Solve · Construction cost data
Quantity takeoff and predictive estimating both assume one thing: that a rate from one project means the same as a rate from another. Across countries, currencies, suppliers and years, it usually does not. J Labs standardises the procurement data first, then builds the pricing layer on top of something that compares.
Summary
J Labs makes construction cost data comparable before anything is built on it. The tools in this space, automated quantity takeoff and predictive estimating, are real and they work; what they assume is that a rate from one project means the same as a rate from another. Across countries that assumption fails quietly. Units differ, currencies differ and so does the date a rate was captured, supplier and material names vary by region and by the person who typed them, and the same cost is recorded at different levels of the breakdown on different jobs. The work is to standardise those first: one unit convention, rates carried with their currency and their date, supplier and material names reconciled to a single register, and the breakdown level made explicit so like is compared with like. Where a figure cannot be reconciled it is flagged rather than averaged in. Only then does a pricing layer mean anything. For a Danish quantity surveying firm running global construction programmes, procurement data was standardised across 6 countries in a six-week cleanup phase, and the pricing intelligence layer was built on top of it. Engagements are fixed scope, fixed timeline and fixed price.
What we automate
The same item measured per square metre on one job and per item on another is two numbers that cannot be averaged. Agreeing one convention and converting the history to it is unglamorous and it is the step everything else depends on.
A rate without the date it was captured is not a rate, it is a rumour. Costs are stored with the currency and the moment they were true, so a comparison across years is a comparison rather than an accident of exchange rates.
The same supplier spelled four ways and the same material described differently by region are the most common reason a cost query returns a third of what it should. Names are reconciled to one register, with the near misses a string match would not catch.
A cost recorded at element level on one project and at component level on another will quietly produce a wrong total. The level is recorded, and comparisons state which level they are made at.
Anything the standardisation is unsure about goes to a person with the original record beside it. A dataset that hides its own gaps produces an estimate nobody can defend when it is questioned.
Benchmarks, rate libraries and estimating support become worth having at this point and not before. This is the order the work goes in, and it is the part most tooling assumes has already happened.
Fit
This is for you if
You have years of procurement data and cannot query it. The history exists in a system or a stack of spreadsheets, and the answers it should give either take days to assemble or come out wrong often enough that nobody trusts them.
You work across countries, currencies or decades. Every one of those multiplies the ways a rate can fail to mean what it appears to mean. It is also what makes the data valuable once it compares.
Someone can settle the conventions. One person who can decide the unit convention and the breakdown level and have it stick. That is the shape the Spec covers: specified, built and tested for a fixed price.
This is not for you if
You want quantity takeoff from drawings. That is a mature product category with good tools in it, and buying one is the right move. We work on what the output of those tools gets compared against.
You want a cost benchmark we supply. We do not sell a rate library. The value here is your own history made usable, which is data nobody else has and the reason your estimates can be defended.
The data does not exist yet. If procurement history was never captured in any system, there is nothing to standardise. Start by capturing it, which is a smaller piece of work and not this one.
Systems
The ones you already run. We don't insist on rip-and-replace: the automation layer goes on top of the systems in place, and if a system is genuinely past saving we say so on the discovery call and show our working.
Engagement options
| Engagement | Price | Timeline | What you get |
|---|---|---|---|
| The Spec | €5,000, fixed | Fifteen working days | One clearly defined outcome, designed, built and tested: one data source standardised against conventions you agree, with unreconcilable records flagged rather than absorbed. A free assessment confirms it fits before you pay. |
| Cleanup and pricing layer | Fixed price, quoted after discovery | Typically four to eight weeks | The standardisation across your real history, then the querying and benchmarking layer on top, with weekly check-ins, open staging and documentation written as we go. |
| Bespoke | Multi-phase, to €200,000+ | Multi-month, phased | Cost intelligence across countries, currencies and entities, where the cleanup is the project and the reporting follows it, phased with weekly check-ins. |
None of these is an off-the-shelf product. There is no J Labs software to install: every system is specified in the brief, built for your stack, and owned by you. Every engagement is fixed scope, fixed timeline and fixed price, with acceptance criteria agreed before work starts. See how we work and the Spec.
What we've shipped
Quantity surveying · Global
Procurement data standardised across 6 countries in a six-week cleanup phase, with a pricing intelligence layer built on top of it.
The product that fixes it
The Spec · 5,000 EUR
One source, a convention you can state, and a comparison that either reconciles or gets flagged. That is one outcome with a testable definition of done, which is what the Spec is scoped for.
Live availability
Live slots from our team calendar, weekdays, European working hours. A 20-minute intro or a 45-minute working session, your choice, on Google Meet. You get a one-page summary within 24 hours.
How much time do you need?
Step 1 · Pick a date
Loading availability…
Frequently asked
Or start with our brief templateWe reply within one business day.