Conceptual Estimating (ROM) for Design-Build and Early Project Phases
Most estimating discussions assume complete drawings and specifications as the starting point. The reality for many contractor workflows is that estimates are needed before complete drawings exist: design-build operations producing budgets during schematic design, contractors helping owners explore feasibility during programming, custom builders giving budget guidance during early architect engagement, commercial contractors providing GMP estimates while design is still developing.
These early-phase estimates have different methodology than hard-bid estimating from complete documents. They use Rough Order of Magnitude (ROM) approaches: parametric models based on building type, square-foot pricing from historical data, comparative analysis against similar completed projects. The accuracy targets are different (±15-20% is typical for ROM estimates versus ±3-5% for hard-bid estimates), and the contractual implications are different (ROM estimates inform planning rather than commit to firm pricing).
This article covers how conceptual estimating actually works, what software handles it well, and how to build credible early-phase estimates while managing risk appropriately.
How Conceptual Estimating Differs From Hard-Bid Estimating
The two approaches share some surface similarities but differ in methodology, accuracy expectations, and contractual implications.
What Hard-Bid Estimating Looks Like
Hard-bid estimating starts with complete drawings and specifications. The estimator performs detailed takeoff, applies costs from databases or vendor quotes, builds the estimate from line-item quantities, and produces a price the contractor commits to within reasonable variance.
The methodology is precise: every wall is counted, every fixture is identified, every material is specified. The accuracy target is tight: 3-5% variance from actual cost is typical for well-executed hard-bid estimates. The bid becomes the contract price, with formal change order processes governing modifications.
This approach requires complete documents. Without them, the takeoff has no basis for completeness, and the estimate has no foundation for accuracy.
What Conceptual Estimating Looks Like
Conceptual estimating starts with whatever information is available, often substantially less than complete documents. The estimator works from building type, square footage, quality level, owner-stated goals, and high-level program requirements. Costs come from parametric models (cost per square foot for similar building types), historical data (what the contractor has spent on similar past projects), and adjusted national database figures.
The methodology is approximation: building type drives cost per square foot, quality tier drives modifiers, location drives regional adjustment, complexity drives contingency. The accuracy target is looser: ±15-20% is typical and acceptable for ROM estimates. The estimate informs decisions rather than committing to fixed pricing.
The Different Use Cases
Hard-bid estimating happens at one specific point: when complete documents are issued for bidding and contractors compete for the work. Conceptual estimating happens at multiple points throughout a project's development:
Pre-design (programming): What's the rough cost of this building idea?
Schematic design: What's the cost as the design takes initial shape?
Design development: What's the cost as the design develops in more detail?
Construction documents: Final pricing as documents complete (transitions to hard-bid)
GMP development: Negotiated guaranteed maximum price during design completion
Each phase has different available information, different accuracy expectations, and different contractual implications.
Why Conceptual Estimates Matter
Owners need cost guidance before complete documents exist. Without it, design proceeds without budget reality, often producing designs that have to be substantially redesigned when costs become clear at the end. Conceptual estimates throughout the design process keep budget and design aligned, which prevents expensive late-stage redesign.
For design-build operations specifically, conceptual estimating is foundational to the value proposition. The design-builder commits to a price (or price range) early in the process, with the design developing toward that target. Without strong conceptual estimating capability, design-build doesn't really work.
Pro Tip: When producing conceptual estimates, always communicate the accuracy range explicitly. A ROM estimate presented as "the project will cost $4.2 million" sets expectations that the contractor can't reliably meet. The same estimate presented as "the project will likely cost between $3.8 million and $4.6 million, with $4.2 million as the midpoint" sets expectations that match what conceptual methodology can actually deliver. The explicit range protects both the contractor and the owner from disappointment when the final cost lands somewhere within the predicted range rather than at the predicted point.
How Conceptual Estimating Actually Works
The methodology has specific patterns that produce credible early-phase estimates.
Building Type Cost Models
The foundation of conceptual estimating is parametric cost data organized by building type: cost per square foot for office buildings, retail stores, hospitals, schools, residential homes, etc. The data captures typical construction cost for each building type at a baseline quality level.
Strong cost databases include parametric data for major building types. RSMeans Square Foot Costs, RSMeans Building Construction Cost Data, Marshall & Swift, and similar sources publish this data with regular updates. Internal data accumulated from completed projects produces even more accurate parametric models for the contractor's specific work types.
