Back to blog
Estimation Workflows

When Your Catalog Has No Price: Estimating Uncommon Items Without Blowing Your Margin

Practical strategies for handling provisional unit prices in construction budgets, from benchmarking similar materials to using live price lookups—and how an AI cost-intelligence tool fills the gaps.

Jorge de los Santos25/6/20265 min read

Every estimator knows the feeling: you’re building a budget in Presto, the BC3 is almost complete, and then you hit a line item that simply isn’t in any of your catalogs. Maybe it’s a specialty waterproof membrane, a custom steel bracket, or an imported fixture with no local reference price. You need to put something in that cell, but too low and you eat the difference later; too high and you lose the bid. This is the art—and risk—of the provisional price.

For a small builder or a busy QS, provisional prices are a leak in margin. A single missing unit price can snowball into a 5–10% cost overrun when the real supplier quote arrives and you’ve already committed to a budget. The fix isn’t just gut feeling. It’s a repeatable process that combines benchmarking, market sourcing, and—when available—a tool that can reach into live, multi-source price catalogs on your behalf.

Why provisional prices are a margin killer

A provisional price is essentially a placeholder. You know you need a certain material or subcontract package, but you don’t have a firm quote or a catalog entry. Common reasons:

  • The item is new to the market and not yet in standard databases.
  • You’re working in a region (Spain, Argentina, or elsewhere) where that product isn’t commonly used, so no local price exists.
  • The specification is still in draft—you have a description but no brand or supplier.

Without a disciplined approach, estimators either copy a similar item’s price without adjustment (risking under-pricing) or apply a blanket 20% “unknown” markup (risking over-pricing and losing the job). Both hurt margin. The goal is to land as close to real cost as possible before the subcontractor or supplier quote arrives.

That’s where a workflow built on cross-referencing and live price discovery becomes essential.

A three-step workflow for provisional unit prices

Whether you’re using Presto, a spreadsheet, or another platform, the same logic applies. Start with what you know, then systematically close the gap.

Step 1: Deconstruct the item.
Break the provisional item into its measurable components: material type, unit of measurement (m², kg, piece), expected waste factor, and installation method. For the waterproof membrane, you might know it’s a liquid-applied acrylic, applied at 1.2 mm thickness, so material consumption is about 2 kg/m². That gives you a quantity to price.

Step 2: Benchmark using proxy items.
Look for the closest match in your main catalog—same material family, similar application. Adjust the unit price by a simple factor: more complex installation? Add 10–15%. Higher performance spec? Add 20%. In Omnicost’s catalog, because it normalizes units across providers, you can quickly compare the price of an acrylic vs. a cementitious membrane and see a range. That range becomes your provisional price floor and ceiling.

Step 3: Live price look-up.
If no proxy is close enough, you need a real supplier price. Instead of calling five suppliers or waiting for email quotes, a cost-intelligence tool like Omnicost can crawl its multi-source catalog for any item description you feed it. You enter “liquid acrylic waterproof membrane 1.2 mm” and it returns live, structured prices from the catalogs it indexes—complete with provider, delivery zone, and date. You pick the median, apply a 5% contingency, and you have a defensible provisional price in minutes, not days.

Concrete worked example: Uncommon waterproof membrane in a renovation budget

You’re preparing a budget for a bathroom renovation in Barcelona. The architect specifies a liquid-applied polyurethane waterproof membrane for the wet areas. Your regular BC3 catalog has prices for cementitious membranes and mastic asphalt, but nothing for this specific product.

Step 1 – You define the unit of measurement as m². The spec says thickness 1.5 mm, coverage 1.5 kg/m². You estimate material quantity: 180 m² needed → 270 kg.

Step 2 – You find a proxy: cementitious membrane is priced at €8.50/m² (installed in a similar scenario). Polyurethane is more flexible and requires careful surface prep—you estimate 15% more labor and a higher material cost. Using Omnicost’s catalog comparison, you see that polyurethane products from different providers average €12.00–€15.00/m² for the material only. You set a provisional material price of €13.50/m².

Step 3 – To validate, you use Omnicost’s live price crawl. You enter “polyurethane liquid membrane 1.5 mm Barcelona”. The tool returns three hits from regional suppliers: €12.80, €13.20, and €14.10 per m² (including standard packaging, excl. VAT). You take the median (€13.20), apply a 5% market volatility buffer (€13.86), and enter that as your provisional unit price in the Presto BC3 budget.

You also add a note in the estimate: “Provisional price – to be confirmed with supplier prior to ordering; adjust budget if actual quote exceeds €14.50/m².”

The total provisional amount for this line is now €13.86 × 180 m² = €2,494.80. Without this workflow, you might have used the old cementitious price (€8.50) and been €964 under real cost—or doubled it blindly to €17.00 and possibly lost the bid.

How an AI cost-intelligence tool protects margin

A manual three-step workflow works, but it’s slow and inconsistent across your team. Omnicost automates the heavy lifting:

  • Multi-source price normalization means you’re not reinventing proxies each time—the catalog already groups similar materials, so you see price ranges instantly.
  • Live crawling lets you discover real market prices for unique items without leaving the estimation environment. The tool pulls from supplier catalogs, public databases, and market reports, so your provisional prices are rooted in current data.
  • The AI estimation agent can spot provisional entries and flag them for review, or suggest a price based on pattern-matching from thousands of similar projects. It doesn’t invent numbers; it surfaces what the market is actually quoting.

And because Omnicost works with the formats you already use (BC3, FIEBDC-3, Presto), the workflow slots into your existing process. You don’t have to change how you think about estimates—you just add a layer of real-time intelligence where the gaps are widest.

Conclusion: Turn provisional prices from guesswork into data

The next time you hit a line item with no catalog price, don’t label it “to be confirmed” and move on. That’s a margin leak waiting to happen. Instead, deconstruct the item, benchmark with proxies, and—if Omnicost is part of your toolkit—run a live price lookup. You’ll enter not a guess but a well-supported number that protects your margin on bid day and through project execution.

In construction cost estimation, the gap between provisional and actual is where wars of attrition get won or lost. Close that gap with data, and you keep your margin intact.

Jorge de los Santos

Founder, Omnicost

Jorge is the founder of Omnicost, where he builds AI-powered construction cost intelligence — a continuously updated, multi-source price catalog and an estimating agent for the construction industry.