CH 00 / 06 — OVERTURE

Who it's for — a field study in four chapters

The people who carry the project.

Four seats hold every capital project together: the contractor defending margin, the owner buying truth, the program director seeing across the portfolio, and the controls lead keeping the number honest. Each is accountable for a truth they can only partly see. This is a study of those four seats — and of what changes when all of them read from one record.

Begin — Chapter 01

Chapter 01 of 06

Why projects fail

Capital projects do not fail in a moment. They fail in the gaps.

A nine-figure project is not one system — it is a schedule in one tool, a cost ledger in another, a contract in a PDF, a risk register in a spreadsheet, and the truth of the site in a thousand emails. Each discipline is competent alone; the failure lives in the space between them, where a delay takes weeks to become a cost, and a cost takes a month to become a report.

By the time the numbers agree, they describe a project that no longer exists. Every month those versions are reconciled by hand into a report that is out of date before it is read, and the failures that follow are not failures of engineering — they are failures of record.

That is why the answer is not another tool. It is one record: kept live, kept honest, and readable from every seat. What follows are the five failure modes we built Verian against.

EXHIBIT A

The gaps between disciplines

Schedule lives with the planners, cost with the controllers, changes with contracts, and reality with the field. Each system is internally consistent and mutually contradictory, and nobody owns the joins between them. A slipped activity takes three weeks to become a revised forecast and a month to appear in a report — the project does not drift in any one system; it drifts in the space between them, where nobody's job description lives.

EXHIBIT B

The cold record

The evidence that decides a dispute is created months before anyone knows there will be one. By the time a claim is contested, the daily reports are archived, the photos are scattered across phones, and the institutional memory has changed employers. A record that has to be reconstructed is a record that has already lost — and cold records settle for cents on the dollar.

EXHIBIT C

Filtered truth

Every layer between the field and the funder summarizes, softens, and selects. None of it is lying, exactly — it is reporting, and reporting has an author, and the author has a position. The people carrying the most financial exposure routinely hold the least direct evidence about how the project is actually going, and bad news travels upward at the speed of the monthly cycle — which is to say, too late to act on.

EXHIBIT D

The portfolio that won't roll up

Ask twelve projects for their status and you will get twelve formats, twelve definitions of percent complete, and twelve different dates “as of.” The program office spends the first third of each month translating before it can aggregate, and the resulting portfolio view is a hand-built artifact describing last month. The one question leadership actually asks — where is the next failure forming — cannot be answered from data that never rolls up the same way twice.

EXHIBIT E

The reconciliation tax

The most expensive people in project controls spend most of their month making systems agree instead of making decisions: exporting, mapping, keying, and explaining the differences between copies of the same numbers. It is skilled work that produces nothing — the numbers were once identical and were separated only by the tools that hold them. It is treated as the cost of doing business. It is actually the cost of not having one record.

Act II — The Four Seats

Four seats at the same table. Four versions of the truth.

FIG. 01 — FIELD CONDITIONS · CONTINUOUS RECORD

Chapter 02 of 06

EPC Contractors

The margin lives in the record.

The seat

It is month eleven and the client's engineer has rejected your extension of time. You know the delay was theirs — the late IFC drawings, the access that never came — but knowing is not evidence. Your proof is spread across daily reports nobody indexed, emails on three servers, a baseline revised nine times, and a superintendent's memory. Your commercial team spends six weeks reconstructing what happened in week fourteen. Meanwhile the next change order ages past its notice period, and the margin you bid quietly becomes the margin you hope for.

With Verian

The record is built while the work happens, not reconstructed after it is contested. Every daily report, RFI, delay event, photograph, and change is captured against the schedule activity and cost code it touches — time-stamped, attributable, linked. When an EOT is challenged, the forensic narrative of cause, notice, impact, and cost assembles from the living record in hours, not weeks. Notice periods are tracked before they lapse. You defend claims with contemporaneous evidence, price change while it is still negotiable, and protect margin at the moment it is actually decided: in the record.

5–15%

of contract value typically lost to unrecovered change and disputed claims

Hours

not weeks, to assemble a claim-ready delay file with its evidence attached

Day one

every site record time-stamped and linked to schedule and cost, before any dispute exists

A claim is won or lost the day the record is made — not the day it is argued.

Chapter 03 of 06

Project Owners & Developers

Independent

EAC and delay analysis, computed from source data — not reported to you

Monthly → live

the cadence of true schedule and cost visibility

First to know

critical-path slip and float erosion flagged as they emerge, weeks before the report

Funded first. Informed last.

The seat

You are funding a nine-figure project, and your primary instrument for seeing it is a monthly report written by the party you would contest a claim against. It arrives on the fourth — forty pages, professionally bound in optimism. The EAC has not moved in five reporting periods, which is itself the warning. Your lender wants a forecast to completion you can stand behind; what you have is the contractor's number, a month old, built on assumptions you cannot inspect. When the slip finally surfaces, it surfaces as a fait accompli — the float is gone, the levers have expired, and your first real decision is which bad option to fund.

With Verian

The owner holds an independent line of sight that does not route through the contractor's narrative. The platform reads the same schedule submissions, cost data, and change registers the delivery team works from, and computes its own analysis — critical-path movement, float erosion, earned value against actuals, an estimate at completion built from evidence rather than assurances — continuously, not monthly. When the contractor's number and the independent number diverge, you see the divergence the week it opens, with the drivers underneath it. Governance stops being a ritual of receiving reports and becomes an act of verification.

The most expensive sentence in capital projects is “we only just found out.”