Quality Tier Modifiers
Buildings of the same type at different quality levels have substantially different costs. A starter office building might run $180 per square foot. A mid-tier office building might run $260. A high-end office building might run $360 or more. Quality tier modifiers adjust the baseline parametric cost to match the project's actual quality level.
Strong conceptual estimating includes explicit quality tier discussion with the owner: what level of finish, what level of mechanical systems, what level of structural sophistication. The conversation aligns expectations and produces estimates that match the actual project intent.
Regional Adjustment
Construction costs vary significantly by region. Boston is different from Phoenix. Both are different from rural Kansas. National parametric data includes regional adjustment factors that localize the costs for specific markets. Strong conceptual estimating applies regional adjustment automatically.
Complexity and Site Factors
Beyond building type and quality, project-specific factors affect cost: site conditions, access, schedule pressure, owner sophistication, building code stringency. These factors get applied as modifiers to the parametric baseline. A typical office building on a difficult site costs more than a typical office building on an easy site.
Historical Project Comparison
For contractors with significant historical data, comparison to similar completed projects produces more accurate ROM estimates than parametric data alone. The new hospital is similar to the hospital we built two years ago, with adjustments for size, location, and design changes. Coverage of historical data can be found in our construction cost databases area.
Building the ROM Estimate
The actual estimate construction follows a structured pattern:
1. Identify building type and approximate size
2. Apply baseline parametric cost
3. Adjust for quality tier
4. Adjust for regional location
5. Adjust for complexity and site factors
6. Apply contingency appropriate to the design phase
7. Produce a range rather than a point estimate
The output is an estimate range that reflects actual ROM methodology rather than false precision.
Contingency for Design Phase
Different design phases warrant different contingency levels. Programming-phase estimates might carry 25-30% contingency to account for the substantial unknowns. Schematic design might use 15-20%. Design development 10-15%. By the time construction documents are complete, contingency drops to typical hard-bid levels (3-5%) because the unknowns have been resolved.
The contingency isn't padding; it's honest acknowledgment that design completion produces information that affects costs. ROM estimates without appropriate contingency understate the actual cost range.
Case Study: A 60-person commercial design-build firm produced ROM estimates throughout their projects' design phases as standard practice. Their typical project went through 4 estimate iterations: programming (±25% contingency), schematic design (±18%), design development (±12%), and final construction documents (±5%). Each iteration informed design decisions and kept budget aligned with design intent. They tracked actual outcomes against their conceptual estimates: programming-phase estimates landed within ±22% of final cost on average (target was ±25%), schematic-phase estimates within ±15% (target ±18%), design-development within ±9% (target ±12%), and final estimates within ±4% (target ±5%). The accuracy at each phase was consistently slightly better than the published targets, which built strong owner trust over time. The lesson was that conceptual estimating accuracy improves with disciplined methodology and historical tracking. Operations that approach ROM estimating systematically produce better outcomes than operations that treat early-phase estimates as guesses dressed up as numbers.
How to Build Conceptual Estimating Capability
For contractors developing conceptual estimating capability, several patterns produce reliable results.
Build Internal Parametric Data Over Time
National parametric data is a useful starting point, but internal data accumulated from completed projects produces more accurate estimates for the contractor's specific work patterns. After 5-10 completed projects of a given building type, internal data typically outperforms national data for that type.
This requires tracking total project cost organized by building type and program, not just by cost code. The data structure is different from typical cost code tracking: total cost per square foot for each completed project, with attributes captured (quality tier, regional factors, complexity).
Use Software That Supports ROM Estimating
Some estimating platforms include explicit support for parametric and ROM estimating: Sage Estimating includes parametric models, several specialty platforms (DESTINI Estimator, BuildSoft) target the conceptual estimating workflow specifically.
Generic estimating platforms can be used for ROM with appropriate setup, but the workflow is less efficient than purpose-built tools. For operations doing significant ROM work, dedicated conceptual estimating capability earns its cost.
Document Assumptions Explicitly
ROM estimates depend heavily on assumptions: building type, quality tier, location, complexity, contingency level. Document these assumptions explicitly with every ROM estimate. When the design develops and costs evolve, the documented assumptions explain why the cost changed.
