Why Builders Lose Money on Their Best Jobs (And How to Fix It)

Sound Familiar?

Margin is disappearing on your best jobs if:

• Your clients never complain about change orders because you rarely submit them — even when scope clearly expanded after the contract was signed
• Small additions and repairs done during a project get absorbed as 'goodwill' without a documented cost or conversation
• Your overhead is allocated as a flat percentage of revenue across all jobs, regardless of how much management complexity each job actually consumed
• Labor hours for time spent troubleshooting, coordinating, or doing warranty-adjacent work during the project get logged to the job generically without cost code detail
• You finished a job, the client was happy, and 60 days later you're still not sure whether you made money on it
• Your project managers report jobs as 'going well' while your job cost report shows budget erosion you can't explain

What We Found

Scenario 1: Scope Creep Without Change Orders — The Most Common Margin Leak

Scope creep is the gradual, often invisible expansion of what the project covers beyond what was contracted. The client asks for a slightly different window configuration during framing. They decide to paint the garage while you're there. They ask for an extra outlet in a room that wasn't wired for one. Each request feels small. Each one takes real time and material. And in most construction businesses, most of them never get documented or billed.

The reason builders absorb these changes is almost never malice. It's a combination of relationship dynamics ("I don't want to nickel-and-dime them"), time pressure ("it'll take longer to write up the change order than to just do it"), and poor systems ("I'll track it at the end" — which never happens). The result is the same regardless of motivation: real costs with no corresponding revenue.

The Math Is Worse Than It Looks

A job with $8,000 in untracked scope changes on a $200,000 project represents a 4-point margin hit. If your target margin is 15% and you expected $30,000 in gross profit, you finished with $22,000. That's not a cost overrun problem — it's a documentation and billing problem. The work was done. The cost was incurred. The money just wasn't collected.

Multiply that across a $3M business with modest scope creep on every project and the Gap Selling framework identifies this precisely: Gap Selling analysis: a builder running $3M in annual revenue who leaks just 5% through undocumented changes, untracked labor, and informal scope creep loses $150,000 in profit annually — equivalent to 3 full additional projects just to break even on the leak.

The System That Closes This Gap

The fix is a standing rule, enforced by both the owner and project managers: any scope addition that requires more than 30 minutes of labor or more than $200 in materials gets a change order before work starts. Not after. Not as a batch at the end. Before.

The change order doesn't need to be complex. It needs to exist: a written description of the work, a price, and a client signature (electronic is fine). Most PM platforms — JobTread, Buildertrend, CoConstruct — make this a 5-minute process. The discipline is the hard part. The technology is the easy part.

When clients push back on change orders for small items, the framing that works is: "I want to be transparent about what every addition costs rather than building it into the final invoice as an unexplained number." Most clients, once they understand this is about transparency and not about extracting money, accept it. The ones who don't accept it are telling you something important about how billing conversations will go at project close.

Scenario 2: Unbilled Extras — The Goodwill That Costs You the Business

Unbilled extras are the cousin of scope creep. Where scope creep is expanded scope that goes undocumented, unbilled extras are additions that get done, sometimes get tracked, but never get billed because the decision was made (explicitly or implicitly) to absorb them as goodwill.

Common examples: fixing a minor item discovered during punch list that wasn't in scope, touching up paint on an adjacent wall the client noticed, correcting a sub's minor workmanship issue without a back-charge, doing a small repair during a walk-through that "only took 20 minutes." Every one of these has a cost. Labor, materials, PM time. Most of them never appear on an invoice.

Why This Happens

The impulse behind unbilled extras is understandable — builders want clients to feel well taken care of, want to protect the relationship, want to generate referrals. Those are legitimate goals. The problem is using direct job cost absorption as the mechanism to achieve them.

There is a difference between a strategic goodwill gesture — deciding consciously to absorb a specific cost because the relationship value or referral potential justifies it — and a reflexive default to absorption because the billing conversation feels uncomfortable. The first is a business decision. The second is a systems failure.

The Measurement Problem

Most builders who absorb extras don't track what they're absorbing. They have a vague sense that they "give a lot away," but they can't quantify it. The first step is measurement: track every unbilled extra for 90 days. Log it to a job, categorize it (goodwill, sub correction, scope ambiguity, warranty-adjacent), and put a dollar value on it. At the end of 90 days, you'll know what goodwill is actually costing you — not as a feeling, but as a number.

The builders who do this exercise almost always find the number is larger than expected. A business doing $3M in revenue often discovers it's absorbing $30,000–$60,000 annually in unbilled extras that were never consciously approved as business development investments. At that point, the choice becomes deliberate: decide which extras are worth absorbing and which ones should be billed. That's a business decision. Currently it's just a leak.

The System: Pre-Authorized Goodwill Budget

The most effective mechanism I've seen: a per-project goodwill budget of $500–$1,500 depending on project size. PM managers have authority to absorb costs up to that threshold without change orders for genuine goodwill gestures. Anything above the threshold requires approval. This gives PMs flexibility to build client relationships without requiring change order discipline on every $75 touch-up — but it puts a ceiling on the absorption and makes it a tracked, intentional decision rather than an invisible default.

