Start from actual hours, not the last fee
The last fee you quoted tells you what you charged, not what the work took. Pull the timesheet hours from two or three completed projects that match on scope, client type, and size, broken out by phase and by role. If the hours ran over budget on those jobs, the actual hours are the honest starting point.
Adjust for real differences: a larger building, an extra phase, a client known for long review cycles. Write each adjustment down so a reviewer can follow it.
Worked example: hours by phase and role
A firm is pricing the design of a building similar to two it completed recently. After adjusting the past actuals for scope, the estimate by phase is 180 hours for schematic design, 320 for design development, 520 for construction documents, and 220 for construction administration: 1,240 hours in total.
Spread across roles, and costed at each role's direct labor cost (salary per hour, before overhead):
- Principal: 100 hours at $85 = $8,500
- Project manager: 250 hours at $60 = $15,000
- Project engineer: 580 hours at $45 = $26,100
- Technician: 310 hours at $32 = $9,920
- Total: 1,240 hours and $59,520 of direct labor cost
Apply the multiplier
The multiplier turns direct labor cost into fee. It has to cover overhead and leave a profit. Your breakeven multiplier is 1 plus your overhead rate: a firm with a 160 percent overhead rate breaks even at 2.6 times direct labor. Anything above that is profit, before any write-downs.
At a target multiplier of 3.0, the labor fee is $59,520 times 3.0 = $178,560. The same result comes from billing rates set at 3.0 times each role's cost: $255 for the principal, $180 for the project manager, $135 for the engineer, and $96 for the technician.
Add contingency and expenses
Contingency covers known unknowns, such as an unclear scope item or a review process you have not been through. Size it to the risk, write down what it is for, and do not use it to hide padding. In the example, 8 percent of the labor fee is $178,560 times 0.08 = $14,285, rounded.
Add reimbursable or direct expenses, such as travel, printing, and permit fees, as their own line. With $6,000 of expenses the total is $178,560 + $14,285 + $6,000 = $198,845, which the firm proposes as $199,000.
Check against what similar work actually earned
Before sending the number, look at the outcome of the past projects you based it on. Suppose one of them earned $165,000 of net service revenue on $63,000 of direct labor cost. Its achieved multiplier was $165,000 divided by $63,000, about 2.62, barely above the 2.6 breakeven. The firm planned a profit on that job and ended up close to even.
That is a warning, not a footnote. Either the hours ran over or the fee was too low for the scope. The new estimate already uses the actual hours, which helps. If the client will not accept $199,000, reduce the scope to match the fee rather than the multiplier, or treat it as a no-go.
Match the contract type to the risk
A fixed fee puts the risk of extra hours on you, so it needs the contingency and a tightly written scope. Time and materials puts that risk on the client, and a not-to-exceed limit caps it, which leaves you exposed if the cap is set at your estimate with no room. On qualifications-based selection, such as federal architect-engineer work under FAR Subpart 36.6, the fee is negotiated after the firm is selected, so arrive with this breakdown ready to defend line by line.
Where Hubity fits
Hubity rolls hours logged on project tasks up to each project, so past hours by project are on hand when you price similar work, and it answers questions about past proposals and rate sheets from the firm's own records with the sources it used.