Without explicit assumptions, ROM estimate evolution looks like the contractor changing their mind. With explicit assumptions, evolution looks like reasonable response to design development.
Communicate Range Honestly
The temptation is to present ROM estimates as point figures because owners prefer single numbers. The discipline is to present ranges that match actual ROM accuracy. "The project will likely cost between $3.8M and $4.6M" is more useful than "The project will cost $4.2M" if the actual accuracy is ±15%.
Owners initially resist ranges but appreciate the honesty when costs land within the predicted range rather than missing a falsely-precise point estimate.
Update Regularly Through Design
Conceptual estimates aren't static. They update as design develops, as material prices move, as site conditions become clearer. Regular updates throughout design (typically at major design phase transitions) keep the estimate aligned with current information.
Operations that produce ROM estimates once and never update them produce predictable disappointments when final costs land outside the original prediction.
Connect to Hard-Bid Estimating
The transition from conceptual to hard-bid estimating happens when construction documents complete. Strong workflow uses the conceptual estimate as the starting point, with detailed takeoff and pricing refining toward the final hard-bid number. This continuity preserves the relationship between conceptual and final estimates.
Operations that treat conceptual and hard-bid as separate workflows often produce conceptual estimates that don't connect cleanly to final pricing, which damages trust when costs evolve unexpectedly.
Pro Tip: Treat your historical project data as the most valuable input to conceptual estimating capability. Every completed project should produce a structured record: total cost, square footage, building type, quality tier, location, complexity factors. This data accumulates into a parametric library specific to your operation that produces dramatically better ROM estimates than any external database. The discipline to capture this data systematically is what separates operations whose conceptual estimating improves over time from operations whose conceptual estimating stays approximate forever.
Conceptual Estimating Earns Its Place When Done Properly
ROM and conceptual estimating fill a real gap in the construction workflow. Owners need cost guidance before complete documents exist, and operations that can provide credible early-phase estimates have a meaningful advantage over operations that can only price from complete drawings. Design-build operations specifically depend on this capability for their value proposition.
The discipline isn't simple but isn't impossibly complex either. Building type cost models, quality tier modifiers, regional adjustment, project-specific factors, appropriate contingency, and honest range communication produce ROM estimates that meet owner needs while protecting the contractor from overpromising precision the methodology can't deliver.
The foundational explainer on estimating software can be found here: What is Estimating Software? Coverage of cost databases that support conceptual estimating lives here. The connection to design-build workflows can be found in our construction design software section. For coverage of how conceptual transitions to hard-bid, see 'How to Choose Estimating Software.'
Frequently Asked Questions
What's the typical accuracy of conceptual estimates?
Industry typical accuracy for ROM estimates is ±15-25% depending on the design phase: programming phase typically ±25%, schematic design ±18%, design development ±12%, with construction documents transitioning to hard-bid accuracy of ±3-5%. These ranges are baselines; specific operations should set their own targets based on historical performance and project complexity. Operations with strong internal parametric data and disciplined methodology often outperform these published targets.
When should contractors use conceptual estimating?
Conceptual estimating fits any phase before complete construction documents exist: programming, schematic design, design development. It's particularly important for design-build operations (where the contractor commits to pricing as design develops), GMP contracts (where guaranteed maximum prices are negotiated during design), and any owner relationship where budget guidance during design helps keep the project on track. Operations doing only hard-bid work from complete documents don't need conceptual estimating capability.
Can I do conceptual estimating in regular estimating software?
Yes, but it's less efficient than purpose-built tools. Most full estimating platforms can be configured for parametric estimating with appropriate setup: building type templates, square-foot cost data, quality tier modifiers. The setup work is meaningful but not impossible. Operations doing significant conceptual estimating typically benefit from platforms with built-in parametric capability (Sage Estimating, DESTINI, BuildSoft) over platforms designed primarily for hard-bid estimating.
How do I present a conceptual estimate to an owner without overpromising?
Use ranges rather than point estimates: "The project will likely cost between $3.8M and $4.6M with $4.2M as the midpoint" rather than "The project will cost $4.2M." Document assumptions explicitly: building type, quality tier, location adjustments, contingency level. Communicate the basis for the estimate: "This is based on similar projects we've completed, parametric data for this building type, and current market conditions." This framing protects both contractor and owner from disappointment when final costs land somewhere within the predicted range rather than at a falsely-precise point.