Chapter 04 of 06

Program Directors

A portfolio is not twelve reports stapled together.

Twelve projects. One answer.

The seat

Ask your program where it stands and you get twelve answers in eleven formats. Every project team codes cost differently, baselines differently, defines complete differently. Your PMO spends three weeks a quarter normalizing it all into one deck — by which time it describes the program as it was, not as it is. Somewhere in the portfolio, three projects share the same critical subcontractor, the same escalating commodity, the same slipping weather window — and no roll-up you currently have will show you that before it becomes a program-level event.

With Verian

Every project — whatever its contractor, template, or toolchain — resolves into one canonical data model, so the portfolio stops being an aggregation exercise. Portfolio EAC, schedule health, contract exposure, and risk concentration roll up natively — same structures, same definitions, same day — and every number drills down: portfolio to project, project to activity, activity to the site record that produced it. Cross-project exposure appears as a pattern, not a post-mortem. You reallocate capital and attention while doing so still changes the outcome, and you answer the board from evidence, not assembly.

12 → 1

disparate project files resolved into a single portfolio model, no translation layer

Same day

from data cut-off to a trustworthy program view — not weeks

4 levels

of drill-down: portfolio → project → activity → source record

Chapter 05 of 06

Cost & Controls Leads

The month-end that ate the month.

The seat

The month has a shape, and you dread it. Close the actuals, export from the ERP, export from the scheduler, chase the commitments file, then begin the ritual: making four systems that disagree tell one story, mapped by hand across a forty-tab workbook only one person understands. The variance report is due Friday; Thursday night you are tracing a two-million-dollar discrepancy to a mis-mapped cost code. The analysis you were actually hired for — why the variance, what it means, what to do — gets whatever hours remain, which is usually none. And when the auditors come, the trail lives in a folder structure only you can explain.

With Verian

Reconciliation happens continuously in the data model, not monthly in your spreadsheet — schedule, actuals, commitments, changes, and budget land against one coding structure, and earned value, variance, and forecast update as the data does. Anomalies — the mis-mapped code, the double-booked commitment, the burn rate that broke trend — surface when they occur, each carrying the lineage to prove where the number came from. The month-end pack becomes an output, not a project, and the audit trail is not a folder you assemble under pressure — it is a property of the data itself. You stop manufacturing the numbers and start interrogating them. Which is the job.

Controls should be a discipline, not a data-entry job.

The ledger

of a controls team's month typically consumed by manual reconciliation — returned to analysis40–60%
earned value and variance detection, not a month-end batchContinuous
of figures carrying full lineage to source: audit-ready by construction, not by scramble100%

Four signatures. One record.

Each of these four puts their name to a number someone else will test. The only signature that holds is the one backed by the same record everyone else signed against.

EPC Contractors

sign their name to the claim

Project Owners & Developers

sign their name to the capital decision

Program Directors

sign their name to the board certification

Cost & Controls Leads

sign their name to the month-end close

Chapter 06 of 06

Custody

The record is only worth what protects it.

Everything on this page rests on one assumption: that a project's canonical record can be trusted with the most sensitive commercial data a project generates — bid margins, claims positions, board-level forecasts — and trusted more than any copy of it. That trust is an engineering and governance posture, not a feature. Verian is built for organizations whose projects are commercially sensitive, contractually contested, and occasionally classified — and it treats the record accordingly, from tenancy to exit. This is the room the record lives in.

Isolated by tenant, encrypted by default

Every customer's record lives in its own isolated tenant — no commingled data, no cross-tenant visibility under any configuration. Data is encrypted in transit with TLS 1.2+ and at rest with AES-256, and encryption is not a tier or a toggle: it is the only mode the platform runs in. No customer's data is ever used to train models for anyone else.

Deployed on your terms

Run Verian as managed cloud, inside a dedicated VPC in your own environment, or fully air-gapped for classified and critical-infrastructure programs. It is the same platform in every posture — choosing stricter isolation never means losing capability — and data-residency requirements are honored by design, not by exception.

Access through your identity

Single sign-on via SAML 2.0 puts your identity provider in charge of who gets in, and SCIM provisioning keeps access aligned with your organizational reality — leavers lose access the moment they leave, without a ticket. Role-based permissions mirror project governance down to the line: joint-venture partners, owner's engineers, and subcontractors each see exactly their slice, and nothing else.

An audit trail that cannot be edited

Every write to the record is append-only: who changed what, when, and from what prior value, preserved immutably and exportable for independent verification. There is no administrative back door to tidy history. The same forensic standard the platform applies to your project applies to itself — which is precisely what makes the record defensible in a dispute.

AI you can interrogate

No number that reaches a commercial position is a black box. Every AI-generated figure — an EAC, a delay attribution, a clause risk score — carries its provenance: the source records it was drawn from and the basis for its conclusion, presented for human confirmation, never silent action. If you cannot ask a number where it came from, it does not belong in front of a board.

Your data, and your exit

The record is yours — contractually and practically. Full-fidelity export in open, documented formats is available at any time, including at offboarding, covering the data and its complete history, with certified deletion on request. We consider the right to leave a security feature, and we build for customers who test it before they sign. Independently audited against SOC 2 Type II, with controls aligned to ISO 27001.

SOC 2 Type II · ISO 27001 · AES-256 · TLS 1.2+ · SAML 2.0 · SCIM

End of field study

See the record from your seat.

A 30-minute working session on your real schedule, cost, and change data — from whichever seat you hold. Not a canned demo.

Request a demo