Episode Description:
A building automation project can look profitable when it is sold and tell a very different story when it closes out.
Often, the problem is labor.
Hardware costs are relatively straightforward to quote. Labor requires judgment. If that judgment misses part of the work, overlooks site conditions, or relies on assumptions that no longer match reality, those missing hours can quickly consume project margin.
In Episode 565 of the Smart Buildings Academy Podcast, you’ll explore how to approach labor estimation so the number reflects the work your team will actually perform.
If you estimate, sell, manage, or execute building automation projects, this episode will help you think differently about what should happen before a proposal goes out the door and what your completed projects can teach you about the next estimate.
Topics Covered
• Why building automation labor estimates can drift so far from actual hours
• The phases of controls work that need to be considered during estimating
• Where overlooked and uncertain labor can put project margin at risk
• How sales and operations can align before a project is booked
• Why completed job data should influence future estimates
A defensible labor estimate is not about adding more hours. It is about selling the work at a number your operations team can trust and your project can deliver.
Click here to download or listen to this episode now.Podcast Video
A building automation project can look profitable when it is booked and tell a very different story when it closes.
Consider a retrofit with 12 air handlers, roughly 140 VAV boxes, a new front end, and integration with an existing chiller plant. The material costs are based on vendor quotes and come in close to expectations. The project is sold with 220 labor hours.
Ten months later, the project closes at 410 hours.
The controllers did not suddenly double in price. The sensors and actuators were not the problem. The margin disappeared through labor.
This is one of the most important challenges in building automation estimating. Material is usually tied to a quote. Labor is built from judgment.
The quality of that judgment can determine whether the margin shown at booking still exists when the project closes.
Labor estimates often fail because the estimator starts with a total number instead of starting with the work.
A project may be assigned 200, 500, or 1,000 hours based on a similar project, an estimating spreadsheet, historical experience, or an experienced salesperson's intuition. Those references can be useful, but a total labor number does not explain how the work will actually be performed.
Building automation projects contain multiple types of labor.
Engineering is different from panel fabrication. Installation is different from programming. Commissioning is different from graphics. Project management and travel can consume significant time without producing a visible installed device.
When all of that work is compressed into one number, omissions become difficult to see.
A stronger estimate begins by breaking the project into the phases of work the team will actually execute.
Before assigning hours, map out the full project lifecycle.
Depending on the organization and project, that can include engineering, submittals, procurement support, panel work, installation, programming, graphics, integration, startup, commissioning, project management, training, closeout, and other required activities.
The exact structure matters less than the discipline behind it.
Every meaningful category of labor should have a place in the estimate.
This creates visibility. Instead of asking whether 500 hours sounds reasonable, an estimator can ask whether the hours allocated to engineering, programming, commissioning, project management, and other phases make sense individually.
That is a much easier estimate to review and defend.
It also gives operations a clearer picture of what sales assumed when the project was priced.
Once the work is separated into phases, estimates should be based on real production data whenever possible.
If your team consistently requires three hours to complete a particular type of controller installation, continuing to estimate two and a half hours creates predictable margin loss.
One completed project may be an exception.
A quarter of projects showing the same pattern is evidence.
Historical job-cost data can help establish realistic unit rates for repeatable work. Those rates might be tied to devices, equipment types, sequences, graphics, integrations, panels, or other measurable units.
The goal is to replace assumptions with evidence.
This does not eliminate professional judgment. It gives that judgment a stronger foundation.
Historical unit rates provide a baseline, not a final answer.
Two projects with the same number of controllers can require very different amounts of labor.
Existing conditions matter. Site access matters. Integration requirements matter. Customer expectations matter. Schedule constraints matter. Crew experience matters.
A retrofit in an occupied facility may require more coordination than comparable work in a new building. An existing chiller plant may involve undocumented sequences, unfamiliar protocols, or equipment that does not behave as expected. A compressed schedule may increase coordination requirements or reduce opportunities to perform work efficiently.
Estimators need to ask what makes this project different from the projects that produced the baseline labor rates.
