Why One Bid Spreadsheet Breaks at Three Active Projects
The first time a sub bills you off a scope you already cut, you learn what version drift costs. I once watched a $14,000 line item get paid twice because two copies of the same bid tab disagreed about who carried the exterior doors. Lock your bids in one dated place before the fourth project lands, not after the second invoice clears.
One bid spreadsheet works fine when you are running a single job. It starts lying to you somewhere around the third active project, and the lie is quiet. Nobody gets an error message. You just start losing money in ways that take a month to surface.
Why does one bid spreadsheet stop working after three active projects?
A spreadsheet stops working because it has no concept of state. It cannot tell you which version is current, who changed a number, or which sub you forgot to chase. Those gaps are survivable on one job because the whole bid universe fits in your head. Run three or four jobs at once and the math turns against you.
Say each project carries 12 trades out to bid and you collect an average of three bids per trade. That is 36 bids a job. At four active jobs you are tracking 144 live bids, each with a scope, a number, an expiration date, and a follow-up owed. No one holds 144 moving items in working memory. The spreadsheet was never the system. Your memory was the system, and the spreadsheet was the notepad.
The Associated General Contractors of America publishes ongoing workforce and cost data showing how thin margins and labor shortages squeeze GC firms (agc.org). When the average commercial GC runs on a single-digit net margin, one mispriced scope gap eats the profit on the whole package. A tracking system that fails silently at scale is a direct threat to that margin.
What breaks first: version drift, missed follow-ups, and no audit trail
Three failures show up in order. Each one has a dollar number attached.
Version drift. You email the bid tab to a PM. He sorts it, fixes a formula, and emails it back. Now two files both named Mechanical_Bids_FINAL.xlsx exist, and they disagree. The drywall number on your copy is last week's. This is how a cut scope gets paid twice. I have seen this across hundreds of bid packages, and the door schedule is almost always where it bites, because doors live in the gap between the GC, the framer, and the door and hardware sub.
Missed follow-ups. A bid spreadsheet has no calendar. The electrical sub said he would firm up his number by Friday. Friday came and went during a different job's preconstruction meeting. You award off three bids when you should have had five, your coverage is thin, and the number you locked is 8 to 12 percent high because you never created real competition. Missed follow-ups do not look like losses. They look like a normal Tuesday.
No audit trail. Six weeks later a sub claims his bid never included the slab insulation. You open the spreadsheet. There is one cell with a number in it and zero history of what that number covered or when it landed. You cannot prove the scope. You eat the change order, often $5,000 to $20,000 on a mid-size commercial slab, because you have no record and a verbal bid is hard to defend (see change order).
How much does manual bid tracking actually cost a growing firm?
Manual bid tracking costs you in three buckets: rebid labor, lost competitive tension, and unrecoverable change orders. Here is how the two approaches line up once you pass three concurrent jobs.
| Failure point | One shared spreadsheet | Purpose-built bid workflow |
|---|---|---|
| Current version | Ambiguous, multiple FINAL files | Single record, last edit wins |
| Scope per bid | One number in a cell | Itemized scope attached to each bid |
| Follow-up reminders | None, lives in memory | Dated tasks per trade |
| Change history | Overwritten, gone | Timestamped, who and when |
| Coverage check | Manual eyeball | Bids-per-trade visible at a glance |
| License / insurance status | Tracked separately or not at all | Tied to the sub on the bid |
Put a number on it. If version drift and missed follow-ups cost you even one double-paid scope and two thin awards a quarter, that is realistically $25,000 to $60,000 a year for a firm doing $8M to $15M in volume. The spreadsheet is free. The failures are not.
What does a purpose-built bid management workflow fix?
A purpose-built workflow fixes the three failures by giving each bid a state, a scope, and a history. It treats a bid as a record, not a cell. That is the whole difference.
Concretely, a real bid workflow does this:
- Itemizes scope on every bid so you can run a true apples-to-apples leveling sheet across three subs instead of comparing one lump number to another lump number. The exclusions are where the money hides, and leveling is what surfaces them (see How to Read a Subcontractor Bid Without Getting Burned on Scope Gaps).
- Stamps every change with who and when, so a disputed change order six weeks out is a five-minute lookup, not a fight.
- Schedules follow-ups per trade so the Friday firm-up does not die in another job's meeting.
- Shows coverage at a glance, so you never award a $400,000 mechanical package off a single bid because you forgot the other two were still pending.
- Ties the bid to the sub's standing, so a number from a sub whose general liability lapsed or whose license is expired gets flagged before you award, not after the slab is poured. Vetting license, insurance, and bond status is the part of sourcing that protects you when a sub fails, and it belongs next to the bid. That linkage is the core of what we built SubSource to do.
A prequalified sub on a tracked bid is a defendable award. You can show the leveling sheet, the scope, the insurance certificate naming your project as additional insured, and the date you locked the number. That is the package that wins an argument and survives an audit.
The U.S. Census Bureau tracks construction spending in the hundreds of billions monthly (census.gov). The firms taking a growing share of that work are the ones that can bid more packages without the tracking falling apart. Capacity to bid is capacity to grow, and the spreadsheet caps it at about three jobs.
When should a GC move off the spreadsheet?
Move off the spreadsheet the moment a second person needs to edit it on a second job. The trigger is concurrency, not company size. A two-person shop running four jobs hits the wall before a ten-person shop running one does. If you are emailing a bid tab back and forth, you have already crossed it.
FAQ
What is bid leveling? Bid leveling is lining up competing subcontractor bids on a single sheet, scope item by scope item, so you compare the same work across each sub and expose what each one excluded. A lump-sum comparison hides those gaps; leveling shows them.
Can I just use more tabs in the same spreadsheet? More tabs delay the failure, they do not fix it. The problems are version drift, no reminders, and no change history. Tabs add none of those three. You still cannot tell which copy is current or prove what a number covered.
How many bids should I have per trade before I award? Aim for three firm, scoped bids per trade on anything material. Awarding off one or two leaves you 8 to 12 percent high on average because you never created competitive tension, and it leaves you exposed if your single bidder walks.
On your next bid package, do one thing: itemize the scope under each sub's number before you compare them. That one habit turns a row of lump sums into a leveling sheet, and it will surface the excluded slab insulation while you can still get it covered.