What is a PCSR?
A Project Cost Summary Report — PCSR — is the single most consequential document a UK construction consultancy produces every month. It is the answer to the only question the client really cares about: where are we on cost, and where will we end up?
Despite that, the PCSR has no formal industry standard. Every consultancy formats it differently, every QS interprets the categories slightly differently, and every Director reads it with a different eye. The underlying structure is universal — committed cost, anticipated cost, forecast cost, out-turn cost — but the presentation, depth, and update cadence vary hugely.
This guide describes the PCSR as it exists across most UK construction consultancies. It explains why so many firms still maintain it in Excel, what breaks when they scale, and what changes when the PCSR moves from a workbook to a live data surface. The terminology is UK-native — JCT, NEC, RIBA, BoQ — and the assumptions are consultancy-side, not contractor-side.
Why Excel-based PCSRs break at scale
Most consultancies start with a single PCSR workbook per project. That works fine for a 5-person practice with 6 live projects. The problem starts somewhere around the 15-project mark.
The first failure mode is maintenance time. Each PCSR takes one to two days to update properly each month — committed cost manually keyed from issued POs, anticipated from approved variations, forecast from a separate cashflow model. Across 30 projects, that\'s 60 days of senior QS time per month — between two and three full-time equivalents producing reports. The opportunity cost is variations not raised, suppliers not pursued, and commercial decisions not made.
The second failure mode is error compounding. A single linked formula breaks on a single cell on a single workbook. The QS who built it spots it immediately; their colleague picking up the project two months later does not. By the time a variance reconciliation flags the issue, the wrong number has been in front of the client for three months.
The third failure mode — the one most directors fixate on — is the portfolio view. Each PCSR is a workbook. There is no way to roll up committed cost across 30 live projects in less than a day. Director-level questions take days to answer. So the directors stop asking; the conversation defaults to the projects on fire, and the slow-burning ones drift unnoticed.
None of this is news to anyone who has run a QS team at scale. It is just rarely articulated in one place.
The four cost categories every PCSR must distinguish
Every PCSR — regardless of presentation — works with four cost categories. Getting these straight is the difference between a PCSR that informs decisions and one that obscures them.
Committed cost. The total contractually committed to date through purchase orders, sub-contracts, and approved appointments. Once a PO is issued, the value is committed. This number only grows.
Anticipated cost. Committed cost plus the cost of approved and pending variations. A variation that has been instructed but not yet priced still adds to anticipated. This is the number directors should be reading; it is the live picture of where the project is heading.
Forecast cost. Anticipated cost plus the QS\'s forecast of remaining variations, contingency drawdown, and provisional sum movement. This is professional judgment expressed numerically. A good QS\'s forecast is consistently within 2% of out-turn; a poor one\'s is 10%+ off.
Out-turn cost. The final cost at project completion. This is locked at closeout. The historical PCSR — committed, anticipated, forecast at every monthly snapshot, vs the eventual out-turn — is the single best dataset a consultancy has for improving its forecasting. Most consultancies don\'t archive this data systematically and lose the learning.
How committed / anticipated / forecast / out-turn flow
In a well-run PCSR, the four categories are not independent columns; they are a flow. A change in one cascades through the others.
When a PO is issued, committed cost moves up. Anticipated and forecast both move up by the same amount (assuming no change to the variation pipeline). When a variation is instructed, anticipated moves up. Once the variation is priced and approved, that incremental cost moves from anticipated into committed. Forecast either moves up (if the variation was unforeseen and the QS has revised expectations on remaining work) or stays flat (if the variation was already inside the forecast envelope).
The trouble with maintaining all this in Excel is that the cascade has to be manually applied. Update committed, then update anticipated, then update forecast. Most failure modes happen on the third update — the QS forgets to revise forecast after a major variation, and the forecast reads stale for a month before someone catches it.
In a live PCSR — where committed flows from the PO module, anticipated from the variation module, forecast from the cashflow module, and out-turn locks at closeout — the cascade happens automatically because the four categories share the same underlying data.
How a real-time PCSR changes the consultancy\'s monthly cycle
The traditional consultancy cost cycle runs monthly: a week to build the PCSR, a week to internally review, a week to package for the client, a week of client conversation, then start the next cycle. By the time the client is reading the PCSR, the data is three weeks old.
A live PCSR collapses this. The cost picture is current as of the last PO and the last approved variation — typically minutes ago. The internal review becomes a check on commentary and exception flags, not the numbers themselves. The client conversation is on what the data means, not whether it\'s right.
The Director-level effect is even more pronounced. Portfolio-level questions — total committed across all live projects, projects with anticipated above 105% of target, variations rate by sector — go from “I’ll get back to you next week” to “let me show you on the screen”. Monday-morning briefs become 30-second scans.
None of this requires the QS to do less commercial work. It just removes the maintenance work that prevents them from doing more of it.
When to outgrow the spreadsheet
Excel is a brilliant tool for the first 6–10 projects. It is a dangerous one past 25. The transition point varies, but three signals consistently appear when a consultancy has outgrown spreadsheet-based PCSRs.
Signal one: a senior QS has flagged that they spend more time maintaining workbooks than doing commercial work. This is usually expressed as frustration before it becomes a metric. Listen to it.
Signal two: the Director can\'t answer portfolio-level commercial questions in real time. If the answer to "what\'s our total committed cost across all live projects?" is "give me until Tuesday", the consultancy has outgrown its tooling.
Signal three: a near-miss has happened. A formula broke, a workbook got over-written, a junior QS\'s mistake reached the client. Most consultancies move off Excel within six months of their first material near-miss. The ones that don\'t typically have a worse one shortly after.
Moving off spreadsheet PCSRs is not a technology upgrade; it\'s an operational maturity step. The consultancy that does it sooner gets a quieter compounding advantage — the QS team produces more commercial value per head, the directors make better decisions faster, and the client conversation gets sharper.
Whether the platform is Projavio or another tool, the underlying point is the same: a PCSR is too consequential a number to live in a workbook one formula away from breaking.