This is where conversations with field personnel become valuable.
A technician who worked at the site last year may know more about the labor risk than anyone looking at the drawings.
Ask before the bid goes out.
What happened on the previous project? What conditions slowed the team down? Is the existing system documented accurately? Are there access restrictions? Does the customer have unusual commissioning or documentation expectations?
That information can change an estimate before it becomes a contractual commitment.
Margin does not disappear only through major estimating errors.
Small categories of labor can accumulate across a project.
Travel, meetings, coordination, troubleshooting, return trips, documentation, training, commissioning support, integration testing, and closeout activities all consume time.
If those activities are predictable, they should be represented in the estimate.
If they are uncertain, the uncertainty should be addressed deliberately rather than ignored.
There is an important difference between work that can be counted and risk that cannot be predicted precisely.
Countable work belongs in the estimate as labor.
Uncertainty may require contingency, qualifications, exclusions, or another appropriate commercial mechanism.
Treating both categories as invisible overhead makes it difficult to understand why projects exceed their labor budgets.
A labor estimate is built on assumptions.
Those assumptions should not exist only in the estimator's head.
If the estimate assumes a specific number of site visits, state it. If existing equipment must meet certain conditions, document those conditions. If particular integration work is excluded, make the exclusion clear.
Qualifications and exclusions help connect the price being offered to the work being promised.
They also create alignment between the customer, sales team, and operations team.
When project conditions change, documented assumptions provide a reference point for determining whether the work still matches what was originally sold.
That can be important for change management and margin protection.
One of the most effective estimating checks can also be one of the simplest.
Have someone from operations review the labor estimate before the proposal is submitted.
They do not necessarily need to review the entire commercial proposal. Focus the review on labor.
A project manager or experienced lead technician can look at a phase-by-phase estimate and identify gaps quickly.
They may notice that commissioning is missing, the expected number of site visits is unrealistic, an integration is more complicated than estimated, or the programming hours do not match the sequence requirements.
A short review before the sale can prevent hundreds of hours of disagreement after the project is booked.
It also changes the relationship between sales and operations.
Instead of operations receiving a labor budget they had no role in evaluating, both teams have already agreed that the planned hours are achievable.
Estimating should not end when the project is sold.
Completed projects provide the data required to improve future estimates.
The key is to compare estimated and actual labor at a useful level of detail.
A project that exceeded its total labor budget by 150 hours tells you that something went wrong. It does not tell you where.
Phase-level tracking can reveal whether the variance came from engineering, installation, programming, graphics, commissioning, project management, integration, or another category.
Those patterns can then update future estimating rates.
If one job exceeds the expected labor for a certain task, investigate it.
If the same task consistently exceeds the estimate across several months or a quarter, the estimating standard may be wrong.
Update it.
Each completed project should make the next estimate more accurate.
Improving labor estimating does not mean increasing every estimate.
Sometimes detailed estimating produces a lower number.
A repeatable VAV project may require fewer hours than an estimator would assign if every controller were treated as custom work. Standardized programming, templates, proven workflows, and experienced crews can create real efficiencies.
Those efficiencies should be reflected in the estimate too.
The objective is not to create the largest defensible labor number.
It is to create the right number.
A strong labor estimate should reflect how the project will actually be delivered, the conditions under which the team will perform the work, and what the organization has learned from previous projects.
A disciplined labor estimating process can follow a repeatable cycle:
This creates a feedback loop between sales, estimating, operations, and completed project data.
The result is not simply a better spreadsheet.
It is a project that sales can price with confidence, operations can execute against, and leadership can evaluate using meaningful data.
Material pricing usually starts with a quote. Labor starts with judgment.
Building automation companies that strengthen that judgment with project phases, historical data, site-specific knowledge, operational review, and closed-loop feedback can build estimates that hold up from booking through closeout.
That is the standard that matters: not a project that looks profitable when it is sold, but one that still looks profitable when the work is complete.
For a deeper discussion and insights from the field, listen to this episode on the Smart Buildings Academy podcast.