18.2 Verifying the Basis of Estimate & Analyzing Estimates
Key Takeaways
- The basis of estimate documents purpose and class, scope and drawing revisions, methodology, pricing and productivity sources, execution assumptions, escalation and currency basis, allowances, contingency method, exclusions, and known risks.
- Verification tests internal consistency (drawing revisions, escalation base dates, contingency versus class, allowance and contingency overlap) and external credibility (labor rates, productivity, quotes, freight, exchange rates).
- Composition analysis expressed as percentages exposes structurally wrong estimates — an unusually low indirect percentage often means indirects were factored from an undersized direct base.
- Sensitivity analysis typically shows a small number of drivers dominating, and craft productivity is usually the largest, which is where verification effort, risk response, and contingency should concentrate.
- When authorized budget falls below the estimate without a matching scope, quality, or schedule reduction, the gap must be recorded in writing at the time rather than absorbed silently into work packages.
18.2 Verifying the Basis of Estimate & Analyzing Estimates
Blueprint tasks 1.F verify basis of cost estimates, 1.G analyze cost estimates, and 1.H work with estimators to establish baseline budget sit at the centre of Domain 1. They cover what a cost professional does with an estimate after it exists and before it becomes a budget.
1. What a Basis of Estimate Actually Is
The basis of estimate (BOE) is the document that makes a number interpretable. Without it, an estimate is an unsourced assertion. AACE treats the BOE as a deliverable in its own right, and a competent reviewer reads it before looking at a single cost.
A complete BOE states:
| Component | Content |
|---|---|
| Purpose and class | What decision the estimate supports; AACE class and expected accuracy range |
| Scope basis | The specific drawings, specifications, equipment lists, and revision numbers priced |
| Methodology | How each portion was estimated — takeoff, factored, parametric, quoted, allowance |
| Pricing basis | Sources and dates for labor rates, material prices, and equipment quotations |
| Productivity basis | Assumed labor productivity and its source, plus any adjustment factors |
| Execution assumptions | Contracting strategy, work hours, shift patterns, site access, weather allowances |
| Escalation basis | Constant or current dollars; the index used; the base date and forecast period |
| Currency and exchange | Rates used and their date |
| Allowances | Each allowance, its coverage, and how it was derived |
| Contingency | The method (deterministic, risk-based, probabilistic) and the confidence level |
| Exclusions | What is deliberately not in the number |
| Risks and opportunities | Known items that could move the estimate, and their direction |
[!WARNING] Exclusions become claims. An exclusion buried on page 40 of a BOE and never surfaced to the decision-maker is, in practice, an unfunded scope item. Every review should extract the exclusion list and confirm someone senior has seen it and owns it.
2. Verifying the Basis (Task 1.F)
Verification is not proofreading. It tests whether the stated basis is internally consistent and externally credible.
Internal consistency checks:
- Does the priced scope match the drawing revisions listed? A BOE citing Rev B while the estimate prices Rev D is a real and common defect.
- Are the escalation base date and the pricing dates compatible? Pricing from quotes dated 18 months ago, escalated from today's base date, silently loses 18 months of escalation.
- Is contingency consistent with the stated class? A Class 4 estimate carrying 5% contingency is asserting a confidence its scope definition cannot support.
- Do the allowances and the contingency overlap? If an allowance already covers undefined small-bore piping and contingency also carries a line for it, the same uncertainty is funded twice.
External credibility checks:
- Are quoted labor rates consistent with the current agreement or market for that location and craft?
- Is the assumed productivity achievable on this site, in this season, with this labour supply?
- Do the equipment quotes carry realistic delivery terms, and is freight priced from the actual origin?
- Does the currency assumption reflect a rate anyone can actually transact at?
3. Analyzing the Estimate (Task 1.G)
Analysis asks what the estimate tells you, beyond its total.
Composition analysis
Break the estimate into direct labor, direct material, subcontract, equipment, indirects, escalation, and contingency, and express each as a percentage. Compare against the peer set. The shape of an estimate is diagnostic: an unusually low indirect percentage often means indirects were factored from an undersized direct base.
Sensitivity analysis
Identify the parameters that actually move the total. A structured sensitivity run typically shows a small number of drivers dominating.
Worked sensitivity. A $180,000,000 estimate has these components and plausible ranges:
| Parameter | Base | Range tested | Effect on total |
|---|---|---|---|
| Craft productivity factor | 1.00 | 0.90 to 1.35 | −$7.2M to +$25.2M |
| Structural steel price | $2,400/t | −10% to +25% | −$2.6M to +$6.5M |
| Escalation rate | 3.0%/yr | 1.5% to 6.0% | −$4.1M to +$8.9M |
| Equipment quotes | firm | −2% to +5% | −$0.9M to +$2.3M |
Productivity dominates: its unfavourable case moves the estimate more than the other three combined. That is where verification effort, risk response, and contingency should concentrate — and it is exactly the insight a total-only review cannot produce.
Trend and reconciliation
Compare against the previous estimate class and against the funding request. Every movement should be attributable to scope, quantity, rate, escalation, or contingency. Chapter 14 covers the probabilistic treatment; the point here is that an unreconciled movement is an unexamined risk.
4. Working with Estimators to Establish the Baseline Budget (Task 1.H)
The handover from estimate to budget is a specific collaboration, not a file transfer. Four things must happen:
- Agree the scope split. Which parts of the estimate become control accounts, and at what level. The estimate structure and the control structure are rarely identical, and the mapping must be explicit.
- Agree what is base and what is provision. Base estimate to control accounts; contingency held centrally; management reserve outside the baseline entirely.
- Preserve the basis with the budget. The BOE travels with the budget so that, twelve months later, a variance can be tested against the assumption that produced it.
- Record the reconciliation. Estimate total, adjustments, and authorized budget, line by line. Where the authorized budget is lower than the estimate — which happens — the difference must be recorded as a known, accepted funding gap rather than quietly absorbed into the work packages.
[!IMPORTANT] A budget cut without a scope cut is a deferred overrun. If authorization comes in below the estimate and no scope, quality, or schedule change accompanies it, the project starts with a negative variance that will surface later as an overrun. The cost professional's obligation — and, under the Canons of Ethics, the requirement to include all relevant and pertinent information — is to state this in writing at the time, not at the first cost report that reveals it.
A cost professional reviews the basis of estimate for a Class 3 estimate and finds that it cites drawing package Revision B, that the equipment pricing is drawn from quotations dated 18 months earlier, and that escalation is applied from a base date of today. What is the most significant defect?
Sensitivity analysis on a $180 million estimate shows these unfavourable-case impacts: craft productivity +$25.2M, structural steel price +$6.5M, escalation rate +$8.9M, equipment quotes +$2.3M. How should this change the team's approach?
An estimate is submitted at $240 million. The capital committee authorizes $215 million with no reduction in scope, quality, or schedule, and directs the project controls lead to allocate the $215 million across the control accounts. What is the cost professional's correct action?