Scenario 3: Overhead Mis-Allocation — Why Simple Jobs Subsidize Complex Ones

Overhead allocation is one of the least-discussed sources of margin distortion in construction financial systems. Most builders allocate overhead as a flat percentage of revenue across all jobs: if overhead is 20% of revenue, every dollar of revenue on every job carries a 20% overhead burden. The logic is simple. The problem is that it's wrong.

Overhead is not evenly consumed by all jobs. A $400,000 cookie-cutter spec home with standard scope, minimal coordination, and a two-person PM team consumes far less overhead than a $400,000 custom renovation with 12 subcontractors, daily client interactions, significant design coordination, and two rounds of city inspections. Same revenue. Vastly different overhead consumption. Flat-rate allocation makes the simple job look less profitable than it is and makes the complex job look more profitable than it is.

Why This Matters for Job Selection

When overhead is mis-allocated, you're making job selection and bid decisions based on inaccurate profitability data. You may be steering toward complex, high-overhead projects because they appear to generate strong margin on your job cost reports — when in reality the overhead they consume is being subsidized by your simpler jobs. The builders who run purgatory — good revenue, flat profits — often discover this is one of the mechanisms: they're winning and executing complex, high-overhead work while believing it's their most profitable project type.

Better Overhead Allocation: The Activity-Based Approach

A more accurate allocation model assigns overhead based on actual resource consumption. The simplest version: weight overhead by a combination of revenue percentage AND estimated PM hours. A job that requires 60 PM hours carries more overhead than a job requiring 20 PM hours, even at the same revenue. This two-factor model is more work to implement than a flat percentage but produces far more accurate job-level margin data.

The even simpler version for builders who don't want to rebuild their overhead model: start tracking PM hours by project. After two to three quarters, you'll have data on which project types actually consume overhead versus which ones your flat rate is overcharging. Use that data to adjust your overhead percentage by project type — a lower rate for simple, well-defined scope and a higher rate for complex, coordination-intensive work.

What Accurate Overhead Allocation Changes

When overhead is allocated accurately, two things change immediately: (1) you can see which project types are genuinely most profitable, not just highest revenue, and (2) your bids for complex work include the true overhead cost rather than a subsidized flat rate. That second point means your bids may be higher on complex jobs than they used to be — which is correct. You were underpricing complexity. Accurate allocation corrects that systematically.

Scenario 4: Invisible Labor Leakage — Hours That Work But Never Show Up

Labor leakage is the fourth mechanism — and arguably the hardest to fix because it requires behavioral change from the field team, not just a system configuration. Labor leakage happens when real labor hours are consumed on a job but never logged, or logged incorrectly, so they disappear from the job cost report.

The most common patterns:

• Drive time logged to general overhead rather than the specific job (or not logged at all)
• Morning setup and afternoon breakdown time not captured in daily logs because field teams log the productive hours and forget the prep time
• Troubleshooting and coordination time by the foreman or lead carpenter — time spent on the phone with subs, reviewing plans, solving problems — logged at a generic crew rate rather than the actual burdened cost of who was doing it
• Return trips for punchlist, warranty-adjacent items, or material runs that never get logged to the job
• Owner time on the project — the builder who is managing the job, solving problems, and running client meetings but not logging any of that time as a job cost because "that's just what I do"

The industry term for this pattern is clear: Phantom Resource — Labor, materials, or subcontractor time that was consumed on a job but never logged — hours that worked but never showed up in the job cost report.. The resource worked. The cost was incurred. It simply never appeared in the job cost report.

The Cost of Not Tracking It

An owner who spends 40 hours managing a specific project — estimating, client meetings, sub coordination, punch list walks — at a burdened cost of $75/hour has $3,000 of cost on that job that never appears anywhere in the financials. On a $180,000 project targeting 15% gross margin, that $3,000 alone moves the margin from 15% to 13.3%. Multiply by 12 projects a year and you've identified $36,000 in annual cost that the financial reports don't show — and that explains why the P&L doesn't match what the job cost reports suggest it should be.

The System: Time Logging Discipline

The fix requires two things: a consistent daily log process that captures all job-coded time (including travel, setup, troubleshooting, and coordination), and owner willingness to log their own time at a realistic burdened rate.

For field teams, the daily log should be non-negotiable: every crew member logs their hours, by job, with cost code, at end of day. Most PM platforms have mobile apps that make this a 3-minute process. For owners, the discipline is tracking at least the most time-intensive projects explicitly. You don't need perfect owner time tracking for every email and phone call — but the major blocks of time (full-day site visits, 3-hour estimating sessions, extended client meetings) should be logged to the relevant job.

Once labor logging is consistent, re-run your bid-to-actual analysis on labor by cost code. You'll likely find that actual labor hours are consistently higher than estimated — and now you'll know which specific categories are the gap, which lets you correct the template rates and improve future estimates.

If these patterns resonate — scope creep going unbilled, extras being absorbed, overhead allocation working against you, labor disappearing from your cost reports — the place to start is a direct conversation about where your specific business is leaking. A strategy call gets specific fast: 45 minutes to map your current margin structure and identify the highest-dollar fixes.

Frequently Asked Questions