25 percent |
Material grade and specification | 5 to 20 percent |
Excluded preliminaries and site costs | 8 to 15 percent |
Quantity takeoff discrepancies | 5 to 12 percent |
Contingency treatment | 3 to 10 percent |
Subcontractor coverage gaps | 5 to 15 percent |
Overhead and margin | 8 to 20 percent |
Only the final row represents genuine commercial competition. Everything above it is noise that must be removed before the numbers mean anything.
The bid leveling method
Step one: build the master scope register
Compile every line item that appears in any bid received into a single list. This combined list defines the true full scope of the project, because between them the bidders have usually identified everything the work requires.
Step two: build a comparison matrix
Line items become rows, bidders become columns. Empty cells are the output of the exercise. Each blank identifies work that one bidder has excluded and the others have priced.
Step three: reconcile quantities
Where two bidders state materially different quantities for the same element, at least one is wrong. Resolve the discrepancy against the drawings before pricing discussions begin. An inflated takeoff inflates every dependent line item proportionally.
Step four: normalize specifications
Compare unit rates only where the specified product grade is equivalent. Two line items with the same description and a threefold price difference almost always reflect different products rather than different margins.
Step five: calculate the adjusted total
Add the estimated cost of every omitted item back to each bid. This adjusted figure is the only number that supports a decision, and it frequently reverses the apparent ranking.
Step six: assess delivery risk
Price is one dimension. A bidder with the lowest adjusted total but no comparable project history, a stretched schedule, and thin working capital may still be the more expensive choice once delays and rework are counted.
Line items most often excluded
These are the items that vanish from aggressive bids and return later as variations.
Line item | Typical value | Omission frequency |
Waste handling and disposal | 1 to 3 percent of contract | Very high |
Temporary works, hoarding, protection | 1 to 4 percent | High |
Permits, fees and statutory approvals | 1 to 4 percent | High |
Scaffolding and access equipment | 2 to 6 percent | High |
Design coordination and shop drawings | 1 to 3 percent | Medium |
Commissioning and certification | 0.5 to 2 percent | Medium |
Builders work in connection with services | 2 to 5 percent | Medium |
Final clean and handover documentation | 0.5 to 1.5 percent | Medium |
Winter working provisions | 1 to 3 percent | Seasonal |
The cumulative value of these exclusions typically lands between twelve and twenty percent of contract value, which is almost exactly the gap that separates the lowest bid from the median in most tender returns.
Qualifying the bidder, not just the bid
A leveled price tells you what the work should cost. It does not tell you whether the firm can deliver it.
Qualification check | What it reveals |
Licensing and registration status | Legal authority to perform and certify the work |
Insurance coverage and limits | Whether risk actually transfers under the contract |
Financial statements and payment history | Capacity to fund the work before milestone payments |
Comparable completed projects | Whether the experience is genuinely transferable |
Current workload and committed resources | Whether your schedule is realistic for them |
Subcontractor relationships | Stability of the supply chain behind the price |
Licensing verification is jurisdiction specific and easy to underestimate. In heavily regulated markets, credentials are matched to categories of work rather than to firms in general, and a contractor authorized for one category may be unable to certify another. Public directories such as listings of general contractors in New York show license class and active status side by side, which makes the check a matter of minutes rather than a request that has to travel through the contractor.
Fixing the problem upstream
Most bid leveling difficulty is created before a single bid arrives, by issuing an inconsistent brief. When each bidder receives a slightly different description of the work, divergent assumptions are guaranteed.
Four practices remove most of the variance at source:
1. Issue one written scope document to every bidder, without exception
2. Provide a pricing schedule with fixed line items that bidders must complete
3. Specify materials by grade or performance standard rather than by brand alone
4. Require exclusions to be listed explicitly rather than left implicit
The pricing schedule is the highest leverage of the four. When every bidder returns the same rows in the same order, leveling collapses from a multi day exercise into a spreadsheet comparison. This is the operating principle behind using a construction procurement platform instead of email tendering: the scope is defined once, every bidder responds against the same structure, and the returns arrive already comparable.
How many bids to seek
Number of bids | Outcome |
One | No market reference, price cannot be validated |
Two | Direction only, no reliable median |
Three to five | Optimal, market level becomes visible |
Six or more | Diminishing returns, tender period extends |
Beyond five, additional bids rarely change the decision, and long bidder lists reduce the effort serious contractors are willing to invest in pricing carefully.
Frequently asked questions
What is bid leveling in construction?
Bid leveling is the process of adjusting bids so they cover identical scope, quantities and specifications before comparing them. Without it, price differences reflect differing assumptions rather than genuine commercial competition.
Should I always select the lowest bid?
Only after adjustment. Once omitted items are added back and specifications are normalized, the lowest bid often ceases to be the lowest. Delivery capability, financial stability and schedule realism should then be weighed alongside the adjusted price.
How much can structured procurement save?
Organizations that move from informal quoting to structured competitive tendering with a fixed pricing schedule typically report savings of fifteen to thirty percent, driven mainly by eliminating post award variations rather than by cutting margins.
What if a bid is dramatically lower than the others?
Treat it as an error until proven otherwise. Request a written confirmation of scope coverage and a breakdown of the divergent line items. Genuine efficiency exists, but an outlier of thirty percent or more usually indicates a missed section of work.
How long should a tender period be?
Two to four weeks for a mid sized project. Compressing the period below two weeks tends to produce padded pricing, because bidders add contingency to cover what they did not have time to verify.
Are fixed price or measured contracts better?
Fixed price transfers quantity risk to the contractor and carries a premium of five to ten percent. Measured contracts settle against actual quantities and can be cheaper where the takeoff is reliable. Fixed price suits refurbishment with unknown conditions; measured contracts suit well documented new build.